ABSTRACT Title of Document: ZOOPLANKTON ECOLOGY IN THE CHESAPEAKE BAY ESTUARINE TURBIDITY MAXIMUM, WITH EMPHASIS ON THE CALANOID COPEPOD EURYTEMORA AFFINIS Scott S. Lloyd, Ph. D., 2006 Directed By: Professor Michael R. Roman, Horn Point Laboratory, University of Maryland Center for Environmental Science The estuarine turbidity maximum (ETM) region of Chesapeake Bay, located near the limit of saltwater intrusion, is characterized by high total suspended solid (TSS) concentrations, high light attenuation, and high densities of zooplankton. Due to high light attenuation, primary production is generally low in ETMs, yet the Chesapeake Bay ETM region is often considered a ?hot spot? of zooplankton abundance within the Bay. The omnivorous copepod Eurytemora affinis is especially prevalent in the ETMs of Chesapeake Bay and its tributaries and in ETM regions worldwide. In order to determine the factors influencing 1) zooplankton distribution and abundance in the Chesapeake Bay ETM, 2) E. affinis reproduction in the Chesapeake Bay and Choptank River ETMs, and 3) zooplankton position maintenance, cruises in the Chesapeake Bay and Choptank River ETMs were conducted in 1996 and 2001-2003. Laboratory experiments examining the egg production cycle of E. affinis were also performed. The cruise results show that zooplankton taxa within the Chesapeake Bay ETM region tend to be distributed along a salinity gradient from up-estuary to down-estuary, with cladocerans being most common in low salinity/freshwater regions, E. affinis found in slightly higher salinities than cladocerans, and mysids and the copepod Acartia tonsa found in more mesohaline conditions. Eurytemora affinis appears to be contained in the ETM by freshwater limiting its up-estuary extent and biological interactions with A. tonsa and salinity tolerances limiting its down-estuary abundance. Grazing and egg production results indicate that E. affinis production is not food-limited in the ETM region and that this copepod?s particle selection ability favors its success in the ETM over that of A. tonsa. Laboratory egg production experiments also suggest that the most accurate estimates of E. affinis egg production (and of all brooding copepods in general) are achieved by incorporating both a temperature-based estimate of interclutch duration (the time between successive clutches) with a temperature-based estimate of egg hatching time. ZOOPLANKTON ECOLOGY IN THE CHESAPEAKE BAY ESTUARINE TURBIDITY MAXIMUM, WITH EMPHASIS ON THE CALANOID COPEPOD EURYTEMORA AFFINIS By Scott Steven Lloyd Dissertation submitted to the Faculty of the Graduate School of the University of Maryland, College Park, in partial fulfillment of the requirements for the degree of Doctor of Philosophy 2006 Advisory Committee: Professor Michael R. Roman, Chair Professor H. Rodger Harvey Professor Edward D. Houde Professor Lawrence P. Sanford Professor Diane K. Stoecker Associate Professor Patrick Kangas, Dean?s Representative ? Copyright by Scott Steven Lloyd 2006 ii Dedication To my beautiful, motivating wife and three furry children. iii Acknowledgements I would first like to thank my advisor, Mike Roman, for his attention to detail, exceptional analytical abilities, and patience. These qualities, mixed with an easy- going nature, left me feeling confident and renewed after every meeting. Mike is a very gifted and enthusiastic scientist, and I am very fortunate to have worked with him. I would also like to thank my other committee members, Rodger Harvey, Ed Houde, Pat Kangas, Larry Sanford, and Diane Stoecker, who have helped me create a document that is superior to anything I thought myself capable of producing. I owe everyone at Horn Point Laboratory a debt of gratitude ? from conversations to late-night filtering, everyone at HPL has helped to shape my graduate experience. For fear of forgetting to include someone, it would be wise for me not to list the names of those who have been especially helpful, but I never claimed to be wise. First, I would like to thank Adam Spear, Dave Kimmel, Hallie Adolph, Xinsheng, Zhang, Angela Padeletti, Carey McGilliard, Matt Reaugh, Deb Kennedy, Danielle Jester, Rebekah, Baker, and everyone else in the ?Zoop Group? for their help - this would not have been possible without you. I would also like to particularly thank Steve Suttles, Elizabeth North, Laurie Van Heukelem, Steph and Jeff Alexander, Anne Willey, Mike Mallonee, Darlene Windsor, Patrick Maloney. Each of you significantly contributed to my success, either at a particular stage in my research or throughout my tenure at Horn Point. To the many others who deserve a specific ?thank you?, please know that I am truly grateful for your contributions. iv I would also like to acknowledge and thank the National Science Foundation and Horn Point for funding this research through a variety of fellowships and grants. Finally, I would like to especially thank my family. Beth, thank you for often lighting a fire beneath me and also letting me know when it?s time for a break. You have been amazing. To my brother, grandparents, aunts, uncles, cousins, nieces, nephews, and in-laws ? you have always been so supportive during this time in my life, even when you had no idea what I was doing. This research has truly been a collaborative effort and should include a chapter consisting only of names. v Table of Contents Dedication..................................................................................................................... ii Acknowledgements......................................................................................................iii Table of Contents.......................................................................................................... v List of Tables .............................................................................................................. vii List of Figures.............................................................................................................. ix CHAPTER 1: General Introduction.............................................................................. 1 INTRODUCTION .................................................................................................... 1 RESEARCH QUESTIONS ...................................................................................... 2 Chapter I: Zooplankton distribution, abundance, and diversity........................... 2 Chapter 2: Egg production and interclutch duration........................................... 3 Chapter 3: Food quantity, quality, and E. affinis egg production ....................... 6 Chapter 4: Tidally-timed vertical migration......................................................... 8 SUMMARY.............................................................................................................. 9 LITERATURE CITED ........................................................................................... 10 CHAPTER II: Zooplankton distribution, abundance, and diversity in the................. 17 Chesapeake Bay estuarine turbidity maximum........................................................... 17 ABSTRACT............................................................................................................ 17 INTRODUCTION .................................................................................................. 18 MATERIALS AND METHODS............................................................................ 20 RESULTS ............................................................................................................... 23 DISCUSSSION....................................................................................................... 31 TABLES ................................................................................................................. 40 FIGURES................................................................................................................ 43 LITERATURE CITED ........................................................................................... 63 CHAPTER III: Incorporating interclutch duration into estimates of brooding copepod egg production ............................................................................................................ 69 ABSTRACT............................................................................................................ 69 INTRODUCTION .................................................................................................. 70 MATERIALS AND METHODS............................................................................ 73 RESULTS ............................................................................................................... 76 DISCUSSION......................................................................................................... 78 TABLE.................................................................................................................... 84 FIGURES................................................................................................................ 85 LITERATURE CITED ........................................................................................... 92 vi CHAPTER IV: Egg production and grazing by the copepod Eurytemora affinis in the Chesapeake Bay estuarine turbidity maximum........................................................... 95 ABSTRACT............................................................................................................ 95 INTRODUCTION .................................................................................................. 96 MATERIALS AND METHODS............................................................................ 97 RESULTS ............................................................................................................. 105 DISCUSSION....................................................................................................... 109 TABLES ............................................................................................................... 117 FIGURES.............................................................................................................. 120 LITERATURE CITED ......................................................................................... 133 CHAPTER V: Tidally-timed vertical migration of zooplankton in the Chesapeake Bay estuarine turbidity maximum............................................................................. 139 ABSTRACT.......................................................................................................... 139 INTRODUCTION ................................................................................................ 140 METHODS ........................................................................................................... 143 RESULTS ............................................................................................................. 146 DISCUSSION....................................................................................................... 148 TABLES ............................................................................................................... 160 FIGURES.............................................................................................................. 163 LITERATURE CITED ......................................................................................... 174 CHAPTER VI: General Conclusions....................................................................... 178 SUMMARY AND CONCLUSIONS ................................................................... 178 LITERATURE CITED ......................................................................................... 184 COMPLETE LITERATURE CITED ....................................................................... 185 vii List of Tables 2.1. Median, (bold values), minimum (min), and maximum (max) values for all abiotic variables and the most abundant zooplankton groups. dO = dissolved oxygen (mg l-1), Temp = Temperature (?C), Salinity = PSU, Chl-a = chlorophyll-a (g l-1). All zooplankton abundances are in number m-3 except for Jelly Volume (the sum of ctenophore and sea nettle volume, ml m-3). S = surface, B = bottom, M = medium. 2.2. Shannon-Weiner diversity index means (H ? 1 SD). October 2001 and 2002 H values are significantly lower than May and July of the respective years and lower than October, 1996 (Tukey-Kramer means comparison,  = 0.05). 3.1. Results of linear regressions performed between model outputs of egg production (EP) versus EP estimates by the egg-ratio (E/A) and clutch size (CS) methods. m = regression slope; b = Y-intercept. 4.1. Mean monthly values for nutritional and abiotic variables (minus microplankton) measured in the Chesapeake Bay (2002) and Choptank River (2003) ETM regions. C = carbon; N = nitrogen; Phaeo = phaeophytin; TSS = total suspended solids; CH2O = carbohydrates; Prot = proteins; NQ = nutritional quotient (sum of CH2O, Prot, and Lipids divided by TSS). 4.2. Mean monthly microplankton biomass (pg C ml-1) as measured in the Choptank River in 2003. Tin = tintinnids; Con = conical ciliates; Sph = spherical ciliates; Hetero = Heterocapsa spp.; P. min = Prorocentrum minimum; Karlo = Karlodinium spp.; Oxy = Oxyrrhis spp.; Gyro = Gyrodinium spp.; Gymno = Gymnodinium spp.; Pperi = Protoperidinium spp.; UnID Dino = unidentified dinoflagellates; Diat = diatoms; Myr/Meso = Myrionecta spp. or Mesodinium spp.; AFlag = autotrophic flagellates (includes cryptophytes). 4.3. Mean carbon-specific ingestion rates, I, [g C ingested (g copepod body C)-1] for E. affinis in the Estuarine Turbidity Maximum (ETM) and downstream from the ETM for each prey type, listed by size class: 10-20 mm, >20 mm; N=3. Autoflag = Autotrophic flagellates. Standard deviations are in parentheses. Total C ingested shows the sum of positive ingestion rates only. 5.1. Median, minimum, and maximum values for Total Suspended Solids (TSS, mg l-1) and Zooplankton Biovolume (BV, mm3 m-3) grouped for both surveys. Surface = 3-6m, Middle = 6-9m, Bottom = 9-12m. 5.2. Correlation coefficients by depth of pairwise correlation analyses between TSS and BV for both cruises. Surface = 0-4 m; Middle = 4-8 m, Bottom = 8-12 m. Values with an asterisk are statistically significant (p < 0.001). viii 5.3. Calculated water displacements (km) for both surveys at three depths (3.5m, 7.5m, and 11.5m). Total displacement is the net average displacement over two tidal cycles except for 7.5 and 11.5m depths for the May 14-15 survey, which are calculated over one tidal cycle. Negative values indicate displacement down-estuary (ebb direction), positive values indicate displacement up-estuary (flood direction). ix List of Figures 2.1. Map of upper Chesapeake Bay showing the ETM region (shaded). 2.2. Total monthly Susquehanna River streamflow as measured at the Conowingo dam for the three study years. The line is the monthly mean from 1968 to 2003. 2.3. May 1996 axial transect through ETM region. The lines in each plot represent isohalines. Crosses represent the mean depths of Tucker trawl samples. River km is the distance down-estuary (south) from Havre de Grace. Zooplankton abundances are in number of individuals m-3. TSS = total suspended solids (mg l-1). All samples were taken at night. 2.4. May 2001 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.5. May 2002 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.6. July 1996 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.7. July 2001 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.8. July 2002 axial transect through ETM region. See Figure 3 caption for an explanation of the plots. The black bar is over stations sampled at night. 2.9. October 1996 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.10. October 2001 axial transect through ETM region. See Figure 3 caption for an explanation of the plots. The black bar is over stations sampled at night. 2.11. October 2002 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 2.12. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all May (1996, 2001, and 2002) data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. x 2.13. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all July data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. 2.14. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all October data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. 2.15. Relationship between Shannon-Weiner diversity index (H) and Salinity (S; psu). Diversity (H) was regressed on log salinity. 2.16. a) Regressions between both maximum TSS and River km (ETM location; distance down-estuary from Havre de Grace) versus total monthly Susquehanna River streamflow in the month prior to the measurements. The TSS versus streamflow regression is the solid linear regression line and has an r2 of 0.49 (TSS = 6.0?10-9(Streamflow) + 24.1); the River km versus streamflow regression was calculated using log10(streamflow) and has an r2 of 0.42 (River km = 6.9 ? Log10(Streamflow) ? 110.0). Both regressions are significant at the 0.05 alpha level. b) Regressions between the locations of A. tonsa and E. affinis peaks versus total monthly Susquehanna River streamflow in the same month as the cruises. The A. tonsa peak location versus streamflow regression is the stippled line and has an r2 of 0.50 (A. tonsa = 5.0?10-9(Streamflow) + 35.6), and the E. tonsa peak location versus streamflow regression has an r2 of 0.61 (E. affinis = 7.0?10-9(Streamflow) + 14.0). Both regressions are significant at the 0.05 alpha level. 2.17. Welch?s ANOVA between E. affinis prosome length and water depth [surface (0-4 m), middle (4-8 m), and bottom (8+ m) waters]. Prosome length is significantly greater in deeper waters (p < 0.05). Diamonds represent the 95% confidence interval. 2.18. Regression between log-transformed chlorophyll-a concentration versus log- transformed E. affinis abundance for all 2001 and 2002 May data (p < 0.001). 2.19. Chlorophyll-a (chl-a) and Eurytemora affinis (E. affinis) concentrations versus salinity. The dashed line represents the calculated chl-a concentration decrease by dilution alone; the solid line is the logarithmic fit to actual chl-a values versus salinity (p < 0.01) 2.20. Regression between log-transformed chlorophyll-a concentration versus log transformed A. tonsa abundance for all 2001 and 2002 data (p < 0.05). xi 3.1. Conceptual diagram of reproductive cycle of E. affinis showing the two major time components, hatching time and interclutch duration. Hatching time is primarily influenced by temperature, and a similar relationship with temperature is presumed for interclutch duration. 3.2. Relationship between interclutch duration (days) and temperature (?C). Ban (1994) data ? 1 SD. Fitted power function r2=0.98 (does not include Ban (1994) data). 3.3. Comparison of Interclutch Durations for copepods fed 200-m filtered Choptank River water (NW) and those fed Choptank River water plus a Thalassiosira weissflogii supplement (NW+Tw) at the four experimental temperatures. 3.4. Graphical representation of the time components of E. affinis? reproductive cycle. HT = hatching time; ID = interclutch duration; TC = total cycle length (sum of HT and ID). 3.5. Hirst-Sheader (H-S) model weight-specific EP predictions versus actual weight- specific EP. (a) Actual EP values calculated by the egg-ratio (E/A) method. The dashed line is the linear fit to of H-S model output to 2002 data. 2003 data versus predicted values was not significant. (b) Actual EP values calculated by the mean clutch size (CS) method. The dashed line is the linear fit to of H-S model output to 2002 data. Linear fits to both 2002 and 2003 data were significant. ?C & D (1986)? = Crawford and Daborn (1986). See Table 1 for all linear regression statistics. 3.6. Comparison of EP rates estimated by the egg-ration method (E/A) and the mean clutch size method (CS). Dashed line is the 1:1 relationship. 3.7. Dorso-lateral view of gravid E. affinis showing hemispheric nature of the egg mass. The two darker regions of the egg mass (em) are separated by the cleft (light- colored region) below the attached spermatophores (sp). The urosome (ur) is lifted dorsally in this photograph but would normally rest in the cleft region. 4.1. A map of the USA east coast, with Chesapeake Bay and its tributaries enlarged. The estuarine turbidity maximum (ETM) regions of the Chesapeake Bay mainstem and the Choptank River are shown using filled stars. 4.2. May, 2001 contour maps of turbidity (in nephalometric turbidity units, NTU) and E. affinis abundance (# m-3) in northern Chesapeake Bay. Note the relationship between turbidity (NTU) and copepod abundance. River km is measured as the distance downstream from Havre de Grace. 4.3. Linear regressions of clutch size versus female prosome length for 2002 (Chesapeake Bay) and 2003 (Choptank River). Only completely formed clutches were used in this analysis. xii 4.4. Regression tree analysis using the residuals from the 2002 (Chesapeake Bay) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See Table I caption for abbreviation information. 4.5. Regression tree analysis using the residuals from the 2003 (Choptank River) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See the captions for Tables I and II for abbreviation information. 4.6. Regression tree analysis using the residuals from the 2002 (Chesapeake Bay) and 2003 (Choptank River) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See the captions for Tables I and II for abbreviation information. 4.7. Linear regressions of female prosome length and estimated mean temperature during development for 2002 (Chesapeake Bay) and 2003 (Choptank River). 4.8. Carbon-specific egg production rates plotted versus temperature. Linear regression is for 2002 (Chesapeake Bay) data and is significant at p < 0.01. 4.9. Comparisons of C-specific egg production rates from this and other studies to the Hirche (1992) model of E. affinis egg production under food replete conditions. 4.10. Calculations for particulate carbon concentrations (POC) concentrations required for copepod maintenance (Cm; copepod carbon assimilation = respiratory demands, stippled line) and non-food limited copepod growth (Cc, solid line). Actual POC concentrations for 2002 and 2003 are also plotted. 4.11. Eurytemora affinis egg production rates versus modeled rates from Huntley and Boyd (1984). The modeled rates are based on temperature and copepod weight and assume no food limitation. The r2 for the linear regression is 0.43 (p < 0.05). 4.12. Results of gut fluorescence measurements following grazing experiment. Chlorophyll-a has been converted to carbon units in order to C-specific ingestion on autotrophic prey. 4.13. Means of E. affinis abundance at Chesapeake Bay Program sampling stations from 1986-1995. 1990 and 1993 are considered years of enhanced Susquehanna River freshwater input (?wet? years). 1986, 1987, 1988, 1989, and 1992 are considered ?normal? and ?dry? years. The years following wet years (1991, 1994, and 1995) were not included in this analysis because E. affinis abundance those years xiii was highly influenced by resting eggs laid the previous year (Kimmel and Roman 2004). 5.1. Map of the United States east coast with the Chesapeake Bay ETM region magnified. Solid circles and the star represent the approximate locations of all stations of the 25-hour surveys. The star represents the location of the station at river km 30. The dotted line perpendicular to the bay axis represents the dividing line between up-estuary and down-estuary stations and is at approximately river km 28. 5.2. Comparison of TAPS Biovolume (BV) measurements with zooplankton counts from pump samples. 5.3. Salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1), and current velocity (bottom plot; positive values indicate flood currents, negative values indicate ebb currents) for the first survey (May 8-9). The solid line in the current plot represents currents at 11m depth, the stippled line represents currents at 3 m depth. The black bar above the top frame marks stations that were sampled at night. 5.4. Salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m- 3), total suspended solids (middle contour plot; mg l-1), and current velocity (bottom plot; positive values indicate flood currents, negative values indicate ebb currents) for the second survey (May 14-15). The solid line in the current plot represents currents at 11m depth, the stippled line represents currents at 3 m depth. The black bar above the top frame marks stations that were sampled at night. 5.5. Multiple fixed-station contour plots of salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1) versus river km (distance from Havre de Grace, Maryland) for the first survey (May 8-9). Transects were surveyed chronologically by letter (a-l). Tidal stage is given for each transect. The bold line represents the 1 PSU isohaline, each isohaline to the right increases by 2 PSU. 5.6. Multiple fixed-station contour plots of salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1) versus river km (distance from Havre de Grace, Maryland) for the second survey (May 14-15). Transects were surveyed chronologically by letter (a-k). Tidal stage is given for each transect. The bold line represents the 1 PSU isohaline, each isohaline to the right increases by 2 PSU. 5.7. Regression analyses between both zooplankton biovolume (BV) and total suspended solids (TSS) versus current velocity (Vel) for transect stations up-estuary from river km 28. Bottom = 8-12 m depth; Middle = 4-8 m depth; Surface = 0-4 m depth. xiv 5.8. Regression analyses between both zooplankton biovolume (BV) and total suspended solids (TSS) versus current velocity (Vel) for transect stations down- estuary from river km 28. Bottom = 8-12 m depth; Middle = 4-8 m depth; Surface = 0-4 m depth. 5.9. Contour plots of Tucker trawl data from Chesapeake Bay ETM region, May, 2002. Line contours ? salinity (PSU); Color contours ? Eurytemora affinis and cladoceran (Bosmina longirostris and Daphnia spp.) abundance (individuals m-3); X- axis is distance down-estuary from Havre de Grace, Maryland. 5.10. A hypothetical snapshot of zooplankton biovolume (BV; each represents approx. 500 mm-3 of BV) and total suspended solid (TSS; relative intensity indicated by shading) concentrations during maximum flood and ebb conditions in the Chesapeake Bay ETM region in May, 2002. The bold arrows in both frames represent relative current velocities in surface (1-4 m depth), middle (4-8 m depth) and bottom (8-12 m depth) portions of the water column, and the stippled line shows the mean isohaline structure. On the seaward side of the mid-point between all transects (28 km from Havre de Grace), BV concentrations in the bottom and middle portions of the water column are significantly greater during flood than ebb. On the landward side, BV concentrations are significantly greater in surface and middle waters during ebb than flood. During flood, bottom scouring resuspends both TSS and epibenthic zooplankton in both landward and seaward regions, especially in bottom waters. During the slack before ebb, TSS concentrations decrease as sediments settle out of the water column. Zooplankton swimming behaviors and zooplankton resuspension during the following ebb may result in significantly greater BV in surface and middle waters than during flood. Additionally, zooplankton swept down-stream from the Susquehanna River might add to zooplankton BV in surface waters, especially during ebb tides. The locations of BV maxima during the tidal cycle suggest that zooplankton tend to be at the depths of maximum current velocities, resulting in a ?conveyor belt? of zooplankton movement. TSS concentrations are generally greater in both landward and seaward regions during flood than ebb at all depths, probably due to greater bottom stress during flood. 5.11. Linear regression of zooplankton biovolume (BV) versus salinity using all May data. xv 1 CHAPTER 1: General Introduction INTRODUCTION The estuarine turbidity maximum (ETM) region of Chesapeake Bay, located near the limit of saltwater intrusion, is traditionally characterized by high total suspended solid (TSS) concentrations, high light attenuation, and high densities of zooplankton and anadromous fish larvae (North and Houde 1999, Roman et al. 2001). Phytoplankton productivity is generally low in the ETM region due to light limitation (Wofsy 1983, Kemp and Boynton 1984, Fisher et al. 1988). Therefore, food webs in ETMs tend to be based on allochthonous detrital inputs, with important microbial components (Irigoien and Castel 1997, Crump et al. 1998, North and Houde 2001). Abundances of the calanoid copepod E. affinis peak in the Chesapeake Bay ETM region. Eurytemora affinis is omnivorous, feeding on phytoplankton (Vijverberg 1980, Gasparini and Castel 1997), detritus (Heinle et al. 1977), protozoa (Berk et al. 1978, Merrell and Stoecker 1998), and bacteria (Boak and Goulder 1983). It is usually considered to be an epibenthic copepod (Heinle and Flemer 1975, Simenstad 1990), and gravid females, which presumably are visible to visual predators, are often rare in the water column (Vuorinen 1987). The high abundance of E. affinis and other zooplankton in the most turbid regions of estuaries leads to the question: what factors influence zooplankton abundance and distribution in ETM regions? One hypothesis is that the high turbidity provides a refuge from visual predators, especially fish. Therefore, high copepod abundances in turbid regions may simply be the result of lower mortality rates 2 (Kimmerer et al. 1998). The ETM may also provide the optimal temperature-salinity- food regime for growth and reproduction. North and Houde (2001) suggest that survival of white perch and striped bass larvae in the Chesapeake Bay ETM is determined by freshwater input into the system, which affects temperature, salinity, and zooplankton abundance in this region. Temperature by itself may explain over 90 % of all growth rate variability for marine copepods (Huntley and Lopez 1992) and is a primary factor in determining developmental and reproductive rates for estuarine copepods (Katona 1970, Nagaraj 1992, Ban 1994). RESEARCH QUESTIONS The following research chapters attempt to describe zooplankton ecology in the Chesapeake Bay and Choptank River estuarine turbidity maximum (ETM) regions. This research focuses on factors influencing the distribution, abundance, and reproductive rates of the calanoid copepod Eurytemora affinis. Specifically, I seek to 1) resolve relationships influencing zooplankton distributions, abundances, and diversity; 2) consider a more direct approach to estimating brooding (egg-carrying) copepod egg production; 3) determine whether or not E. affinis egg production is food-limited in the Chesapeake Bay and Choptank River ETMs and what other factors may influence egg production rates; and 4) determine if E. affinis vertically migrates to maintain its position within the Chesapeake Bay ETM region. Chapter I: Zooplankton distribution, abundance, and diversity 3 The goal of Chapter 1 is to determine the primary factors influencing zooplankton location and abundance, particular those of E. affinis. The various zooplankton taxa exhibit abundance peaks in different regions throughout the ETM region, and these distributions are likely the result of biological and physical interactions which change temporally and spatially. Resolving these interactions may allow for predictions of zooplankton abundance and distribution to be made as the physical environment changes seasonally, between years, and over longer periods of time. Chapter 2: Egg production and interclutch duration Brooding copepods (those that carry their eggs until hatching) differ from broadcast spawning copepods (those that release their unhatched eggs) in several ways, one way being that a brooding female cannot produce a new clutch of eggs until after her present clutch hatches or the unhatched (usually dead) eggs are released (Runge and Roff 2000). This adds an important time component to egg production rate ? hatching time (HT). Several studies have focused on temperature effects on HT (also known as egg development time or embryonic development time), which is often defined differently by different authors (Heinle and Flemer 1975, Vijverberg 1980, Ban and Minoda 1992, Andersen and Nielsen 1997). Only one study has measured the effects of salinity on embryonic development time, though, as stated above, the authors did not incorporate a temperature-salinity interaction effect (Ishikawa et al. 1999). Another time period in brooding copepod egg production must also be considered ? interclutch duration (ID). Interclutch duration is the period from the 4 hatching of the last egg of a clutch to the extrusion of the first egg of another clutch. Ban (1994) found a significant effect of temperature on E. affinis ID but a nonsignificant food concentration effect. There do not appear to be any studies on the effects of salinity on ID. Traditional egg production measurements involve incubating copepods in bottles filled with filtered water, natural water, or water with a food amendment (Runge and Roff 2000) over a period of 24 hours to capture any diel periodicity in egg production (White and Roman 1992). Although spawning copepods appear to lay distinct clutches (Mauchline 1998), they lay them over shorter periods than do brooding copepods, often during a 24-hour period. Measured HTs for E. affinis range from over 14 days at temperatures below 5?C to ~1 day at 25?C (Andersen and Nielsen 1997). Such long HTs mean that traditional incubation techniques may not be applicable for brooders. One solution to this problem is to run incubations in large bottles (4 l) containing male and female copepods and periodically remove subsamples to enumerate nauplii and eggs female-1 (Burdloff et al. 1999). Counts for these subsamples are then compared to subsamples taken at the beginning of the incubations. The major assumption of this method is that each subsample contains the same ratio of gravid to nongravid females and is a representative sample of nauplius concentration (nauplii are uniformly distributed). Thus, there is the potential for much error with this method, and egg production rates seem low for this method compared to other methods [compare Burdloff et al. (1999) to Katona (1975) and Ban (1994)]. 5 The ?egg ratio method?, developed by Edmondson et al. (1962), for estimating the egg production rates of rotifers has also been used to estimate copepod egg production (e.g.Edmondson et al. 1968, Sabatini and Ki?rboe 1994). The egg ratio is egg abundance (E, per unit volume) to total number of females (A, per unit volume), giving units of eggs female-1. The egg ratio alone does not estimate production rate, but if the HT can be determined based on environmental conditions (temperature, salinity, etc.) at the time of copepod capture, production rate can theoretically be determined (Runge and Roff 2000). According to Edmondson et al. (1962), egg production rate (EP) should equal the egg ratio divided by HT: ( ) EEP A HT  =   . Therefore, the only time component one needs to estimate EP using this method is HT. Theoretically, ID and HT could be used directly in EP measurements by dividing average sample clutch size by the sum of both time components: CSEP HT ID  =   + . The major goals of Chapter 2 are to determine the influence of temperature on measurements of E. affinis ID, use these ID values to estimate EP from field-collected E. affinis samples, compare these values to the EP values obtained through the egg- ratio method, and compare both types of EP estimates to models of copepod growth. An evaluation can then be made concerning which approach to measuring copepod egg production is most useful and/or valid. 6 Chapter 3: Food quantity, quality, and E. affinis egg production Measurements of copepod egg production are often used to estimate reproductive rates as well as larval and juvenile biomass production, as both often depend on the same environmental factors (Berggreen et al. 1988, Runge and Roff 2000). In theory, adult females allocate carbon that would be used for somatic growth (adult copepod nonreproductive biomass is relatively constant) towards the production of eggs. There are conditions, such as during periods of food limitation, where somatic growth of immature stages and reproductive output by adults are not equivalent, and the above assumption does not hold (Hopcroft and Roff 1998). The major factors known to affect egg production rates (and growth rates) for most calanoid copepod species are temperature, food concentration and food quality. Body size is also positively related to absolute reproductive output, so growth rates are usually expressed as weight-specific growth (day-1). Temperature has long been recognized as a primary factor affecting the physiological rates of poikilotherms (Blehr?dek 1935, cited by McLaren et al. 1966). As stated above, Andersen and Nielsen (1997) analyzed E. affinis egg HTs versus temperature (T) from five studies and found that a power function explained 87% of the variability. Huntley and Lopez (1992) pooled data for 33 species of marine copepods corresponding to a temperature range of 32.4?C, and described two functions: one power function describing generation time (days) in terms of temperature (r2 = 0.91) and one describing weight- specific growth rate (day -1) in terms of generation time (r2 = 0.96). The relationship between temperature and a physiological rate, such as egg production, does not need to be modeled as a power function. Temperature may have a linear effect (Ambler 7 1985) or a parabolic effect (White and Roman 1992) on Acartia tonsa egg production. A parabolic effect of temperature on egg production is logical, as most physiological processes have an optimum temperature range above and below which the rates decline (Mauchline 1998). Such differences of temperature effects do not necessarily represent a physiological difference between populations but may reflect variability in the other factors affecting egg production (Ambler et al. 1986, J?nasd?ttir 1994, Hirst and McKinnon 2001). Food quantity and quality are also known to affect egg production rates (Mauchline 1998, Runge and Roff 2000). Differentiating between quantity and quality, though, is somewhat subjective, because one must define quantity in an ecologically relevant manner. Food availability has traditionally been measured in terms of carbon, chlorophyll, or both (Heinle and Flemer 1975, Runge and Roff 2000). Chlorophyll, however, is often not a good predictor of estuarine copepod egg production due to allochthonous carbon inputs to estuaries (Ambler 1986, White and Roman 1992, Burdloff et al. 1999). In turbid regions of estuaries, other food sources, such as microzooplankton and detritus, become important (Berk et al. 1978; Heinle et al. 1977). The ratios of both chlorophyll and the sum of protein, carbohydrate, and lipid concentrations to TSS are correlated to E. affinis egg production in the Gironde estuary (Burdloff et al. 1999). Acartia spp. egg production has been negatively correlated with seston C:N (Ambler 1986) and positively correlated with protein and specific fatty acid concentrations (J?nasd?ttir 1994, J?nasd?ttir et al. 1995) and microzooplankton abundance (White and Roman 1992). 8 The focus of Chapter 3 is to determine how biotic (food quantity and quality) and abiotic factors influence E. affinis clutch sizes (and thus egg production rates), and if any of these factors limit egg production in the Chesapeake Bay and Choptank River ETMs. Grazing experiments were also conducted to determine the diet of adult E. affinis both in the ETM and down-estuary from the ETM (where A. tonsa typically dominates the zooplankton community). Chapter 4: Tidally-timed vertical migration Several researchers have demonstrated that E. affinis uses tidally-timed vertical migration to maintain its position within an estuary (Hough and Naylor 1991, Simenstad et al. 1994, Morgan et al. 1997). In these studies, E. affinis was generally more abundant higher in the water column during a flood tide and deeper during ebb, thus being associated with the lowest velocity regions within the water column. Morgan et al. (1997) found that the principal lunar tidal component (M2) described nearly half of all variability in E. affinis densities, further supporting the idea of a tidally-induced vertical migration. Vertical migrations based on the tidal cycle are common in the larvae of other, meroplanktonic organisms such as crabs (Cronin and Forward 1983, Tankersley and Forward 1994, Zeng et al. 1999), fish (Dauvin and Dodson 1990, Joyeux 1999), and molluscs (Roberts et al. 1989). Other authors, however, have demonstrated that E. affinis populations may be sustained within estuaries by passive means. Castel and Viega (1990) determined that the swimming abilities of E. affinis are ineffective in retaining the population within the turbid Gironde estuary. This finding and the relationship between E. affinis abundance and suspended matter lead them to the conclusion that the copepods 9 behave as passive particles to maintain themselves within the Gironde estuary. Kimmerer et al. (1998) found that macrozooplankton (mysids and amphipods) exhibited tidally-induced vertical migrations that maintained their positions within San Francisco Bay (higher in the water column during flood, lower during ebb), but the copepods E. affinis and Pseudodiaptomus forbesi did not. Eurytemora affinis? position maintenance strategy apparently either differs with location / hydrodynamic properties of the estuary or is masked by certain hydrodynamic characteristics. In Chapter 4, acoustically-determined zooplankton biovolume measurements are used to resolve the vertical distributions of zooplankton with respect to tidal stage and magnitude in the Chesapeake Bay ETM region. The major goal was to determine if E. affinis actively regulates its vertical position within the water column to enhance its retention in the ETM. The results are then compared to a model by Roman et al. (2001), which was developed using data from a year with markedly different hydrologic characteristics. SUMMARY Together, these four chapters should provide a better understanding of zooplankton ecology in the Chesapeake Bay ETM; the factors that influence E. affinis distribution and abundance; how E. affinis is retained in the Chesapeake Bay ETM; and, if production of E. affinis is food limited in the ETM. 10 LITERATURE CITED Ambler, J. W. 1985. Seasonal factors affecting egg production and viability of eggs of Acartia tonsa Dana from East Lagoon, Galveston, Texas. Estuarine, Coastal, and Shelf Science 20:743-760. Ambler, J. W. 1986. Effect of food quantity and quality on egg production of Acartia tonsa Dana from East Lagoon, Galveston, Texas. Estuarine, Coastal and Shelf Science 23:183-196. Andersen, C. M., and T. G. Nielsen. 1997. Hatching rate of the egg-carrying estuarine copepod Eurytemora affinis. Marine Ecology Progress Series 160:283-289. Ban, S. H. 1994. Effect of temperature and food concentration on postembryonic development, egg-production and adult body-size of the calanoid copepod Eurytemora-affinis. Journal of Plankton Research 16:721-735. Ban, S., and T. Minoda. 1992. Hatching of diapause eggs of Eurytemora-affinis (Copepoda, Calanoida) collected from lake-bottom sediments. Journal of Crustacean Biology 12:51-56. Blehr?dek, J. 1935. Temperature and living matter. Protoplasma, Monograph 8:1- 277. Berk, S. G., D. C. Brownlee, D. R. Heinle, H. J. Kling, and R. R. Colwell. 1978. Ciliates as a food source for marine planktonic copepods. Microbial Ecology 4(1):27-40. Berggreen, U., B. Hansen, and T. Kioerboe. 1988. Food size spectra, ingestion and growth of the copepod Acartia tonsa during development: Implications for determination of copepod production. Marine biology 99:341-352. Boak, A. C. and R. Goulder 1983. Bacterioplankton in the diet of the calanoid copepod Eurytemora sp. in the Humber Estuary. Marine Biology. 73:139-149. Burdloff, D., S. Gasparini, B. Sautour, H. Etcheber, and J. Castel. 1999. Is the copepod egg production in a highly turbid estuary (the Gironde, France) a function of the biochemical composition of seston? Aquatic Ecology. 34:165- 175. Castel, J and J Viega. 1990. Distribution and retention of the copepod Eurytemora- affinis-hirudinoides in a turbid estuary. Marine Biology 107(1):119-128. Cronin, T. W., and R. B. Forward, Jr. 1983. Vertical migration rhythms of newly hatched larvae of the estuarine crab, Rhithropanopeus harrisii. Biological Bulletin 165:139-153. 11 Crump, BC, JA Baross, and CA Simenstad. 1998. Dominance of particle-attached bacteria in the Columbia River estuary, USA. Aquatic Microbial Ecology 14(1):7-18 Dauvin, J.-C. and J. Dodson. 1990. Relationship between feeding incidence and vertical and longitudinal distribution of rainbow smelt larvae (Osmerus mordax) in a turbid well-mixed estuary. Marine Ecology Progress Series 60:1- 12. DeMott, W. R. 1998. Utilization of a cyanobacterium and a phosphorus-deficient green alga as complementary resources by daphnids. Ecology 79:2463-2481. Dubois, M, KA Gills, JK Hamilton, PA Rebers, and F Smith. 1956. Colorimetric method for determination of sugars and related substances. Analytical Chemistry 28:350-356. Edmondson, W. T. 1968. A graphical model for evaluating the use of the egg ratio for measuring birth and death rates. Oecologia 1:1-37. Edmondson, W. T., G. W. Comita, and G. C. Anderson. 1962. Reproductive rate of copepods in nature and its relation to phytoplankton population. Ecology 43:625-634. Elser, J. J., and J. Urabe. 1999. The stoichiometry of consumer-driven nutrient recycling: Theory, observations, and consequences. Ecology 80:735-751. Fisher, TR, LW Harding, DW Stanley, et al. 1988. Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson Estuaries. Estuarine, Coastal, and Shelf Science 27:61-93. Froelich, P. N. 1988. Kinetic control of dissolved phosphate in natural rivers and estuaries: A primer on the phosphate buffer mechanism. Limnology and Oceanography 33:649-668. Gasparini, S., and J. Castel. 1997. Autotrophic and heterotrophic nanoplankton in the diet of the estuarine copepods Eurytemora affinis and Acartia bifilosa. Journal of Plankton Research 19:877-890. Gismervik, I. 1997a. Stoichiometry of some marine planktonic crustaceans. Journal of Plankton Research 19:279-285. Gismervik, I. 1997b. Implications of zooplankton stoichiometry on distribution of N and P among planktonic size fractions. Journal of Plankton Research 19:343- 356. 12 Goldman, J. C. 1979. Temperature effects on steady-state growth, phosphorus uptake, and the chemical composition of a marine phytoplankter. Microbial Ecology 5:153-166. Goolish, E. M., and R. S. Burton. 1989. Energetics of osmoregulation in an intertidal copepod: Effects of anoxia and lipid reserves on the pattern of free amino acid accumulation. Functional Ecology 3:81-89. Hasle, G. R. 1978. The inverted microscope method. in A. Sournia, editor. Phytoplankton manual. UNESCO, Paris. Heinle, D. R. and D. A. Flemer. 1975. Carbon requirements of a population of the estuarine copepod Eurytemora affinis. Marine biology 31:235-247. Heinle, D. R., R.P. Harris, J.F. Ustach, and D.A. Flemer. 1977. Detritus as food for estuarine copepods. Marine Biology 40:341-353. Hirst, A. G. and A. D. McKinnon 2001. Does copepod egg production represent adult female copepod growth? A call to account for body weight changes. Marine Ecology Progress Series 223:179-199. Hopcroft, R. R., and J. C. Roff. 1998. Zooplankton growth rates: The influence of female size and resources on egg production of tropical marine copepods. Marine Biology 132:79-86. Hough, A. R., and E. Naylor. 1992. Endogenous rhythms of circatidal swimming activity in the estuarine copepod Eurytemora-affinis (Poppe). Journal of Experimental Marine Biology and Ecology. 161:27-32. Howarth, R. W., H. S. Jensen, R. Marino, and H. Postma. 1995. Transport To and Processing Of P in Near-shore and Oceanic Waters. in H. Tiessen, editor. Phophorus in the Global Environment: Transfers, Cycles, and Management. John Wiley and Sons, West Sussex, England. Huntley, M. and M. D. G. Lopez. 1992. Temperature-dependent production of marine copepods: a global synthesis. The American Naturalist 140:201-242. Irigoien, X., J. Castel, and B. Sautour. 1993. In-situ grazing activity of planktonic copepods in the Gironde Estuary. Cahiers De Biologie Marine 34:225-237. Irigoien, X., and J. Castel. 1997. Light limitation and distribution of chlorophyll pigments in a highly turbid estuary: the Gironde (SW France). Estuarine, Coastal and Shelf 44:507-517. 13 Ishikawa, A., S. Ban, and N. Shiga. 1999. Effects of salinity on survival and embryonic and postembryonic development of Eurytemora affinis from a freshwater lake. Plankton Biology and Ecology 46:113-119. Jacobs, F., W. Burton, and I. Moss. 1985. Patterns of zooplankton abundance, species composition, and biomass in upper Chesapeake Bay. Estuaries 8:78A. J?nasd?ttir, S. H. 1994. Effects of food quality on the reproductive success of Acartia tonsa and Acartia hudsonica: Laboratory observations. Marine biology 121:67-81. J?nasd?ttir, S. H., D. Fields, and S. Pantoja. 1995. Copepod egg production in Long Island Sound, USA, as a function of the chemical compostion of seston. Marine Ecology Progress Series 119:87-98. Joyeux, J. C. 1999. The abundance of fish larvae in estuaries: Within-tide variability at inlet and immigration. Estuaries 22:889-904. Katona, S. K. 1970. Growth characteristics of the copepods Eurytemora affinis and E. herdmani in laboratory cultures. Helgol?nder wiss. Meeresunters. 20:373-384. Katona, S. K. 1975. Copulation in the copepod Eurytemora affinis (Poppe, 1880). Crustaceana 28:89-95. Kemp, W. M., and W. R. Boynton. 1984. Spatial and temporal coupling of nutrient inputs to estuarine primary production: The role of particulate transport and decomposition. Bulletin of Marine Science 35:522-535. Kimmel, D. G., and B. P. Bradley. 2001. Specific protein responses in the calanoid copepod Eurytemora affinis (Poppe, 1880) to salinity and temperature variation. Journal of Experimental Marine Biology and Ecology 266:135-149. Kimmerer, W. J., J. R. Burau, and W. A. Bennett. 1998. Tidally oriented vertical migration and position maintenance of zooplankton in a temperate estuary. Limnology and Oceanography. 43:1697-1709. Koski, M. 1999. Carbon:nitrogen ratios of Baltic Sea copepods-indication of mineral limitation? Journal of Plankton Research 21:1565-1573. Mauchline, J. 1998. The biology of calanoid copepods. Academic Press, Inc., 525 B St., Suite 1900 San Diego, CA 92101-4495 USA. McLaren, I. A. 1966. Predicting development rate of copepod eggs. Biological Bulletin 131:457-469. 14 Merrell, JR and DK Stocker. 1998. Differential grazing on protozoan microplankton by developmental stages of the calanoid copepod Eurytemora affinis Poppe. Journal of Plankton Research 20(2):289-304. Morgan, C., JR Cordell, and CA Simenstad. 1997. Sink or swim? Copepod population maintenance in the Columbia River estuarine turbidity maxima region. Marine Biology 129:309-317. Nagaraj, M. 1992. Combined effects of temperature and salinity on the development of the copepod Eurytemora-affinis. Aquaculture 103:65-71. North, EW and ED Houde. 2001. Retention of white perch and striped bass larvae: Biological-physical interactions in the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):756-789. Parsons, T. R., Y. Maita, and C. M. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford. 187pp. Porter, J., D. Deere, M. Hardman, C. Edwards, and R. Pickup. 1997. Go with the flow - use of flow cytometry in environmental microbiology. FEMS Microbiology Ecology 24:93-101. Poulet, S. A. 1983. Factors controlling utilization of non-algal diets by particle- grazing copepods: A review. Oceanologica acta 6:221-234. Redfield, A. C. 1958. The biological control of chemical factors in the environment. American Scientist 46:206-221. Richman, S., D.R. Heinle, and R. Huff. 1977. Grazing by adult estuarine copepods of the Chesapeake Bay. Marine Biology. 42:69-84. Roberts, D., D. Rittschof, D. J. Gerhart, A. R. Schmidt, and L. G. Hill. 1989. Vertical migration of the clam Mercenaria mercenaria (L.) (Mollusca: Bivalvia). Environmental correlates and ecological significance. Journal of Experimental Marine Biology and Ecology 126:271-280. Roddie, B. D., R. J. G. Leakey, and A. J. Berry. 1984. Salinity-temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth Estuary. Journal of Experimental Marine Biology and Ecology 77:191-211. Roman, MR, DV Holliday, and LP Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. 15 Runge, J. A. and J. C. Roff. 2000. The measurement of growth and reproductive rates. Pages 401-444 in R. P. Harris, P.H. Wiebe, J. Lenz, H.R. Skjokdal, and M. Huntley, editors. ICES Zooplankton Methodology Manual. Academic Press, San Diego. Sabatini, M., and T. Kiorboe. 1994. Egg-production, growth and development of the cyclopoid copepod Oithona-similis. Journal of Plankton Research 16:1329- 1351. Sanford, L. P., S. E. Suttles, and J. P. Halka. 2001. Reconsidering the physics of the Chesapeake Bay estuarine turbidity maximum. Estuaries 24:655-669. Schubel, J. R. 1968. Turbidity maximum of the northern Chesapeake Bay. Science 161:1013-1015. Sharp, J. H., C. H. Culberson, and T. M. Church. 1982. The chemistry of the Delaware Estuary. General considerations. Limnology and Oceanography 27:1015-1028. Simenstad, C. A., L. F. Small, and C. D. McIntire. 1990. Consumption processes and food web structure in the Columbia River estuary. Progress in Oceanography 25:271-297. Simenstad, C. A., C. A. Morgan, J. R. Cordell, and J. A. Baross. 1994. Flux, passive retention, and active residence of zooplankton in Columbia River estuarine turbidity maxima. Olsen & Olsen, Fredensborg, Denmark. Smith, PK, RI Krohn, GT Hermanson et al. 1985. Measurements of protein using bicinchonic acid. Analytical Biochemistry 150:76-85. Solorzano, L., and J. H. Sharp. 1980. Determination of total dissolved phosphorus and particulate phosphorus in natural waters. Limnology and Oceanography 25:754-758. Sterner, R. W. 1990. The ratio of nitrogen to phosphorus resupplied by herbivores: Zooplankton and the algal competitive arena. American Naturalist 136:209- 229. Strom, S. L., R. Benner, S. Ziegler, and M. J. Dagg. 1997. Planktonic grazers are a potentially important source of marine dissolved organic carbon. Limnology and Oceanography 42:1364-1374. Tankersley, R. A., and R. B. Forward, Jr. 1994. Endogenous swimming rhythms in estuarine crab megalopae: Implications for flood-tide transport. Marine biology 118:415-423. 16 Touratier, F., L. Legendre, and A. Vezina. 1999. Model of copepod growth influenced by the food carbon:nitrogen ratio and concentration, under the hypothesis of strict homeostasis. Journal of Plankton Research 21:1111-1132. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10:317-340. Vuorinin, I. 1987. Vertical migration of Eurytemora (Crustacea, Copepoda) ? A compromise between the risks of predation and decreased fecundity. Journal of Plankton Research 9(6):1037-1046. Walve, J., and U. Larsson. 1999. Carbon, nitrogen and phosphorus stoichiometry of crustacean zooplankton in the Baltic Sea: implications for nutrient recycling. Journal of Plankton Research 21:2309-2321. White, J. R., and M. R. Roman. 1992. Egg production by the calanoid copepod Acartia tonsa in the mesohaline Chesapeake Bay: The importance of food resources and temperature. Marine Ecology Progress Series. 86:239-249. Wofsy, S. C. 1983. A simple model to predict extinction coefficients and phytoplankton biomass in eutrophic waters. Limnology and Oceanography 28:1144-1155. Zeng, C., P. Abello, and E. Naylor. 1999. Endogenous tidal and semilunar moulting rhythms in early juvenile shore crabs Carcinus maenas: Implications for adaptation to a high intertidal habitat. Marine Ecology Progress Series 191:257-266. 17 CHAPTER II: Zooplankton distribution, abundance, and diversity in the Chesapeake Bay estuarine turbidity maximum ABSTRACT Estuarine turbidity maximum (ETM) regions of many estuaries have been shown to contain high abundances of various zooplankton taxa. Previous research in the Chesapeake Bay ETM has been on a relatively large scale (10s of kilometers between stations) or has not focused on the different zooplankton taxa that comprise this community. I conducted high-resolution sampling in the Chesapeake Bay ETM in the spring, summer, and fall in 1996, 2001, and 2002. Freshwater input into the ETM region via the Susquehanna River was higher than normal in 1996 (a wet year), average in 2001 (a normal year), and low in 2002 (a dry year). The calanoid copepod Eurytemora affinis, cladocerans (Bosmina longirostris and Daphnia spp.), and amphipods (Gammarus spp.) were especially abundant during periods of high freshwater input. These zooplankton groups favor low salinity waters and were found further down-estuary during the wet year than in the dry or normal year. Their abundances, especially those of E. affinis, tended to stay above average for much of the year in 1996. Abundances of the copepod Acartia tonsa were highest in 2002, the dry year, and densities were elevated in regions where freshwater/low salinity species tend to dominate in the wet and normal years. Not only were zooplankton abundances and distributions related to environmental variables, but it was clear that biological interactions also influenced the spatial partitioning of the zooplankton community. Specifically, mysid predation is suggested, based on correlations with 18 both A. tonsa and E. affinis. Further, there appear to be interactions between A. tonsa and E. affinis themselves that may explain these species? distributions. Zooplankton taxa diversity tended to be lower during the fall when freshwater input was near or slightly below normal, and diversity in general was inversely correlated with salinity. INTRODUCTION The Susquehanna River discharges into the head of Chesapeake Bay and annually provides 50% of all freshwater input into the bay (Schubel and Pritchard 1986).The salt front created between this seaward flowing freshwater and landward moving tidal water is a site of enhanced suspended sediment concentrations, known as the estuarine turbidity maximum (ETM) region (Schubel 1968, Roman et al. 2001). The ETM region of Chesapeake Bay, located near the up-estuary limit of saltwater intrusion, is traditionally characterized by high total suspended solid (TSS) concentrations, high light attenuation, and high densities of zooplankton and anadromous fish larvae (North and Houde 2001, Roman et al. 2001). Though primary production tends to be low in this region (Wofsy 1983, Kemp and Boynton 1984, Fisher et al. 1988), heterotrophs are abundant, possibly supported by senescent freshwater phytoplankton, protozoa, free and particle-attached heterotrophic bacteria, and detritus (Irigoien and Castel 1997, Crump et al. 1998, North and Houde 2001). The calanoid copepod Eurytemora affinis often dominates the ETM zooplankton community (Heinle and Flemer 1975, Castel and Viega 1990, Kimmerer et al.1998). Eurytemora affinis is omnivorous, feeding on phytoplankton (Vijverberg 1980, Gasparini and Castel 1997), detritus (Heinle et al. 1977), protozoa 19 (Berk et al. 1978, Merrell and Stoecker 1998), and bacteria (Boak and Goulder 1983). It is usually considered to be an epibenthic copepod (Heinle and Flemer 1975, Simenstad 1990), and gravid females, which presumably are visible to visual predators, are often rare in the water column (Vuorinen 1987). Zooplankton distributions vary along the horizontal salinity gradient in ETM regions (Roman et al. 2001, Simenstad 1994, Irigoien et al. 1993) and vertically (Roman et al. 2001) in ETM regions and often move as the ETM itself moves with changes in freshwater inputs, winds, and tidal cycles. Although zooplankton samples have been collected monthly since 1984 by the Chesapeake Bay Monitoring Program (http://www.chesapeakebay.net/data/index.htm) at selected stations in the main stem of Chesapeake Bay (Jacobs et al. 1985), there have been few studies of species composition and abundances on small spatial scales (on the order of kilometers) in upper Chesapeake Bay. Such data can provide information on the effects of abiotic factors affecting zooplankton distribution and abundance and on biological interactions within the ETM region. High zooplankton densities in the ETM suggest that there are benefits of retention in this region, such as physiological optima (Hough and Naylor 1992, Kimmerer et al. 1998), avoidance of visual predators, and/or a greater reproductive output (Kimmerer et al. 1998, Roman et al. 2001). Hough and Naylor (1992) showed that E. affinis collected from portions of the Conwy estuary displayed tidally induced migratory behaviors to maintain themselves within a specific range of salinities. These behaviors varied depending on the sampling location and tidal range (spring versus neap). In tidal pools along the St. Lawrence estuary, E. affinis has been found 20 to migrate into areas of dense algal biomass along the shore during the day and back into the less turbid central regions during the night, presumably to avoid predation by sticklebacks (Castonguay and FitzGerald 1990). Eurytemora hirudinoides, a related species, performs diel vertical migration in the Archipelago Sea to escape predation (Vuorinen 1987). However, studies have also shown that some E. affinis populations do not actively select their location within estuaries, relying on passive behaviors (like a non-living particle) to remain in the ETM region (Kimmerer et al. 1998, Roman et al. 2001). The purpose of this paper is to describe the variability in abundance and distribution of different zooplankton groups in relation to each other and to abiotic factors within the Chesapeake Bay ETM region. The goal was to determine the primary factor(s) influencing zooplankton location and abundance, particular those of E. affinis. Five major zooplankton groups were identified whose abundance peaks in the ETM region at least part of the year. Principal component analyses suggest that physical and biological interactions influence distribution and abundance of these taxa. Linear and quadratic regression analyses suggest that larger E. affinis individuals may be able to control their location within the ETM and have a significant effect on autotrophic biomass. MATERIALS AND METHODS Mesozooplankton samples were collected using a multi-net,1-m2, 280-?m Tucker trawl in May (spring), July (summer), and October (autumn) of 1996, 2001, and 2002 in the Chesapeake Bay ETM region (Figs. 1 and 2) aboard the R/V Cape 21 Henlopen. Each net was opened at one of three depths (surface: ~0 to 4 m; middle: ~4 to 8 m; bottom: ~8 to 12 m) and deployed for 2 minutes at 5 to 7 stations along the channel of the ETM, for a total of 6 minutes spent sampling. The exceptions were the 1996 cruises, in which only two depth strata (surface and bottom) were sampled. The average volume filtered in a 2-minute deployment over all 9 cruises was 122 m3. Once onboard, the plankton samples were preserved in ethyl alcohol for future species identification and enumeration. Ctenophore (Mnemiopsis leidyi) and sea nettle (Chrysaora quinquecirrha) displacement biovolumes were also measured whenever collected. This sampling regime allowed for zooplankton composition and abundance to be quantified and compared spatially throughout the ETM region, between seasons, and between years with differing environmental conditions. Physical data were collected just prior to Tucker trawl deployment using a Seabird CTD equipped with a transmissiometer, fluorometer, and oxygen sensor. Zooplankton samples were split using a plankton splitter, and organisms were identified and enumerated under a dissecting microscope. Three separate subsamples were counted and averaged for each sample. For most subsamples, at least 200 individuals of the most abundant taxon were counted to minimize errors common to small sample sizes. In samples with the lowest densities of zooplankton, at least 50 individuals of the most abundant taxon were counted. The taxa enumerated included Eurytemora affinis (life stage, sex, and female reproductive state), Acartia tonsa (life stage and sex), Cyclops spp., barnacles (nauplii and cyprids), ostracods, polochaetes larvae, harpacticoid copepods, Podon spp., Bosmina spp., Daphnia spp., Leptodora kindtii, amphipods, mysids, decapod larvae (including crab zoea), mollusks (veligers), 22 and mites. The biological groups included in statistical analyses include numerical abundances of E. affinis, A. tonsa, cladocerans (Podon spp. + Bosmina spp., + Daphnia spp. + Leptodora kindtii), mysids (Neomysis spp.), amphipods (Gammarus spp.), chlorophyll-a (chl-a) concentrations and biovolume of the ctenophore Mnemiopsis leidyi (for July cruises only). Temperature, salinity, and total suspended solids (TSS) served as the physical variables, and river km (down-estuary from Havre de Grace) and water depth were also included in statistical analyses. Taxon diversity was calculated using the Shannon-Weiner index: H = - (P ? Log(P)), where H = diversity index and P = proportion of each taxon. This diversity index was chosen because it is sensitive to rare species (Odum 1983), and the data were skewed so that the two most abundant taxa for each cruise represented 97 to 98% of all individuals. Shannon-Weiner indices can also be used in mean-comparison statistical methods, such as ANOVA. Diversity is comprised of two components, richness (number of taxa per unit volume) and evenness (the apportionment of each individual among the taxa). Evenness (E) can be calculated from H using the formula: E = H/log(S), where S = the number of taxa. Log(S) represents the maximum possible diversity, and, assuming that the total number of taxa remains the same between treatments (cruises), E varies proportionately with changes in H (Odum 1983). Because of this relationship, results from analyses involving H can be inferred to correspond to E, and thus E is not considered further. 23 Because of the many variables measured, principal component analyses (PCA) was used to reduce the dimensionality of the data and to visually detect relationships between variables. Principal components analysis is a linear transformation approach to simplifying complex data sets. Principal components (PCs) are determined by plotting all data points (and their corresponding axes) in space (in this study, there are 11 axes) and running a new axis though the data ?cloud? so that this new axis accounts for the maximum amount of variance of the original data. This is the first principal component (PC 1). The second principal component (PC 2) is orthogonal (independent) to PC 1 and accounts for more variance than the third, PC, and so on. A biplot of PC 1 and PC 2 can visually help determine relationships between the original variables (Sokal and Rohlf 1994). The contribution of each variable to a PC (the eigenvectors) is known as the PC loadings, and these allow PCs to be characterized in terms of the original variables. All statistical analyses were conducted using JMP v. 5.01 (SAS Institute), and contour plots were created using Surfer v. 8 software (Golden Software). RESULTS Total daily streamflow values for May, July, and October in the lower Susquehanna River show 1996 streamflow to be above average for all months, with most days at or above the 37-year mean (Fig. 2). In May of 2002, the week prior to the sampling cruise had above-average freshwater streamflow, while streamflow during the week prior the May, 2001 cruise is below average. July streamflow conditions are reversed from the May pattern, with 2002 having low-flow conditions 24 preceding the cruise and 2001 streamflow being near average. October streamflow prior to the sampling cruises was below average during both 2001 and 2002. Physical variables change predictably throughout the year in response to temperature and changes in freshwater input (Table 1). Mesozooplankton annual comparisons by season May. Eurytemora affinis peak abundances in 1996 are five times greater than in 2001. It is possible that the 1996 sampling stations did not include locations with the highest densities of this copepod (Table I). Most zooplankton groups were located 20 to 30 km further up-estuary in 2001 compared to 1996, presumably due to decreased Susquehanna flow in 2001, and those groups favoring freshwater or low salinity (cladocerans, amphipods, E. affinis) are generally much more abundant in 1996 than in 2001 (Fig. 3 and 4). Eurytemora affinis is very abundant in May 2002 (>200 l-1) and peak about 20 km above the salt front (Fig. 5). Springtime cladoceran densities in 2002 are comparatively low and are divided between a Bosmina spp. population above the salt front and a Daphnia spp. population in freshwater. Podon spp. and Leptodora kindtii were found at very few stations and at densities never exceeding 0.01 individuals l-1, so are not considered separately in any analyses. Acartia tonsa densities in 2002 are at their highest springtime values measured during all three cruises (2.5 l-1; Fig. 5). July. Maximum densities of E. affinis in 1996 are >20 l-1, their highest summer values for all years (Table I, Fig. 6). These densities are comparable to maximum abundances of A. tonsa; under conditions of normal freshwater input (2001). Eurytemora affinis summer abundances decrease drastically from springtime 25 densities and are much lower than maximum A. tonsa values (Fig. 7). E. affinis abundances in 2001 are two orders of magnitude lower than 1996 values, while A. tonsa densities are at similar levels as 1996 and were greater down-estuary. Mysids and cladocerans are at lower concentrations in 2001, while amphipod densities increase from 1996 values (Fig. 7). Maximum densities of A. tonsa in 2001 are the highest measured out of all 9 cruises, peaking at values over 37 individuals l-1. Eurytemora affinis peaks further upstream in 2002 than in 1996 or 2001, presumably in response to differences in salinity structures between these years. Measured cladoceran abundances are also lower in 2002 than in 1996, possibly because their peak densities are landward of the survey area, and Daphnia spp. and Bosmina spp. distributions always overlap. Daphnia spp. is an order-of-magnitude more abundant than Bosmina in 2001 and 2002, and both genera are at equivalent densities in 1996. Amphipod and mysid concentrations are low in 2002 (Fig. 8). October. Eurytemora affinis in 1996 is at levels similar to those in July, likely due to high Susquehanna River streamflow in that year. Cladocerans and amphipods are also much more abundant in 1996 than in other years where they are often absent from October measurements (Fig. 9). In 1996, Bosmina spp. and Daphnia spp. distributions again overlap, but Bosmina spp. is generally an order-of-magnitude more abundant. Maximum A. tonsa densities in 1996 are between maximum values for 2001 and 2002 (Figs. 9, 10, and 11). It is possible that the peak in A. tonsa abundance during October 1996 occurred further down-estuary from the transect area due to the salt front being pushed further seaward compared to the other years. Low abundances were found for all zooplankton groups in 2001. The following October 26 (2002), A tonsa abundances were highest in October for all three years. Eurytemora affinis is conspicuously absent from the 2002 survey. Principal components analyses. Principal components (PC) analyses for each of the three months (for all three years) show varied relationships between organisms and their environment as the year progresses. In May, E. affinis is positively related to river km (distance down-estuary from Havre de Grace) and negatively related to chl-a concentration and TSS (Fig. 12). The orthogonal placement of E. affinis versus A. tonsa, mysids, salinity, and depth in the biplot suggest that there is no relationship between E. affinis and these variables. Acartia tonsa is positively related to salinity and negatively to dissolved oxygen (DO) concentration, which is likely due to this copepod species being found further down-estuary than E. affinis where DO concentrations tend to be lower. When Daphnia spp. and Bosmina spp. were considered separately in PC analyses, they were found to overlap with each other on all monthly biplots. They are therefore grouped as ?Cladocerans? for all PC analyses. In May, cladocerans are positively related to DO and negatively related to salinity. Temperature appears to have a minimal influence on any biological variable (Fig. 12). The greatest loadings in the first PC (PC1) come from physical variables and A. tonsa. Total suspended solids, E. affinis, river km, chl-a, and DO are strongly loaded in PC2 (Fig. 12). The PCA on July data again shows no relationship between copepod species (Fig. 13). Acartia tonsa abundance is unrelated to temperature, DO, and salinity, while E. affinis shows a positive relationship with temperature and DO and a negative relationship with salinity and ctenophore biovolume. Both A. tonsa and mysids have 27 negative relationships with chl-a. Cladocerans are negatively related to both river km and salinity. (Fig. 13). The loadings for PC1 are dominated by physical variables, and the biological variables (zooplankton and chl-a) are heavily loaded into PC2 (Fig. 13). The October PCA suggests that E. affinis, A. tonsa, river km, and amphipods all strongly correlated with one another; cladocerans showed a weak correlation with these variables. All other variables were orthogonal to the above variables and thus not related. Whereas TSS and DO were strongly related to each other, they were negatively related to salinity, depth, chl-a, temperature, and mysids, which were positively correlated (Fig. 14). With the exception of cladocerans, PC1 is again primarily influenced by physical variables. Biological and physical variables both influence PC2, with E. affinis and amphipods being loaded most heavily. Mesozooplankton seasonal comparisons by year. 1996. The spring Tucker trawl survey shows maximum densities of E. affinis around 70 l-1 that overlap with a population of A. tonsa with maximum densities an order of magnitude lower (Fig. 3). However, A. tonsa abundances generally peak further seaward in the mesohaline region of Chesapeake Bay. Amphipod abundance peaks further up-estuary. Cladoceran abundance peaks at the same distance down- estuary as amphipods, but the cladocerans occurred near the surface. Both amphipods and cladocerans are generally more abundant in the freshwater regions of Chesapeake Bay. All zooplankton populations shift 20 to 40 km upstream from spring to summer, presumably due to decreased freshwater input via the Susquehanna River (Fig. 6). Acartia tonsa abundance increases 10- to 100-fold in July; this copepod generally 28 experiences its abundance peak in the summer. Densities of E. affinis remain at levels comparable to springtime values. Copepod, amphipod, and mysid abundance peaks tend to overlap, with mysid abundance also extending further down-estuary. Cladoceran densities are at a maximum up-estuary of the salt front and are dominated by Bosmina spp (Fig. 6). Eurytemora affinis remains at the July abundance levels into October and is spatially separated from A. tonsa with very little overlap (Fig. 9). 2001. Maximum springtime E. affinis densities are less than in 1996, though still at the same order of magnitude. Cladocerans, amphipods, and E. affinis are all located landward of the salt front and are spatially separated from both A. tonsa and mysids, which peak further down-estuary (Fig. 4). In July, both cladocerans and amphipods are again found primarily on the freshwater side of the salt front, with E. affinis being found both up- and down-estuary of the salt front. Mysids and A. tonsa are located seaward of the salt front. Their summer abundances increase greatly from springtime densities, but peak density values cannot be determined because the transect did not extend far enough down-estuary to completely sample both populations (Fig. 7). Abundances of all zooplankton are lower in October, with cladocerans and amphipods completely absent in the sampled region. Because the salt front is landward of our most up-estuary station it is possible these zooplankton groups and E. affinis are limited to the lower salinity/freshwater of the Susquehanna River. Maximum turbidity and E. affinis lags behind the salt front for most surveys but does appear to follow its movement up- and down-estuary (Fig. 10). 2002. Maximum springtime densities of A. tonsa and E. affinis in 2002 are the highest of all three years sampled, with abundances of A. tonsa exceeding those of E. 29 affinis. All zooplankton are clustered between 30 and 40 km down-estuary with the exception of mysids, which do not appear until around 50 km. The zooplankton and turbidity are below the salt front, which was moving down-estuary due to an ebb tide (Fig. 5). The salt front in July is further up-estuary in 2002 than in either of the other two summer transects. Eurytemora affinis, mysids, and A. tonsa are distinctly separated from up-estuary to down-estuary, and A. tonsa densities are the highest recorded during the three sampling years. Maximum amphipod concentrations overlap with maximum mysid concentrations and are the lowest measured for amphipods during the three summers. Cladocerans again appear to be shifted up- estuary into fresher waters (Fig. 8). Very few amphipods and no E. affinis or cladocerans are found in the sampling area in October. Acartia tonsa and mysids are found at relatively high levels (especially A. tonsa) with patchy, non-overlapping distributions (Fig. 11). Diversity Because it is designed to test significant differences between the means of >2 treatments under conditions of unequal sample size, the Tukey-Kramer test was used to examine differences in taxa diversity between months (May, July, and October) and years (1996, 2001, and 2002). Diversity during the October 2001 and 2002 cruises was significantly lower than in May or July of 2001 and 2002. No other diversity comparisons between cruises were significant (Table II). An inverse semi- log regression analysis showed diversity to be negatively related to salinity (p < 0.0001, r2 = 0.48; Fig. 15). 30 Other analyses I was interested in how freshwater input into this region might affect the location and turbidity of the ETM and thus performed a regression between both turbidity (TSS) and distance down-estuary of maximum TSS values (ETM location) and Susquehanna River input (measured at the Conowingo dam). There is not a significant relationship between TSS or ETM location and streamflow for the same month as the cruises, but regressions between TSS or ETM location and Susquehanna River flow for the previous month were significant (Fig. 16a). Acartia tonsa and E. affinis peak abundance locations are related to freshwater input during the same month of study (Fig. 16b). An ANOVA comparing adult E. affinis female prosome length and the depth at which the females were collected shows that females nearest the bottom are significantly larger than those higher in the water column (Fig. 17). This analysis was based on data from the spring samples, when E. affinis was most abundant. Using log-transformed values for A. tonsa, E. affinis, and chlorophyll-a (after adding 1 to abundance/concentration measurements), I grouped chlorophyll-a data by month (May, July, or October) for 2001 and 2002 and regressed it against the log- transformed values of abundance for both copepod species. During the May peak in E. affinis abundance in both 2001 and 2002, log-transformed chlorophyll-a concentrations are negatively related to the log of E. affinis abundance (Fig. 18). To determine the decrease in chl-a entering the ETM via the Susquehanna River that can be attributed to dilution, the mean of peak concentrations of chl-a at maximum ebb tides (corresponding to chl-a maxima) and maximum flood tides (corresponding to 31 chl-a minima) were calculated from a 25-hour survey in the ETM region in May, 2002. These values were then used to create a linear regression of chl-a versus salinity to compare to the actual relationship of chl-a versus salinity for all May cruises (Fig. 19). Also, all 2001 and 2002 log-transformed chlorophyll-a data were negatively related to the log abundance of A. tonsa, though to a lesser degree than to E. affinis abundance (Fig. 20). DISCUSSSION This study examined zooplankton abundances in upper Chesapeake Bay on a smaller spatial scale than in previous studies. Using 15 years of monitoring data collected at fixed sites throughout Chesapeake Bay, Kimmel and Roman (2004) determined that freshwater input into the oligohaline region of Chesapeake Bay (via the Susquehanna River) is the primary environmental force affecting zooplankton abundances. Many of the zooplankton species/groups (i.e., E. affinis and cladocerans) have relatively narrow salinity tolerances and tend to track along a specific salinity range. For example, in May 1996, a year of above-average freshwater input, I found an abundance of zooplankton groups that favor low salinities or freshwater, and the salt front was located further down-estuary than during 2001 or 2002. It has been suggested that E. affinis abundance and distribution is influenced by both temperature and salinity, with increases in temperature decreasing the maximum salinity tolerance of this species (Kimmel and Bradley 2001). The same may hold true for other estuarine zooplankton, especially those that dominate in the freshwater/saltwater transition area. 32 Because E. affinis is often considered a major food source for larval fish (i.e. striped bass and white perch) and mysids, especially in the spring when anadromous fish spawn in the ETM region (North and Houde 2001, Winkler and Greve 2004; Shoji et al. 2005), a positive relationship between TSS and E. affinis abundance might represent an adaptation to reduce predation by visual predators (Gilmurray and Daborn 1981, Castel and Viega 1990, Roman et al. 2001). Female E. affinis collected during spring cruises were larger at depths greater than 8 m than higher in the water column. Eurytemora affinis is a brooding copepod, thus gravid females are especially visible to predation and may actively seek or become passively entrained into areas of reduced visibility (Roman et al. 2001). Large and/or gravid individuals may also appear to accumulate in deeper or higher turbidity waters due to greater predation outside of these regions. E. affinis abundance and turbidity were not positively correlated for any of the cruises. Morgan et al. (1997) also found that turbidity patterns in the Columbia River estuary did not match E. affinis distributions. If E. affinis actively migrates to maintain its position within the ETM region (Vuorinen 1987, Simenstad et al. 1994) or is more readily suspended by tidal currents than sediments (Roman et al. 2001), E. affinis abundance and turbidity would not be expected to coincide. The PCA for July shows a negative relationship between E. affinis abundance and salinity. As stated above, under high temperature conditions, E. affinis may modify its location to be in a general low salinity habitat rather than targeting high turbidity. To test this possibility, I categorized the mean salinity at each station, averaged over depth, as low (0-3), middle (3-6 ), or high (6+ ) and then used the nonparametric Wilcoxon 33 rank sums test to analyze E. affinis abundance each month grouped for all years versus salinity label. The May data show no difference in E. affinis abundance versus salinity label (2 = 2.31, p = 0.31), while E. affinis abundance was significantly higher in the ?low? salinity group than in either of the other salinity groups for both July (2 = 22.27, p < 0.0001) and October (2 = 17.97, p < 0.0001). Salinities between 0 and 5 have been shown to be optimal for E. affinis with the May temperatures ranges of the present study (12.8 to 19.6?C; Roddie et al. 1984, Soltanpou-Gargari and Wellershaus 1985). Nearly all of the study areas during the May cruises would be considered to have ?low? or ?middle? salinities, suggesting that most of this region is optimal habitat (in terms of temperature and salinity) for this copepod. Therefore, a relationship between copepod abundance and salinity in May would not be expected. These results suggest that E. affinis may display active migrations to maintain itself in a particular salinity range. Because our data represent snapshots of physical and biological variables in a dynamic environment, the lack of a significant relationship between E. affinis and salinity in the correlation analyses does not mean that salinity is unimportant in determining this copepod?s location and abundance in the ETM region (as suggested by the Wilcoxon ranked sums analyses). The lack of a direct relationship between the May E. affinis with turbidity and salinity, and the positive relationship with river km suggest that physical variables do not greatly influence E. affinis abundance in the ETM region in spring. With higher temperatures and reduced freshwater input in July the volume of habitat available to E. affinis declines, as does E. affinis abundance (Hoffmeyer 2004, Kimmel and 34 Roman 2004). In October, temperatures decrease to spring values while freshwater input remains close to July levels, and E. affinis concentrations either remain near July values or decrease. The effects of freshwater input on the location of the ETM (the TSS peak) and E. affinis abundance peaks as well as the relationship between E. affinis and salinity in July (when water temperature are greatest) show the importance of freshwater input to sediment and zooplankton dynamics in the Chesapeake Bay ETM region. The general distributional progression of zooplankton in the Chesapeake Bay ETM region from fresh- to saltwater (north to south) was the freshwater copepod Cyclops spp. (data not shown), amphipods, Daphnia spp., Bosmina spp., E. affinis, and A. tonsa, with mysids often covering a large salinity range. North and Houde (2003), working in the Chesapeake Bay ETM in 1998 and 1999, found a similar pattern between Bosmina longirostris and E. affinis. These distributional patterns and previous studies suggest that there are interactions between E. affinis and A. tonsa that help to define their distributions. Acartia tonsa has been shown to preferentially prey on the nauplii of other copepods species, including those of E. affinis, over its own nauplii (Londsdale et al. 1979). Eurytemora affinis may not be able to establish a population outside of the ETM due to such predation pressure. Live collections of A. tonsa from the field often show adults feeding on smaller copepodites, though it is not clear if this prey is of a different species (pers. obs.). Acartia tonsa has been shown to graze less efficiently on autotrophic prey than E. affinis in water collected from the ETM of the Choptank River, a Chesapeake Bay tributary (Lloyd, Chapter 3). A positive correlation between chlorophyll-a and turbidity in July, when A. tonsa 35 abundance peaks, may reflect a grazing rate decrease influenced by the concentration of suspended, inedible particles (Burdloff et al. 2002). Eurytemora affinis, however, appears be able to graze at similar rates on autotrophic prey in water from the ETM and water down-estuary (Lloyd, Chapter 3). In general, peak abundances of E. affinis, and A. tonsa did not overlap along the ETM transect in summer and fall, and in 2001, the distributions appeared to be spatially separated all three cruises. A. tonsa?s salinity tolerances are broader than E. affinis, and A. tonsa has been shown to tolerate salinities ranging from 0 to over 70 (Cervetto et al. 1999). Eurytemora affinis? salinity tolerances generally range from freshwater to around 15, though they have been found at higher salinities in the later winter/early spring (Vaupel-Klein and Weber 1975, Roddie et al. 1984). As noted earlier, increasing temperatures tend to decrease the salinity range of E. affinis (Kimmel and Bradley 2001). Acartia tonsa ingestion rates are possibly too low in the ETM to sustain a population, and E. affinis is limited in its down-estuary distribution by predation pressure from A. tonsa and mysids. Acartia tonsa, in turn, may be limited to surface and middle depths in the summer due to hypoxic conditions (< 2 mg O2 l-1) in bottom waters beginning around river km 40 and continuing seaward . The influence of predation on zooplankton abundance also may influence zooplankton abundance and distribution from summer to autumn as mysids and ctenophores increase in abundance in July and either remain at equivalent levels or decrease into October (i.e. Kimmel and Roman 2004). Whereas there is no significant correlation between E. affinis and mysids in July, there is a strong positive correlation in July between A. tonsa and mysid abundance that could represent the 36 two groups either coexisting without interaction or mysids actively seeking out copepod prey. Among zooplankton, mysids are especially capable swimmers (Kimmerer et al. 1998, Azeiteiro et al. 1999), are zooplanktivorous, and often select for E. affinis before most other prey (Viherluoto and Viitasalo 2001, Chigbu 2004). However, E. affinis has also been shown experimentally to have a greater escape response than Acartia spp. regarding mysid predation and may actually interfere with mysid predation on Acartia spp. (Viitasalo and Rautio 1998). October correlations are similar to July, where mysids and A. tonsa share opposite relationships with the same variables (though not all of these relationships are significant) and have a significant positive relationship with each other. The ctenophore Mnemiopsis leidyi is another predator that is often a large source of mortality upon crustacean zooplankton (Purcell et al. 2001) and may release microzooplankton (ciliates in particular) from copepod grazing mortality (Graneli and Turner 2002). Abundances of M. leidyi usually peak in the late summer into the fall at salinities > 4. Eurytemora affinis and other oligohaline zooplankton may be protected from ctenophore and sea nettle (which usually has a negligible volume in the study area compared to ctenophores) predation due to their salinity preferences and tolerances, as peak abundances of E. affinis and cladocerans in the July and October surveys occur at salinities below those where ctenophores are usually found. Summer PCAs showed a nearly non-existent relationship between both E. affinis and A. tonsa and ctenophores in the ETM region. This may not be the case in other areas of Chesapeake Bay as Purcell and Decker (2005) concluded that ctenophore abundance can significantly reduce A. tonsa abundance in years when populations of 37 the sea nettle, Chrysaora quinquecirrha, a predator of ctenophores, are forced down- estuary due to high freshwater input. Winkler et al. (2003) demonstrated a reduction of autotrophic biomass from freshwater into the low salinity zone of the St. Lawrence Estuary that was greater than a decrease caused by dilution and coincided with high E. affinis concentrations. This decrease in chlorophyll along the ETM occurred prior to any significant change in salinity or turbidity. Eurytemora affinis has been known to preferentially graze phytoplankton under conditions where autotrophic biomass is dwarfed by the mass of heterotrophs, detritus, and inorganic particles (Vijverberg 1980, Gasparini and Castel 1997). The inverse log-log relationship between E. affinis abundance and chlorophyll-a concentration in the May cruises of the present study (Figs. 19 and 20) is similar to results in the St. Lawrence Estuary. The logarithmic decline of chl-a concentrations versus E. affinis and the mismatch between chl-a concentrations and the dilution-based calculations suggest that the up- to down-estuary decrease in chl-a concentrations in May in northern Chesapeake Bay is possibly due to zooplankton grazing. A plot of chlorophyll-a versus A. tonsa abundance suggests that A. tonsa may have impacted autotrophic biomass via grazing during the July and October 2001 and 2002 cruises (Fig. 18). Research in Chesapeake Bay has demonstrated variability on the effects of A. tonsa grazing on phytoplankton standing stock (White and Roman 1992, Cuker and Watson 2002). Abundances of A. tonsa and phytoplankton (using chl-a concentrations as a proxy) in the Chesapeake Bay ETM region often do not overlap, suggesting that copepod grazing may significantly influence the abundance 38 of autotrophs. However, this relationship is much weaker than the relationship between chl-a concentration and E. affinis density, and it could be the result of chl-a dilution from freshwater (the Susquehanna River) to the ETM region, phytoplankton senescence and corresponding chl-a degradation in the ETM, and/or reduced primary production due to light limitation. Cruise comparisons of diversity show little change from May to July throughout the entire ETM region, but, under normal streamflow conditions, diversity does appear to decrease from July to October. The inverse log relationship between diversity and salinity suggests that, at salinities ranging from 0 to 14, diversity decreases from up-estuary to down-estuary. Zooplankton diversity in the St. Lawrence Estuary has been shown to be at a minimum at mesohaline locations (Laprise and Dodson 1994), and it is possible that zooplankton diversity increases from the middle bay to the bay mouth (Tafe 1990, Vieira et al. 2003). Salinities in the ETM region were higher during the October 2001 and 2002 cruises than during May or July for both of those years, and this hydrographic change may explain why diversity was lower during the October 2001 and 2002 cruises. Zooplankton distributions and densities appear to be primarily influenced by freshwater input and to a lesser extent through biological interactions. Eurytemora affinis is a species found in freshwater and brackish habitats. The seaward limit of other zooplankton, such as cladocerans and amphipods, moves up-estuary in July, presumably in response to both reduced streamflow and increased temperatures. High temperatures and salinities in the summer likely increase energy expenditures due to osmoregulation and increased respiration rates and may contribute to the 39 general decline of zooplankton diversity with salinity (Kimmel and Bradley 2001). Copepods, especially E. affinis, in the Chesapeake Bay ETM region appear to have a grazing impact on autotrophic biomass, suggesting that the Chesapeake Bay ETM is not necessarily a net heterotrophic system in the spring, even when turbidities are highest (i.e. Winkler et al. 2003). The filtering efficiency of cladocerans (Gliwicz 1980, Uitto 1996), feeding selectivity of copepods (Bollens and Penry 2003, Tackx et al. 2003), abundant detritus for amphipods (Zimmerman et al. 1979, Sanford et al. 2001), and zooplankton available to mysids creates a dynamic ecosystem in which a change in freshwater input can broadly impact the location and densities of autotrophic and heterotrophic plankton. 40 Cruise Depth dO 2 min max Temp min max Salinity min max Chl-a min max TSS min max May, 1996 S 9.3 8.2 11.0 15.8 15.4 16.5 1.8 0.1 4.0 2.4 2.3 2.3 25.2 12.0 45.1 B 6.9 5.6 10.8 14.2 12.8 16.1 5.6 0.2 8.0 2.8 1.5 5.0 40.7 29.3 63.7 July, 1996 S 7.0 6.4 8.9 25.0 24.6 25.7 3.3 0.3 5.8 8.0 8.0 8.0 63.3 46.4 251.8 B 4.7 2.1 6.2 24.1 22.9 25.6 8.3 0.4 11.9 8.0 8.0 8.0 22.3 1.7 45.5 October, 1996 S 10.3 9.2 12.3 14.1 12.5 15.6 0.3 0.1 4.0 1.9 1.9 2.0 32.4 17.1 51.6 B 7.4 6.8 11.9 15.7 12.8 16.5 5.7 0.2 9.1 1.9 1.8 1.9 16.8 10.9 32.7 May, 2001 S 7.0 4.2 14.2 18.5 17.5 19.5 0.6 0.2 1.9 1.9 1.6 4.5 20.0 15.2 27.2 M 7.0 3.4 7.9 18.2 16.8 19.5 1.1 0.2 4.0 2.1 1.7 2.1 19.6 18.4 32.8 B 6.6 5.3 7.3 17.6 15.0 19.6 2.2 0.2 7.8 2.2 1.8 2.9 40.6 26.4 52.4 July, 2001 S 5.9 5.8 6.0 26.0 25.7 26.4 1.0 0.2 3.1 2.1 1.9 2.5 16.6 15.1 17.4 M 5.9 4.8 5.9 26.0 25.5 26.6 1.2 0.2 4.4 2.5 2.0 2.6 16.7 14.8 19.9 B 5.7 2.1 5.9 25.6 24.2 26.7 1.6 0.2 9.7 2.5 2.2 2.7 28.8 20.8 51.7 October, 2001 S 6.8 6.9 7.4 16.4 20.2 20.7 5.6 4.4 10.9 2.1 1.7 2.8 6.7 4.1 10.2 M 6.8 5.4 7.0 16.4 15.9 17.1 4.6 2.0 10.8 2.2 2.1 2.6 6.5 4.0 11.7 B 6.4 5.4 6.7 16.6 16.2 17.1 6.4 2.4 10.8 1.9 1.7 2.7 11.7 7.6 19.4 May, 2002 S 7.3 6.7 9.9 16.6 16.1 17.0 0.7 0.1 4.2 4.4 3.3 8.5 206.5 200.0 224.0 M 6.7 4.2 9.4 16.5 14.4 17.0 1.3 0.1 10.9 5.0 2.9 9.3 172.4 153.7 212.1 B 5.4 3.3 7.2 14.7 13.5 16.9 8.9 0.1 14.8 2.5 2.3 10.3 114.5 65.3 126.5 July, 2002 S 6.3 6.0 6.9 27.0 25.9 27.1 1.8 0.3 5.4 4.9 2.4 5.8 5.7 4.1 7.7 M 5.9 4.2 6.2 26.3 25.3 26.5 4.7 2.6 7.2 5.1 2.1 5.9 6.0 2.8 12.0 B 3.8 0.4 4.1 25.8 23.0 26.0 4.9 4.7 15.0 2.0 1.5 2.5 11.1 10.3 31.6 October, 2002 S 7.1 6.9 7.4 20.7 20.2 20.7 7.3 4.4 10.9 2.3 1.7 2.8 6.1 4.1 10.2 M 7.0 6.4 7.1 20.9 20.8 21.0 8.0 6.4 13.9 2.6 1.8 2.9 8.9 5.3 9.7 B 6.9 6.2 7.1 21.0 20.9 21.2 8.3 6.6 14.9 2.5 1.8 2.7 12.2 8.3 21.1 TA BL ES Table 2.1 (part 1 of 2). Median, (bold values), minimum (min), and maximum (max) values for all abiotic variables and the most abundant zooplankton groups. dO = dissolved oxygen (mg l -1 ), Temp = Temperature (?C), Salinity = PSU, Chl-a = chlorophyll-a (g l -1 ). All zooplankton abundances are in number m -3 except for Jelly Volume (the sum of ctenophore and sea nettle volume, ml m -3 ). S = surface, B = bottom, M = medium. 41 Cruise Depth Cladocerans min max Amphipods min max Mysids min max Adult E. affinis min max Adult A. tonsa min max May, 1996 S 397.9 0.0 758.9 0.0 0.0 2.9 0.0 0.0 0.0 10587.4 1043.6 29053.8 0.0 0.0 890.1 B 105.7 0.0 352.9 0.0 0.0 5.7 0.0 0.0 0.0 6231.4 1448.6 9146.5 40.3 0.0 64.4 July, 1996 S 0.0 0.0 302.1 0.0 0.0 20.1 0.0 0.0 17.9 0.0 0.0 5518.4 4263.1 232.0 9280.7 B 0.0 0.0 470.7 0.0 0.0 156.9 21.2 0.0 172.2 21.2 0.0 22281.6 2217.7 56.7 34363.9 October, 1996 S 110.2 0.0 243.0 0.0 0.0 5.8 0.0 0.0 0.0 340.3 0.0 863.2 60.0 0.0 16522.3 B 2.3 0.0 264.8 0.5 0.0 66.2 0.0 0.0 0.0 48.6 0.0 25286.4 720.5 0.0 9990.2 May, 2001 S 31.0 7.9 70.2 6.9 0.0 28.6 0.0 0.0 0.0 529.9 264.2 1347.9 40.8 0.0 114.3 M 11.2 3.5 64.0 20.0 1.8 25.6 0.5 0.0 0.8 801.8 140.7 1992.4 10.7 0.0 409.7 B 19.2 6.0 186.4 22.1 6.4 186.4 3.5 0.0 30.6 2307.9 156.7 16673.6 28.1 0.0 1436.6 July, 2001 S 0.0 0.0 37.0 9.5 0.0 64.7 0.0 0.0 43.1 178.6 0.0 267.6 2379.0 0.0 12523.4 M 0.0 0.0 17.5 97.6 28.0 362.4 0.7 0.0 129.2 203.5 84.0 322.0 3996.5 0.0 37261.2 B 0.0 0.0 83.5 32.6 0.0 183.7 0.0 0.0 47.0 75.6 0.0 225.4 75.0 0.0 20690.1 October, 2001 S 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 13.8 0.0 0.0 0.0 862.1 36.9 1432.0 M 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 2.5 3.8 1.9 23.4 693.2 61.3 2694.7 B 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 10.6 1.4 0.0 72.1 1662.3 779.8 4992.8 May, 2002 S 57.4 9.3 242.1 10.1 0.0 27.8 0.0 0.0 5.0 340.9 64.4 986.1 241.4 0.0 538.0 M 81.9 0.0 378.0 17.8 0.8 37.3 0.0 0.0 0.9 598.2 25.8 1278.6 641.6 0.0 2492.4 B 30.5 0.0 138.4 19.9 3.3 138.2 0.0 0.0 32.9 1090.0 35.4 2418.2 675.3 0.0 3610.0 July, 2002 S 23.6 0.0 472.8 0.0 0.0 0.0 0.0 0.0 0.1 47.3 0.0 55.6 1237.7 703.1 19824.5 M 0.0 0.0 0.0 0.0 0.0 14.0 0.7 0.0 14.0 26.4 0.0 56.0 1575.5 1080.5 12545.7 B 0.0 0.0 83.3 0.0 0.0 0.0 0.0 0.0 4.7 93.6 0.0 353.9 1593.3 993.5 2612.8 October, 2002 S 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 13.8 0.0 0.0 0.0 50.8 36.9 1432.0 M 0.0 0.0 0.0 0.0 0.0 0.0 10.5 2.5 31.1 0.0 0.0 0.0 591.0 188.4 815.6 B 0.0 0.0 0.0 0.0 0.0 2.6 5.4 1.1 14.1 0.0 0.0 0.0 579.7 60.3 1319.5 Table 2.1 (part 1 of 2). Median, (bold values), minimum (min), and maximum (max) values for all abiotic variables and the most abundant zooplankton groups. dO = dissolved oxygen (mg l -1 ), Temp = Temperature (?C), Salinity = PSU, Chl-a = chlorophyll-a (g l -1 ). All zooplankton abundances are in number m -3 except for Jelly Volume (the sum of ctenophore and sea nettle volume, ml m -3 ). S = surface, B = bottom, M = medium. 42 Table 2.2. Shannon-Weiner diversity index means (H ? 1 SD). October 2001 and 2002 H values are significantly lower than May and July of the respective years and lower than October, 1996 (Tukey-Kramer means comparison,  = 0.05). May July October 1996 0.59 (0.39) 0.45 (0.31) 0.62 (0.26) 2001 0.76 (0.44) 0.63 (0.58) 0.06 (0.049) 2002 0.89 (0.13) 0.36 (0.41) 0.15 (0.046) 43 FIGURES Figure 2.1. Map of upper Chesapeake Bay showing the ETM region (shaded). Chester River Sassafras River Susquehanna River Patapsco River 76 76.5 76 76.5 39 39.5 L a t i t u d e ( ? N ) Longitude (?W) Choptank River 10 20 30 kilometers 44 Figure 2.2. Total monthly Susquehanna River streamflow as measured at the Conowingo dam for the three study years. The line is the monthly mean from 1968 to 2003. 0 100 200 300 400 500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1996 2001 2002 0 50 100 150 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1996 2001 2002 0 100 200 300 400 500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 1996 2001 2002 Day of Month T o t a l D a i l y S u s q u e h a n n a R . i n p u t ( ? 1 0 6 m 3 ) July October May 2001 2001 2001 1996 and 2002 2002 2002 1996 1996 45 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 1 5001 10001 15001 20001 25001 30001 35001 40001 45001 E. affinis 40 50 60 70 River km 1 1001 2001 3001 4001 5001 6001 7001 8001 9001 A. tonsa 40 50 60 70 River km -20 -10 -0 D e p t h ( m ) 1 76 151 226 301 376 451 526 601 676 Cladocerans 40 50 60 70 River km 1 21 41 61 81 101 121 141 161 181 Amphipods TSS Figure 2.3. May 1996 axial transect through ETM region. The lines in each plot represent isohalines. Crosses represent the mean depths of Tucker trawl samples. River km is the distance down-estuary (south) from Havre de Grace. Zooplankton abundances are in number of individuals m -3 . TSS = total suspended solids (mg l -1 ). All samples were taken at night. 46 -20 -10 -0 D e p t h ( m ) 1 5001 10001 15001 20001 25001 30001 35001 40001 45001 1 1001 2001 3001 4001 5001 6001 7001 8001 9001 1 21 41 61 81 101 121 141 161 181 10 20 30 River Km 1 5 9 13 17 21 25 29 33 37 10 20 30 River Km 1 21 41 61 81 101 121 141 161 181 10 20 30 River km -20 -10 0 D e p t h ( m ) 1 76 151 226 301 376 451 526 601 676 NTU E. affinis A. tonsa Cladocerans Amphipods Mysids TSS Figure 2.4. May 2001 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 47 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 1 251 501 751 1001 1251 1501 1751 2001 2251 E. affinis 1 1001 2001 3001 4001 5001 6001 7001 8001 9001 A. tonsa 10 20 30 40 50 River km -20 -10 -0 D e p t h ( m ) 1 76 151 226 301 376 451 526 601 676 Cladocerans 10 20 30 40 50 River km 1 21 41 61 81 101 121 141 161 181 Amphipods 10 20 30 40 50 River km 1 5 9 13 17 21 25 29 33 37 Mysids TS D e p t h ( m ) TSS Figure 2.5. May 2002 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 48 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 E. affinis 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 A. tonsa 20 30 40 50 60 70 River km -20 -10 -0 D e p t h ( m ) 1 101 201 301 401 501 601 701 801 901 Cladocerans 20 30 40 50 60 70 River km 1 41 81 121 161 201 241 281 321 361 Amphipods 20 30 40 50 60 70 River km 1 11 21 31 41 51 61 71 81 91 Mysids TSS Figure 2.6. July 1996 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 49 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 1 26 51 76 101 126 151 176 201 226 E. affinis 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 A. tonsa 10 20 30 40 River km -20 -10 -0 D e p t h ( m ) 1 11 21 31 41 51 61 71 81 91 Cladocerans 10 20 30 40 River km 1 41 81 121 161 201 241 281 321 361 Amphipods 10 20 30 40 River km 1 11 21 31 41 51 61 71 81 91 Mysids TSS Figure 2.7. July 2001 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 50 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 1 26 51 76 101 126 151 176 201 226 E. affinis 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 A. tonsa 10 20 30 40 River km -20 -10 -0 D e p t h ( m ) 1 101 201 301 401 501 601 701 801 901 Cladocerans 10 20 30 40 River km 1 5 9 13 17 21 25 29 33 37 Amphipods 10 20 30 40 River km 1 11 21 31 41 51 61 71 81 91 Mysids TSS Figure 2.8. July 2002 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 51 -20 -10 -0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 20 30 40 50 60 70 River km 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 A. tonsa 20 30 40 50 60 70 River km -20 -10 -0 D e p t h ( m ) 1 101 201 301 401 501 601 701 801 901 Cladocerans 20 30 40 50 60 70 River km 1 11 21 31 41 51 61 71 81 91 Amphipods 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 E. affinis TSS Figure 2.9. October 1996 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 52 10 20 30 40 50 River km -20 -10 0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 10 20 30 40 50 River km 1 26 51 76 101 126 151 176 201 226 E. affinis 1 2501 5001 7501 10001 12501 15001 17501 20001 22501 A. tonsa 10 20 30 40 50 River km -20 -10 0 D e p t h ( m ) 1 5 9 13 17 21 25 29 33 37 Mysids TSS Figure 2.10. October 2001 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 53 -20 -10 0 D e p t h ( m ) 1 21 41 61 81 101 121 141 161 181 NTU 10 20 30 40 River km -20 -10 0 D e p t h ( m ) 1 11 21 31 41 51 61 71 81 91 Amphipods 1 251 501 751 1001 1251 1501 1751 2001 2251 A. tonsa 10 20 30 40 River km 1 5 9 13 17 21 25 29 33 37 Mysids TSS Figure 2.11. October 2002 axial transect through ETM region. All samples were taken at night. See Figure 3 caption for an explanation of the plots. 54 Figure 2.12. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all May (1996, 2001, and 2002) data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. -0.2 -0.1 0.0 0.1 0.2 0.3 Comp.1 - 0 . 2 - 0 . 1 0 . 0 0 . 1 0 . 2 0 . 3 C o m p . 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1819 20 21 22 23 24 25 26 27 28 2930 31 32 33 34 35 36 37 3839 40 4142 43 -4 -2 0 2 4 6 8 - 4 - 2 0 2 4 6 8 Temp Amphipods Cladocerans d0 E. affinis River km Mysids A. tonsa Salinity Chl-a Depth TSS PC1 P C 2 -0.5 0 0.5 -0.5 0 0.5 Sa lt A. . to ns a De pth Ri ve r k m My sid s E. af fin is Am ph i TS S Ch l-a Cl ad oc era ns dO 2 Te mp PC1 PC2 E i g e n v e c t o r s ( l o a d i n g s ) 55 Figure 2.13. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all July data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. -0.2 -0.1 0.0 0.1 0.2 0.3 Comp.1 - 0 . 2 - 0 . 1 0 . 0 0 . 1 0 . 2 0 . 3 C o m p . 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1819 20 21 22 23 24 25 26 27 28 2930 31 32 33 34 35 36 37 3839 40 4142 43 -4 -2 0 2 4 6 8 - 4 - 2 0 2 4 6 8 Temp Amphipods TSS Cladocerans d02 E. affinis River km Chl-a A. tonsa Salinity Chl-a E. affinis A. tonsa Mysids TSS Chl-a River km Depth Salinity Amphipods Temp dO Cladocerans Jellies PC1 P C 2 -0.5 0 0.5 -0.5 0 0.5 Sa lt A. . to ns a De pth Ri ve r k m My sid s E. af fin is Am ph i TS S Ch l-a Cl ad oc era ns dO 2 Te mp PC1 PC2 E i g e n v e c t o r s ( l o a d i n g s ) 56 Figure 2.14. Principal components analysis (PCA) biplot of principal component 1 (PC 1) versus PC 2 for all October data based on correlations. The relative lengths of arrows represent the amount of variation explained by the two principal components, and arrow direction represents the type of relationship (arrows in same direction mean variables are positively related, arrows at right angle mean variables are unrelated). PC loadings for PC1 and PC2 are shown below the biplot. -0.2 -0.1 0.0 0.1 0.2 0.3 Comp.1 - 0 . 2 - 0 . 1 0 . 0 0 . 1 0 . 2 0 . 3 C o m p . 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1819 20 21 22 23 24 25 26 27 28 2930 31 32 33 34 35 36 37 3839 40 4142 43 -4 -2 0 2 4 6 8 - 4 - 2 0 2 4 6 8 Temp Amphipods TSS Cladocerans d02 E. affinis River km Chl-a A. tonsa Salinity Chl-a E. affinis A. tonsa TSS Chl-a River km Depth Salinity Amphipods Temp dO2 Cladocerans Mysids PC1 P C 2 -0.5 0 0.5 -0.5 0 0.5 Sa lt A. . to ns a De pth Ri ve r k m My sid s E. af fin is Am ph i TS S Ch l-a Cl ad oc era ns dO 2 Te mp PC1 PC2 E i g e n v e c t o r s ( l o a d i n g s ) 57 Figure 2.15. Relationship between Shannon-Weiner diversity index (H) and Salinity (S; psu). Diversity (H) was regressed on log salinity. Salinity D i v e r s i t y ( H ) H = -0.177*Ln(S) + 0.583 r2 = 0.48 0 0.5 1 1.5 0 4 8 12 16 58 Figure 2.16. a) Regressions between both maximum TSS and River km (ETM location; distance down-estuary from Havre de Grace) versus total monthly Susquehanna River streamflow in the month prior to the measurements. The TSS versus streamflow regression is the solid linear regression line and has an r2 of 0.49 (TSS = 6.0?10-9(Streamflow) + 24.1); the River km versus streamflow regression was calculated using log10(streamflow) and has an r2 of 0.42 (River km = 6.9 ? Log10(Streamflow) ? 110.0). Both regressions are significant at the 0.05 alpha level. b) Regressions between the locations of A. tonsa and E. affinis peaks versus total monthly Susquehanna River streamflow in the same month as the cruises. The A. tonsa peak location versus streamflow regression is the stippled line and has an r2 of 0.50 (A. tonsa = 5.0?10-9(Streamflow) + 35.6), and the E. tonsa peak location versus streamflow regression has an r2 of 0.61 (E. affinis = 7.0?10-9(Streamflow) + 14.0). Both regressions are significant at the 0.05 alpha level. 0 10 20 30 40 50 60 70 80 0 1 2 3 4 5 6 7 A. tonsa max E. affinis max 0 10 20 30 40 50 60 70 ETM location NTU max Streamflow (? 109 m3 month-1) R i v e r k m o r T S S R i v e r k m a b 59 Figure 2.17. Welch?s ANOVA between E. affinis prosome length and water depth [surface (0-4 m), middle (4-8 m), and bottom (8+ m) waters]. Prosome length is significantly greater in deeper waters (p < 0.05). Diamonds represent the 95% confidence interval. Bottom Middle Surface P r o s o m e l e n g t h (  m ) Depth 900 800 700 600 60 Figure 2.18. Regression between log-transformed chlorophyll-a concentration versus log-transformed E. affinis abundance for all 2001 and 2002 May data (p < 0.001). 0 0.2 0.4 0.6 0.8 1 1.2 0 1 2 3 4 5 Log10(E. affinis (no. m-3)) L o g 1 0 ( C h l - a (  g l - 1 ) ) 61 Figure 2.19. Chlorophyll-a (chl-a) and Eurytemora affinis (E. affinis) concentrations versus salinity. The dashed line represents the calculated chl-a concentration decrease by dilution alone; the solid line is the logarithmic fit to actual chl-a values versus salinity (p < 0.01). Salinity (PSU) 0 2 4 6 8 10 12 0 10 20 30 40 50 60 70 0 2 4 6 8 10 12 14 16 E . a f f i n i s ( i n d i v i d u a l s m - 3 ) C h l - a (  g l - 1 ) 62 Figure 2.20. Regression between log-transformed chlorophyll-a concentration versus log-transformed A. tonsa abundance for all 2001 and 2002 data (p < 0.05). 0 0.2 0.4 0.6 0.8 1 1.2 0 1 2 3 4 5 May July October Log10(A. tonsa (no. m-3 + 1)) L o g 1 0 ( C h l - a (  g l - 1 ) ) 63 LITERATURE CITED Azeiteiro, U. M. M., L. Jesus and J. C. Marques. 1999. Distribution, population dynamics, and production of the suprabenthic mysid Mesopodopsis slabberi in the Mondego Estuary, Portugal. Journal of Crustacean Biology 19(3):498-509. Berk, S. G., D. C. Brownlee, D. R. Heinle, H. J. Kling and R. R. Colwell. 1978. Ciliates as a food source for marine planktonic copepods. Microbial Ecology 4(1):27-40. Bollens, G. C. R. and D. L. Penry. 2003. Feeding dynamics of Acartia spp. copepods in a large, temperate estuary (San Francisco Bay, CA). Marine Ecology Progress Series 257(139-158. Burdloff, D., S. Gasparini, F. Villate, I. Uriarte, U. Cotano, B. Sautour and H. Etcheber. 2002. Egg production of the copepod Acartia bifilosa in two contrasting European estuaries in relation to seston composition. Journal of Experimental Marine Biology and Ecology 274(1):1-17. Castel, J. and J. Veiga. 1990. Distribution and retention of the copepod Eurytemora affinis hirundoides in a turbid estuary. Marine Biology 107(1):119-128. Castonguay, M. and G. J. FitzGerald. 1990. The ecology of the calanoid copepod Eurytemora affinis in salt marsh tide pools. Hydrobiologia 202(3):125-133. Cervetto, G., R. Gaudy and M. Pagano. 1999. Influence of salinity on the distribution of Acartia tonsa (Copepoda, Calanoida). Journal of Experimental Marine Biology and Ecology 239(1):33-45. Chigbu, P. 2004. Assessment of the potential impact of the mysid shrimp, Neomysis mercedis, on Daphnia. Journal of Plankton Research 26(3):295-306. Crump, B. C., J. A. Baross and C. A. Simenstad. 1998. Dominance of particle- attached bacteria in the Columbia River estuary, USA. Aquatic Microbial Ecology 14(1):7-18. Cuker, B. E. and M. A. Watson. 2002. Diel Vertical Migration of Zooplankton in Contrasting Habitats of the Chesapeake Bay. Estuaries 25(2):296-307. Fisher, T. R., L. W. Harding, Jr., D. W. Stanley and L. G. Ward. 1988. Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson estuaries. Estuarine, Coastal and Shelf Science 27(1):61-93. 64 Gasparini, S. and J. Castel. 1997. Autotrophic and heterotrophic nanoplankton in the diet of the estuarine copepods Eurytemora affinis and Acartia bifilosa. Journal of Plankton Research 19(7):877-890. Gilmurray, M. C. and G. R. Daborn. 1981. Feeding relations of the Atlantic silverside Menidia menidia in the Minas Basin, Bay of Fundy. Marine Ecology Progress Series. 6(3):231-235. Gliwicz, Z. M. 1980. Filtering rates, food size selection, and feeding rates in cladocerans - another aspect of interspecific competition in filter-feeding zooplankton. In W. C. Kerfoot (ed.), Evolution and ecology of zooplankton communities. Graneli, E. and J. T. Turner. 2002. Top-down regulation in ctenophore-copepod- ciliate- diatom-phytoflagellate communities in coastal waters: A mesocosm study. Marine ecology progress series 239:57-68. Heinle, D. R. and D. A. Flemer. 1975. Carbon requirements of a population of the estuarine copepod Eurytemora affinis. Marine Biology 31(3):235-247. Heinle, D. R., R.P. Harris, J.F. Ustach, and D.A. Flemer. 1977. Detritus as food for estuarine copepods. Marine Biology 40:341-353. Hoffmeyer, M. S. 2004. Decadal change in zooplankton seasonal succession in the Bahia Blanca estuary, Argentina, following introduction of two zooplankton species. Journal of Plankton Research 26(2):181-189. Hough, A. R. and E. Naylor. 1992. Endogenous Rhythms of Circatidal Swimming Activity in the Estuarine Copepod Eurytemora-Affinis (Poppe). Journal of Experimental Marine Biology and Ecology 161(1):27-32. Irigoien, X., J. Castel, and B. Sautour. 1993. In-situ grazing activity of planktonic copepods in the Gironde Estuary. Cahiers De Biologie Marine 34:225-237. Irigoien, X. and J. Castel. 1997. Light limitation and distribution of chlorophyll pigments in a highly turbid estuary: The Gironde (SW France). Estuarine, Coastal and Shelf Science 44(4):507-517. Jacobs, F., W. Burton, and I. Moss. 1985. Patterns of zooplankton abundance, species composition, and biomass in upper Chesapeake Bay. Estuaries 8:78A. Kemp, W. M. and W. R. Boynton. 1984. Spatial and temporal coupling of nutrient inputs to estuarine primary production: The role of particulate transport and decomposition. Bulletin of Marine Science 35:522-535. 65 Kimmerer, W. J., J. R. Burau and W. A. Bennett. 1998. Tidally oriented vertical migration and position maintenance of zooplankton in a temperate estuary. Limnology and Oceanography 43(7):1697-1709. Kimmel, D. G. and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Kimmel, D. G. and B. P. Bradley. 2001. Specific protein responses in the calanoid copepod Eurytemora affinis (Poppe, 1880) to salinity and temperature variation. Journal of Experimental Marine Biology and Ecology 266(2):135- 149. Laprise, R., J.J. Dodson. 1994. Environmental variability as a factor controlling spatial patterns in distribution and species-diversity of zooplankton in the St. Lawrence Estuary. Marine Ecology Progress Series 107(1-2):67-81. Lonsdale, D. J., D. R. Heinle and C. Siegfried. 1979. Carnivorous feeding behavior of the adult calanoid copepod Acartia tonsa Dana. J. Exp. Mar. Biol. Ecol. 36(3):235-248. Merrell, J. R. and D. K. Stoecker. 1998. Differential grazing on protozoan microplankton by developmental stages of the calanoid copepod Eurytemora affinis Poppe. Journal of Plankton Research 20(2):289-304. Morgan, C., JR Cordell, and CA Simenstad. 1997. Sink or swim? Copepod population maintenance in the Columbia River estuarine turbidity maxima region. Marine Biology 129:309-317. North, E. W. and E. D. Houde. 2001. Retention of White Perch and Striped Bass Larvae: Biological-Physical Interactions in Chesapeake Bay Estuarine Turbidity Maximum. Estuaries 24(5):756-769. North, E. W. and E. D. Houde. 2003. Linking ETM physics, zooplankton prey, and fish early-life histories to striped bass Morone saxatilis and white perch M. americana recruitment. Marine Ecology Progress Series 260:219-236. Peitsch, A., B. Koepcke and N. Bernat. 2000. Long-term investigation of the distribution of Eurytemora affinis (Calanoida; Copepoda) in the Elbe Estuary. Limnologica Jena 30(2):175-182. Purcell, J. E., J. R. White and M. R. Roman. 1994. Predation by gelatinous zooplankton and resource limitation as potential controls of Acartia tonsa copepod populations in Chesapeake Bay. Limnology and Oceanography 39(2):263-278. 66 Purcell, J. E., T. A. Shiganova, M. B. Decker and E. D. Houde. 2001. The ctenophore Mnemiopsis in native and exotic habitats: U.S. estuaries versus the Black Sea basin. Hydrobiologia 451:1-3. Purcell, J. E. and M. B. Decker. 2005. Effects of climate on relative predation by scyphomedusae and ctenophores on copepods in Chesapeake Bay during 1987-2000. Limnology and Oceanography. 50(1):376-387. Roddie, B. D., R. J. G. Leakey and A. J. Berry. 1984. Salinity-temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth Estuary. Journal of Experimental Marine Biology and Ecology 77(2):191-211. Roman, M., Anne Gauzens, WK Rhinehart, and JR White. 1993. Effects of low oxygen waters on Chesapeake Bay zooplankton. Limnology & Oceanography 38(8):1603-1614. Roman, M. R., D. V. Holliday and L. P. Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227 Sanford, L. P., S. E. Suttles and J. P. Halka. 2001. Reconsidering the physics of the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):655-669. Schubel, J. R. 1968. Turbidity maximum of the northern Chesapeake Bay. Science 161:1013-1015. Schubel, J. R. and D. W. Pritchard. 1986. Responses of Upper Chesapeake Bay to Variations in Discharge of the Susquehanna River. Estuaries 9(4A):236-249. Shoji, J., E. W. North and E. D. Houde. 2005. The feeding ecology of Morone americana larvae in the Chesapeake Bay estuarine turbidity maximum: the influence of physical conditions and prey concentrations. Journal of Fish Biology 66:1328-1341. Simenstad, C. A., L. F. Small and C. D. McIntire. 1990. Consumption processes and food web structure in the Columbia River estuary. Progress in Oceanography 25(1-4):271-297. Simenstad, C. A., C. A. Morgan, J. R. Cordell, and J. A. Baross. 1994. Flux, passive retention, and active residence of zooplankton in Columbia River estuarine turbidity maxima. Olsen & Olsen, Fredensborg, Denmark. Sokal, R. R. and F. J. Rohlf. 1994. Biometry, 3rd Edition. W. H. Freeman: New York 67 Soltanpou-Gargari, A. and S. Wellershaus. 1985. Eurytemora affinis -- one year study of abundance and environmental factors. Veroeff. Inst. Meeresforsch. Bremerh. 20(2):183-198. Tackx, M. L. M., P. J. M. Herman, S. Gasparini, X. Irigoien, R. Billiones and M. H. Daro. 2003. Selective feeding of Eurytemora affinis (Copepoda, Calanoida) in temperate estuaries: model and field observations. Estuarine, Coastal and Shelf Science 56(2):305-311. Tafe, D. J. 1990. Zooplankton and Salinity in the Rufiji River Delta, Tanzania. Hydrobiologia 208(1-2):123-130. Uitto, A. 1996. Summertime herbivory of coastal mesozooplankton and metazoan microplankton in the northern Baltic. Marine Ecology-Progress Series 132(1- 3):47-56. Vaupel-Klein, J. and R. E. Weber. 1975. Distribution of Eurytemora affinis (Copepoda: Calanoida) in relation to salinity: field and laboratory observations. Netherlands Journal of Sea Research 9:3-4. Vieira, L., U. Azeiteiro, P. Re, R. Pastorinho, J. C. Marques, and F. Morgado. 2003. Zooplankton distribution in a temperate estuary (Mondego estuary southern arm: Western Portugal). Acta Oecologica-International Journal of Ecology 24:163-173. Viherluoto, M. and M. Viitasalo. 2001. Effect of light on the feeding rates of pelagic and littoral mysid shrimps: a trade-off between feeding success and predation avoidance. Journal of Experimental Marine Biology and Ecology 261(2):237- 244. Viitasalo, M. and M. Rautio. 1998. Zooplanktivory by Praunus flexuosus (Crustacea : Mysidacea): functional responses and prey selection in relation to prey escape responses. Marine Ecology-Progress Series 174:77-87. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10(4):317-340. Vuorinen, I. 1987. Vertical migration of Eurytemora (Crustacea, Copepoda): A compromise between the risks of predation and decreased fecundity. Journal of Plankton Research 9(6):1037-1046. White, J. R. and M. R. Roman. 1992. Seasonal Study of Grazing By Metazoan Zooplankton in the Mesohaline Chesapeake Bay. Marine Ecology-Progress Series 86(3):251-261. 68 Winkler, G., J. J. Dodson, N. Bertrand, D. Thivierge and W. F. Vincent. 2003. Trophic coupling across the St. Lawrence River estuarine transition zone. Marine Ecology Progress Series 251(59-73. Winkler, G. and W. Greve. 2004. Trophodynamics of two interacting species of estuarine mysids, Praunus flexuosus and Neomysis integer, and their predation on the calanoid copepod Eurytemora affinis. Journal of Experimental Marine Biology and Ecology 308(1):127-146. Wofsy, S. C. 1983. A simple model to predict extinction coefficients and phytoplankton biomass in eutrophic waters. Limnology and Oceanography 28(6):1144-1155. Zimmerman, R., R. Gibson and J. Harrinton. 1979. Herbivory and detritivory among gammaridean amphipods from a Florida seagrass community. Marine Biology 54(1):41-47. 69 CHAPTER III: Incorporating interclutch duration into estimates of brooding copepod egg production ABSTRACT Egg production rates for copepods which brood their eggs are usually based on indirect methods which incorporate a temperature-dependent value for hatching time (= embryonic development) and the ratio of eggs to adult females. An alternative method for estimating brooding copepod egg production is described using the copepod Eurytemora affinis as an example. This method takes into account the life histories of brooding copepods and incorporates hatching time, interclutch duration, and mean clutch size into egg production estimates. Egg production estimated fromboth the clutch size method and the egg-ratio method were compared to estimates from three different empirical models of copepod production. The clutch size method consistently provided a better fit to predicted values from the models, with the egg-ratio method generally overestimating egg production compared to the clutch-size method. Although the clutch size method requires further research to be performed on the relationships between temperature and interclutch duration, it appears to provide more accurate estimates of egg production rates. The close fit between the somatic growth model outputs and egg clutch size-derived estimates suggests that egg production rates for E. affinis can be used to estimate juvenile growth under food-replete conditions. 70 INTRODUCTION Brooding copepods (those that carry eggs until hatching) differ from broadcast-spawning copepods (those that release unhatched eggs) in several ways, one being that a brooding female cannot produce a new clutch of eggs until after her present clutch hatches or the unhatched (usually dead) eggs are released (Runge and Roff 2000). This adds an important time component to egg production rate ? hatching time (HT). Temperature effects on HT (also known as egg development time or embryonic development time) have been the focus of several studies (e.g. Heinle and Flemer 1975, Vijverberg 1980, Ban and Minoda 1992, Andersen and Nielsen 1997). Essentially, HT is the time from extrusion of the first egg into an egg mass to hatching or release of the last egg in that mass. Hatching time values for Eurytemora affinis can be calculated from the equation: 1.04HT 36.8 T-= ? developed by Andersen and Neilsen (1997). This equation incorporates data from several studies and has a regression coefficient of 0.87. Another time component in brooding copepod egg production that must also be considered is interclutch duration (ID; Chow-Fraser and Maly 1991). Interclutch duration, which has receivedless study, is the period from the hatching of the last egg of a clutch to the extrusion of the first egg of another clutch. Ban (1994) found a significant, negative power relationship between temperature and E. affinis ID, while Vijverberg (1980) found reduced E. affinis ID at temperatures >25?C and reported ID values much lower than Ban (1994). Food limitation appeared to influence ID only at 71 near starvation levels (see Ban?s Table III). The above time periods (HT and ID) comprise the egg production cycle for brooding copepods (Fig. 1). Egg production measurements of broadcast-spawning copepod involve incubating animals in bottles filled with filtered water, natural water, or water with a food supplement (Runge and Roff 2000) over a period of 24 hours to capture any diel periodicity in egg production (White and Roman 1992). Measured HT for E. affinis ranged from over 14 days at temperatures below 5?C to ~1 day at 25?C (Andersen and Nielsen 1997). Such long hatching times at cooler temperatures mean that bottle incubation techniques may not be applicable for brooding copepods. One solution to this problem is to conduct incubations in large bottles (i.e.  4 l) containing male and female copepods and periodically remove subsamples to enumerate nauplii and eggs female-1 (Burdloff et al. 1999). Counts from these subsamples are then compared to subsamples taken at the beginning of the incubations. The major assumption of this method is that each subsample contains the same ratio of gravid to nongravid females and is a representative sample of nauplius concentration (nauplii are uniformly distributed). Thus, there is the potential for significant sample error with this method, and egg production rates appear low for this method compared to other methods [compare Burdloff et al. (1999) to Katona (1975) and Ban (1994)]. The ?egg ratio method? developed by Edmondson (1968) to estimate the egg production rates of rotifers has been applied to brooding copepods (e.g. Edmondson et al. 1962, Sabatini and Ki?rboe 1994). The egg ratio is egg abundance (E, per unit volume) divided by the total number of females with and without eggs (A, per unit volume), giving units of eggs female-1 (E/A). The egg ratio alone does not estimate 72 the production rate, but because HT can be determined from temperature at the time of copepod capture, egg production rate can theoretically be determined (Runge and Roff 2000). Assuming that all eggs are viable, egg production rate (EP, eggs female-1 day-1) should equal the egg ratio divided by hatching time: . This definition assumes that ID is incorporated into E/A, because A includes all gravid and nongravid adult females. Therefore, the only time component one needs to estimate EP using this method is HT. Interclutch duration and hatching time could be used directly in EP measurements by dividing the mean clutch size (CS; of completely formed clutches) by the sum of both time components: . where CS = mean clutch size of fully-formed clutches. This method assumes that all mature females are reproductively capable., The measured mean clutch size represents the average potential egg output of all females, thus only gravid females must be included in the calculation. Traditional methods of zooplankton collection (nets, pumps) may undersample bottom depths within the water column (Omori and Hamner 1982), and because E. affinis is often considered an epibenthic copepod, with the proportion of gravid females increasing with depth (Kimmerer et al. 1998) or light attenuation (Castonguay and FitzGerald 1990), measured E/A may not represent the actual population E/A. This would lead to inaccurate estimates of EP. Egg production estimates based on the clutch size method are not dependant on ratios and are also CSEP (ID HT)  =  +   EEP A(HT)  =    73 well-suited for conditions of low copepod abundances, where ratios may be especially misleading. One purpose of this study is to provide a temperature-based model to estimate E. affinis interclutch duration (ID). These values can then be used to provide an EP estimate based directly on the complete reproductive cycle of E. affinis (CS) and can be compared to traditional E/A derived estimates of EP. Such comparisons suggest that E. affinis EP estimates based on the CS method provide a better fit to several empirical models of copepod growth rates than do estimates based on the E/A method. Further, the relationships between the model outputs and actual data suggest that EP can be used as proxy for somatic growth of E. affinis under food satiating conditions MATERIALS AND METHODS Interclutch duration experiments Live E. affinis were collected by a horizontal net tow (mesh size = 200 m) from the Horn Point Laboratory dock on the Choptank River (38.590 N/76.139 W, Cambridge, Maryland, USA), the catch was diluted with ambient water and the stored in a cooler. A 20-l carboy of 200-m filtered water was also collected on site. In the laboratory, four, 2-l aspirator bottles were each filled with 200-m filtered water, and approximately 250 ml of water from beakers containing copepods. The four bottles were placed into one of four environmental chambers set at 5?C, 13.6?C, 22.6?C, and 27.5?C. The bottles were gently bubbled (~1 bubble s-1) and left 74 undisturbed for 24 h under a 12:12 day/night cycle. The ambient water temperature at collection was 10.1?C. Following the 24 h temperature acclimation, 14 male and gravid female pairs from each temperature regime were collected using a wide-bore Pasteur pipette and each pair placed into 50 ml beakers containing 20 ml of 200-m filtered water from the Choptank River. To ensure that the copepods in natural water (NW) were not food limited, the diatom Thalassiosira weissflogii(Tw) was added to four of the fourteen beakers in each of the temperature treatments to achieve a final concentration of 5*104 cells ml-1 (NW+Tw). Diatom concentration was determined by a regression between fluorescence and cell number using a Coulter Counter (r2 = 0.98). An ANOVA was performed using JMP v. 5.0.1 software (SAS Institute) to test for differences in ID between NW and NW+Tw. The incubation beakers were maintained in their respective environmental chambers under a screen to reduce light to ~20% incident levels. Once daily, 80% of the water in each beaker was replaced with 200-m filtered Choptank River water, with care not to pour out the copepods, and T. weissflogii added to the appropriate beakers. Copepods in the 27.5?C and 22.6?C treatments were checked under a dissecting microscope at least every 5 hours for survival and the presence of eggs. Copepods in the 5?C and 13.6?C treatments were checked at least every 8 hours. When the last egg hatched from an egg mass was observed, a female was noted as being nongravid (in the ID phase) and was considered nongravid until the first egg of the next egg mass appeared. Any dead male or female copepods, were replaced using E. affinis cultures maintained at the same temperature and food rations. 75 Mortality was high at 27.5?C, and egg development data was collected for only four copepod pairs in the NW treatment and two pairs in the NW+Tw treatment. Measured egg production versus model predictions Eurytemora affinis were collected in 2002 (Chesapeake Bay) and 2003 (Choptank River) using a plankton net with a 200 m mesh. Egg production rates estimated by the E/A and CS methods were compared to three copepod growth models. The Huntley and Lopez (1992) model assumes that the weight-specific growth rate (G, d-1) of copepods is solely dependant on temperature (T, in ?C) and using published growth rates at different temperatures, they developed the model: . This is the simplest of the three models, as it is independent of copepod size and food concentration. Ikeda and Motoda (1978) developed a copepod growth model predicting G in terms of habitat temperature (T) and individual copepod dry weight in mg (W) and in g carbon (Wc): . . This growth equation is derived from an estimate of respiration that is converted to carbon using a 0.8 respiratory quotient. This model assumes an assimilation efficiency of 0.7 (Conover 1978) and a gross growth efficiency of 0.3 (Roman et al. 2000, Ikeda and Motoda 1978, Hirst and Sheader 1997). Copepod W (mg) for the Ikeda-Motoda model was calculated from the equation (0.111T)G=0.0445e (0.02538T-0.125) (-0.01089T-0.818) c 7.714?10 ?WG= W 76 , where PL is prosome length in m (B?ttger and Schnack 1986). Carbon weight (in g) was calculated using the equation , determined by Kankaala and Johansson (1986). The third growth model is a multiple linear regression developed by Hirst and Sheader (1997): with temperature (T) and copepod body carbon (Wc) This model was developed with a meta-analysis approach using multiple copepod species growth data collected from the poles to the tropics. As copepod mass (Wc) increases this model predicts reduced growth rates. These growth models were developed using primarily somatic growth data. Comparing them to growth data derived from egg production measurements allows for 1) a comparison between the E/A and CS methods for estimating egg production and 2) a verification of the use of EP as a proxy for somatic growth, which is a common use for EP estimates (Runge and Roff 2000). RESULTS The interclutch duration experiments indicated that ID has a strong negative power relationship with temperature that is similar to Ban?s (1994) data (Fig. 2), and is described by the equation -1.523ID=161.62?T . 10 10 clog (g)=0.0246T-(0.2962Log (W ))-1.1355 Log(W)=2.11?Log(PL)-5.29 2.83C=6.25?(PL/1000) 77 Interclutch duration is apparently not food limited, as the NW+Tw treatments did not differ from NW treatments (Fig. 3). Therefore, NW+Tw treatment data with NW data were grouped in the analysis of ID versus temperature (Fig. 2). Summing HT (from Andersen and Nielsen 1997) and ID (from the present study) and plotting versus temperature gives the equation -1.3114TC=169.07?T , where TC = total reproductive cycle length (Fig. 4). The, 2002 weight-specific EP values for Chesapeake Bay fit each of the three growth models better than did the 2003 EP data from the Choptank River, and CS- based estimates have higher coefficients of determination than do E/A-based calculations when plotted versus model predictions (Table 1). The Huntley-Lopez (1992) temperature-based model demonstrates greater EP predictability for 2002 when the CS method is used compared to the E/A method; the 2003 regressions for both methods are not significant (Table 1). The Huntley-Lopez model is a slightly better EP predictor for these data than is the Ikeda-Motoda model, although the coefficients of determination for the two are similar. In 2002, the CS method greatly improves the model fit over the E/A approach for the Ikeda-Motoda model (Table 1). The Hirst-Sheader model shows the best fit for all data and is the only model that significantly fits any 2003 data (CS-derived EP; Table 1, Figure 5). CS-derived EP was calculated using data from Ban (1994) and Crawford and Daborn (1986) and plotted these with 2002 and 2003 data (Figure 5). The Crawford and Daborn (1986) EP values were calculated from measurements of E. herdmani. The three data points from Ban (1994) were measured under satiating food conditions 78 and complement the 2002 data. With the Ban (1994) and 2002 data sets plotted together, the slope between actual EP and modeled EP is 0.90. A comparison of EP estimates derived from the E/A method versus the CS method suggests that as EP increases, the E/A method provides slightly higher EP estimates than the CS method (Figure 6). DISCUSSION Previous research suggests that temperature is the primary variable affecting the time components of the brooding copepod reproductive cycle. Hatching time is probably not affected by food limitation because the eggs are already formed and extruded at the beginning of the measurement. Ban (1994) demonstrated that ID remains relatively stable at a given temperature until food concentrations are nearly nonexistent. The addition of T. weissflogii had no effect on ID in the E. affinis used in our experiments, supporting the idea that ID was not generally food limited. In field studies, E. affinis clutch sizes as well as female size tend to decrease with increasing temperature (Ban 1994, Hirche 1992). Thus, temperature effects are confounded by the strong relationship between female size and clutch size. It is possible that under satiating food conditions, E. affinis of a given size is ?programmed? to produce an egg clutch of particular number regardless of temperature. Therefore, laboratory estimates of ID may not reflect actual ID in the field if the copepods being incubated at the different temperatures are the same size, because larger clutches may take longer to produce. If this is the case, then ID values for this experiment may be overestimated for temperatures >10?C and slightly 79 underestimated for temperatures <10?C because all females used in the experiments were collected at ~10?C, and their sizes (and thus clutch sizes) would be expected to represent the temperature of collection (mean female prosome ? SD = 935.1 ? 80.3 m) instead of their incubation temperatures. Ban (1994) used E. affinis hatched and raised at the experimental temperatures, and his ID patterns are similar to those of this study (Figure 3). Any size effect on E. affinis clutch size appears to exert a minimal control on ID. Therefore, the ID-temperature relationship can be used for field estimates of EP when using the clutch size method. The Ikeda-Motoda model (incorporating temperature and copepod weight) is the most complex of the three models, but predicts G (= EP) no better than the Huntley-Lopez model (incorporating temperature only). This is likely because the Ikeda-Motoda model is a more indirect estimate of EP, having a number of assumptions and calculations prior to calculating a growth value. The high predictability of the Huntley-Lopez model is due in part to the indirect incorporation of female size into the calculation. Generally, water temperature is similar during both an individual?s somatic development and its reproductive period, and the temperature value in the equation accounts for both of these factors. Both models also rely on a temperature-based power function, which may also add to their similarity. The Hirst-Sheader model is a multiple linear regression using log-transformed data to create linear relationships between predictors (temperature and copepod weight) and the regressor (G). This model provides the best fit for all data regardless of year or EP estimation method but is not better than the Huntley-Lopez model. The 80 Wc in this dataset covers a size range of 3.9 g, while the Hirst-Sheader model covers a Wc size range of 511 g. Thus, the Wc range in this study?s E. affinis dataset may simply be too small to capture the effects of Wc on EP. There were large differences in the model fits between 2002 (Chesapeake Bay) and 2003 (Choptank River; Figure 5). Preliminary analysis of available food in both of these regions suggests that these animals were not food limited (Chapter 3). 2002 was a warm, dry year compared to 2003, which had unseasonably cool temperatures and was one of the wettest years on record in this region (http://water.usgs.gov/). Eurytemora affinis flourishes in cool, wet conditions (Kimmel and Roman 2004), and abundances in both systems were notably higher in 2003 than in 2002. Enhanced predation on large, gravid E. affinis females in 2003 by greater number of anadromous fish larvae may increase the proportion of smaller females and/or smaller clutch sizes in a population, which would cause underestimates of mean clutch size, the egg-ratio, and EP. Whatever the cause of the year/system differences, all three models predict CS-derived EP estimates better than EP estimated by the traditional E/A approach. I propose that the CS method be included in any study measuring brooding copepod EP along with either E/A estimates or incubation measurements to allow for further comparisons of the various methods of EP estimation. Specifically, the equation for TC can be used to estimate E. affinis EP once a mean sample clutch size of fully-formed clutches is determined: . The relationship between E/A-derived EP and CS-derived EP supports the idea that the E/A may provide overestimates of EP. The tendency for E. affinis to be found CSEP= TC 81 near the bottom where they are often not sampled, suggests that gravid females would be undersampled and EP thus underestimated. Not all E. affinis populations appear to have the same innate reproductive time components (Andersen and Nielsen 1997). Ban (1994) in Japan and Vijverberg (1980) in the Netherlands found different ID values versus temperature for E. affinis. Andersen and Nielsen (1997) point out that rate differences between populations of a species are often far fewer than between species. Therefore, extrapolating ID values from one population to another for a particular species can be justified. Ideally, estimates of ID versus temperature for a species will include data from many different sub-populations. A large potential error in the clutch-size method of EP estimation is from the determination of mean CS. The standard deviations of E. affinis clutch size sample means from Chesapeake Bay represent from 25 to 50% of the means and are influenced by female prosome variability, whose standard deviations vary by up to 10% of the mean. If incomplete egg masses are used to estimate mean CS, EP will be underestimated. Under food-saturating conditions, the relationship between CS and prosome length (PL) in Chesapeake Bay is linear and has an r2 of 0.85. A complete or nearly complete clutch can also be identified by its hemispheric appearance (Fig. 7). If food is limiting, clutch size would still be expected to be related to prosome length but would have a lower mean number of eggs. Because these growth models were developed using primarily juvenile growth data, E. affinis EP (an easily measured rate) appears to provide a good estimate for E. affinis somatic growth (a more difficult rate to measure) in Chesapeake Bay. Hirst 82 and Sheader (1997) performed a meta-analysis using their multiple regression model and literature values that included copepod juvenile growth and egg production, obtaining an r2 of 0.64 (r2 = 0.74 for juvenile growth data only). The coefficient of determination using the same model with CS-derived E. affinis EP data was 0.88 for 2002 in Chesapeake Bay. All three growth models provide accurate estimates of E. affinis EP in 2002 and can be used to predict EP, with the Hirst-Sheader model providing the best fit. Copepods are not usually food limited in Chesapeake Bay (White and Roman 1992), and, under such conditions, reproductive rates should serve as reliable proxies for somatic growth (Hopcroft and Roff 1998). This research supports the use of egg production rates as a measure for juvenile growth for E. affinis in Chesapeake Bay. Similar to hatching time, interclutch duration appears to be primarily affected by temperature. The addition of the diatom supplement did not have any effect on E. affinis interclutch duration, suggesting that this time component was not food-limited in this experiment. Clutch size-derived EP estimates provided a better fit to the copepod growth models than the egg-ratio estimates. The data also suggest that gravid females were slightly oversampled compared to the rest of the population during high production periods (usually late spring, early summer), i.e., the sample E/A is greater than population E/A (see Figure 6). As noted above, the vertical depth separation often observed between gravid and nongravid E. affinis does not support the use of the E/A method in estuarine systems (Kimmerer et al. 1998, Castonguay and FitzGerald 1990). 83 The better fit to the independent growth models as well as avoiding sampling errors associated with the E/A method, support the use of the clutch size method in addition to the egg-ratio method for estimating brooding copepod reproductive rates. Continued use of the egg-ratio method will allow future data to be compared to previous studies of copepod egg production. Caution should be taken when estimating EP with the clutch size method under low food conditions, as laboratory estimates of ID may not be representative of field ID. Further, it must be noted that because of the poor fit of the 2003 data to growth models, indirect methods of estimating copepod egg production, namely the egg-ratio method and clutch size method, may not always be appropriate, and incubations may provide better estimates. Egg production rates have often been used to estimate somatic growth due to the general ease of determining EP (Berggreen et al. 1988). This approach assumes that any resources that would go into somatic growth for juveniles would be allocated to reproductive growth in adults. Our observations in the Chesapeake Bay suggest growth of E. affinis based on egg production can be used to estimate juvenile somatic growth. 84 TABLE Table 3.1. Results of linear regressions performed between model outputs of egg production (EP) versus EP estimates by the egg-ratio (E/A) and clutch size (CS) methods. m = regression slope; b = Y-intercept. * p < 0.05 ** p < 0.01 *** p < 0.001 Year Model EP method m b r2 2002 Huntley and Lopez (1992) E/A 1.44 0.014 0.54** CS 2.61 -0.16 0.82*** Ikeda and Motoda (1978) E/A 1.41 -0.045 0.52** CS 2.64 -0.18 0.78*** Hirst and Sheader (1997) E/A 0.742 -0.36 0.62** CS 0.99 -0.13 0.84*** 2003 Huntley and Lopez (1992) E/A 0.37 0.29 0.06 CS 1.58 0.10 0.15 Ikeda and Motoda (1978) E/A 0.31 0.20 0.07 CS 1.39 0.04 0.17 Hirst and Sheader (1997) E/A 0.20 -0.71 0.09 CS 0.32 -0.59 0.32* 85 FIGURES Figure 3.1. Conceptual diagram of reproductive cycle of E. affinis showing the two major time components, hatching time and interclutch duration. Hatching time is primarily influenced by temperature, and a similar relationship with temperature is presumed for interclutch duration. 86 Figure 3.2. Relationship between interclutch duration (days) and temperature (?C). Ban (1994) data ? 1 SD. Fitted power function r2=0.98 (does not include Ban (1994) data). 87 Figure 3.3. Comparison of Interclutch Durations for copepods fed 200-m filtered Choptank River water (NW) and those fed Choptank River water plus a Thalassiosira weissflogii supplement (NW+Tw) at the four experimental temperatures. 88 Figure 3.4. Graphical representation of the time components of E. affinis? reproductive cycle. HT = hatching time; ID = interclutch duration; TC = total cycle length (sum of HT and ID). 89 Figure 3.5. Hirst-Sheader (H-S) model weight-specific EP predictions versus actual weight-specific EP. (a) Actual EP values calculated by the egg-ratio (E/A) method. The dashed line is the linear fit to of H-S model output to 2002 data. 2003 data versus predicted values was not significant. (b) Actual EP values calculated by the mean clutch size (CS) method. The dashed line is the linear fit to of H-S model output to 2002 data. Linear fits to both 2002 and 2003 data were significant. ?C & D (1986)? = Crawford and Daborn (1986). See Table 1 for all linear regression statistics. Predicted EP = 0.754(Actual EP) + 0.002 Predicted EP = 0.427(Actual EP) + 0.051 90 Figure 3.6. Comparison of EP rates estimated by the egg-ration method (E/A) and the mean clutch size method (CS). Dashed line is the 1:1 relationship. 91 Figure 3.7. Dorso-lateral view of gravid E. affinis showing hemispheric nature of the egg mass. The two darker regions of the egg mass (em) are separated by the cleft (light-colored region) below the attached spermatophores (sp). The urosome (ur) is lifted dorsally in this photograph but would normally rest in the cleft region. 92 LITERATURE CITED Andersen, C. M., and T. G. Nielsen. 1997. Hatching rate of the egg-carrying estuarine copepod Eurytemora affinis. Marine Ecology Progress Series 160:283-289. Ban, S. H. 1994. Effect of temperature and food concentration on postembryonic development, egg-production and adult body-size of the calanoid copepod Eurytemora-affinis. Journal of Plankton Research 16:721-735. Ban, S., and T. Minoda. 1992. Hatching of diapause eggs of Eurytemora-affinis (Copepoda, Calanoida) collected from lake-bottom sediments. Journal Crustacean Biology 12:51-56. B?ttger, R., and D. Schnack. 1986. On the effect of formaldehyde fixation on the dry weight of copepods. Meeresforschung/Rep. Mar. Res. 31:141-152. Burkill, P. H., and T. F. Kendall. 1982. Production of the copepod Eurytemora affinis in the Bristol Channel. Marine Ecology Progress Series 7:21-31. Burdloff, D., S. Gasparini, B. Sautour, H. Etcheber, and J. Castel. 1999. Is the copepod egg production in a highly turbid estuary (the Gironde, France) a function of the biochemical composition of seston? Aquatic Ecology. 34:165- 175. Castonguay, M., and G. J. FitzGerald. 1990. The ecology of the calanoid copepod Eurytemora affinis in salt marsh tide pools. Hydrobiologia 202:125-133. Chow-Fraser, P., and E. J. Maly. 1991. Factors governing clutch size in two species of Diaptomus (Copepoda: Calanoida). Canadian Journal of Fisheries and Aquatic Sciences. 48:364-370. Conover, R. 1978. Transformations of organic matter. In Kinne, O. (ed.), Marine Ecology, vol. 4, Dynamics. Wiley, Chichester, pp. 221-499. Crawford, P., and G. R. Daborn. 1986. Seasonal variations in body size and fecundity in a copepod of turbid estuaries. Estuaries 9:133-141. Edmondson, W. T. 1968. A graphical model for evaluating the use of the egg ratio for measuring birth and death rates. Oecologia 1:1-37. Edmondson, W. T., G. W. Comita, and G. C. Anderson. 1962. Reproductive rate of copepods in nature and its relation to phytoplankton population. Ecology 43:625-634. Fockedey, N. and J. Mees. 1999. Feeding of the hyperbenthic mysid Neomysis integer in the maximum turbidity zone of the Elbe, Westerschelde and Gironde estuaries. Journal of Marine Systems. 22:2-3. 93 Heinle, D. R., and D. A. Flemer. 1975. Carbon requirements of a population of the estuarine copepod Eurytemora affinis. Marine Biology. 31:235-247. Hirche, H. J. 1992. Egg-production of Eurytemora-affinis - effect of K-strategy. Estuarine Coastal and Shelf Science 35:395-407. Hirst, A. G., and M. Sheader. 1997. Are in situ weight-specific growth rates body-size independent in marine planktonic copepods? A re-analysis of the global syntheses and a new empirical model. Marine Ecology Progress Series. 154:155-165. Hopcroft, R. R., and J. C. Roff. 1998. Zooplankton growth rates: The influence of female size and resources on egg production of tropical marine copepods. Marine Biology 132:79-86. Huntley, M. and M. D. G. Lopez. 1992. Temperature-dependent production of marine copepods: a global synthesis. The American Naturalist 140:201-242. Ikeda, T., and S. Motoda. 1978. Estimated zooplankton production and their ammonia excretion in the Kuroshio and adjacent seas. Fish. Bull. 76:357-367. Kankaala, P., and S. Johansson. 1986. The influence of individual variation on length- biomass regressions in three crustacean zooplankton species. Journal of Plankton Research 8:1027-1038. Katona, S. K. 1975. Copulation in the copepod Eurytemora affinis (Poppe, 1880). Crustaceana 28:89-95. Kimmel, D. G., and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Kimmerer, W. J., J. R. Burau, and W. A. Bennett. 1998. Tidally oriented vertical migration and position maintenance of zooplankton in a temperate estuary. Limnology and Oceanography. 43:1697-1709. Mauchline, J. 1998. The biology of calanoid copepods. Academic Press, San Diego. North, EW and ED Houde. 2001. Retention of white perch and striped bass larvae: Biological-physical interactions in the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):756-789. Omori, M. and W. M. Hamner. 1982. Patchy distribution of zooplankton: Behavior, population assessment and sampling problems. Marine Biology. 72(2):193- 200. 94 Sabatini, M., and T. Kiorboe. 1994. Egg-production, growth and development of the cyclopoid copepod Oithona-similis. Journal of Plankton Research 16:1329- 1351. Roman, M. R., D. V. Holliday, and L. P. Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. Runge, J. A. and J. C. Roff. 2000. The measurement of growth and reproductive rates. Pages 401-444 in R. P. Harris, P.H. Wiebe, J. Lenz, H.R. Skjokdal, and M. Huntley, [eds.], ICES Zooplankton Methodology Manual. Academic Press, San Diego. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10:317-340. White, J. R., and M. R. Roman. 1992. Egg production by the calanoid copepod Acartia tonsa in the mesohaline Chesapeake Bay: The importance of food resources and temperature. Marine Ecology Progress Series. 86:239-249. Williamson, C. E., and N. M. Butler. 1987. Temperature, food and mate limitation of copepod reproductive rates: Separating the effects of multiple hypotheses. Journal of Plankton Research 9:821-836. 95 CHAPTER IV: Egg production and grazing by the copepod Eurytemora affinis in the Chesapeake Bay estuarine turbidity maximum ABSTRACT The calanoid copepod Eurytemora affinis tends to exhibit maximum abundance in the estuarine turbidity maxima (ETMs) of Chesapeake Bayand in estuaries around the world. Autotrophic production is generally low in these turbid regions due to light limitation, yet ETMs usually contain high abundances of zooplankton. In this study, egg clutch size and grazing measurements of E. affinis were conducted to help understand why this copepod species reaches maximum abundance within ETM regions. The comparison of clutch size and numerous biotic and abiotic variables revealed that E. affinis egg production does not appear to be food limited in the Chesapeake Bay or Choptank River ETMs because prosome length and not any measure of food quantity or quality was the best predictor for egg clutch size. Regression tree analysis using residuals from the prosome versus clutch size regression suggested that salinity had a slight positive influence on clutch size in 2002 (Chesapeake Bay; a dry year), while food quality may have had a greater influence on clutch size in 2003 (Choptank River; a wet year). Grazing measurements indicated that E. affinis ingests autotrophs (<20 m equivalent spherical diameter) at higher rates inside the ETM than outside, but grazes on >20 ESD protists at higher rates outside the ETM. Neither of these sources, however, sufficiently accounted for total grazing by E. affinis. The copepod, E. affinis, likely obtains a majority of its ingested carbon from additional sources which were not quantified, such as free-living nanoplankton and detrital communities. Benefits of 96 living in the ETM for E. affinis likely include decreased interspecific interactions (competitive and predator-prey) and abundant food resources (due to E. affinis? particle selection ability). INTRODUCTION The estuarine turbidity maximum (ETM) regions of the main stem of Chesapeake Bay and the Choptank River (Fig. 1), located near the limit of saltwater intrusion, are traditionally characterized by high total suspended solid (TSS) concentrations, high light attenuation, and high densities of zooplankton and anadromous fish larvae (North and Houde 1999, Roman et al. 2001). Phytoplankton productivity is generally low in ETM regions due to light limitation (Wofsy 1983, Kemp and Boynton 1984, Fisher et al. 1988). Therefore, food webs in ETMs tend to be based on allochthonous detrital inputs, with important microbial components (Irigoien and Castel 1997, Crump et al. 1998, North and Houde 2001). Abundances of the epibenthic copepod Eurytemora affinis peak in the Chesapeake Bay ETM region and in ETMs worldwide (Fig. 2; Heinle and Flemer 1975, Hough and Naylor 1992, Gasparini et al. 1999, Kimmel and Roman 2004). Eurytemora affinis is omnivorous, feeding on phytoplankton (Vijverberg 1980, Gasparini and Castel 1997), detritus (Heinle et al. 1977), protozoa (Berk et al. 1978, Merrell and Stoecker 1998), and bacteria (Boak and Goulder 1983). Assuming that E. affinis is retained within the ETM by either passive or active means, greater egg production rates (EPr) within the ETM versus up- and downstream of the ETM will increase population size. Higher EPr within the ETM 97 would suggest that the ETM is a suitable nutritional environment for copepods. Previous research has demonstrated that copepod egg production rates respond positively to increases in food quality. For example, in the Gironde estuary, E. affinis EPr increases with a higher nutritional quotient (NQ; carbohydrates+proteins+lipids:TSS) and with the ratio of chl-a to TSS (Burdloff et al. 2000). Gasparini et al. (1999) demonstrated that high TSS values in the Gironde estuary decreased ingestion of chl-a by E. affinis and suggested that this likely has an adverse effect on its EPr. In Acartia spp., egg production has been negatively correlated to seston C:N (Ambler 1986) and positively correlated with the protein and specific fatty acid concentrations of seston (J?nasd?ttir 1994, J?nasd?ttir et al. 1995), microzooplankton abundance (White and Roman 1992), and chl-a:TSS values (Burdloff et al. 2002). This study compared E. affinis egg clutch sizes throughout the Chesapeake Bay (2002) and Choptank River (2003) ETM regions with biotic and abiotic variables to determine 1) if EPr are consistently higher in regions of high turbidity compared to up- or down-estuary of these areas, and 2) what factors, if any, might limit egg production. I also conducted grazing experiment to determine the diets of E. affinis in the ETM and down-estuary from the ETM. MATERIALS AND METHODS Egg production Field work. I conducted five cruises in 2002 and six in 2003 into the ETM regions of Chesapeake Bay (2002) and the Choptank River (2003), a tributary of 98 Chesapeake Bay (Fig. 1). I selected stations downstream, within, and upstream of the ETM in each location based on salinity and turbidity profiles measured by a Seabird CTD (Chesapeake Bay) or a YSI Salinometer with a weighted sensor (Choptank River). Water samples were collected at mid-depth at each station using either a Niskin bottle or a submersible pump for analysis of chlorophyll-a (chl-a), phaeophytin, total suspended solids (TSS), carbon (C), nitrogen (N), total carbohydrates (CH2O), total protein, and total lipids. Eurytemora affinis was collected by replicate plankton tows from the bottom to the surface with a 64-m mesh size plankton net, 0.5 m in diameter, equipped with a General Oceans flowmeter. Zooplankton samples were immediately fixed in a 5% formalin solution for later processing. Seston analysis. Water samples collected for seston analyses were first sieved through a 64-m sieve and the filtrate collected. The resulting maximum particle size in the filtrate (< 64 m) was expected to correspond to published values of the maximum particle size available for E. affinis grazing (Richman et al. 1977). Filtrate from each station was filtered onto either uncombusted (chl-a, phaeophytin, and TSS) Whatman GF/F filters or GF/F filters that had been combusted at 450?C for 4 hours (C, N, CH2O, protein, and lipids). Chl-a and phaeophytin values were determined fluorometrically using the methods of Parsons and Strickland (1963). I calculated TSS by drying a filter in a 60?C oven to a constant weight and subtracting the initial weight of the filter. Filters for organic C and N analysis were dried at 60?C and analyzed on an Exeter Analytical Inc. CE-440 Elemental Analyzer. Carbohydrates were determined spectrophotometrically using an Hitachi U-3110 spectrophotometer 99 following the protocol of Dubois et al. (1956). Proteins were determined using bicinchoninic acid protein assay reagent (Smith et al. 1985) following the addition of sodium deoxycholate, which aids in protein release from filters (Nguyen and Harvey, 1994). Protein concentrations were then estimated on a Hitachi U-3110 spectrophotometer using BSA as a standard at an absorbance of 560 nm. Lipids were determined gravimetrically by extraction in a 1 CH2CL2:1 MeOH (vol/vol) solution, immersing the capped extraction tube in an ice-cold sonicating bath for 10 min., then following the Bligh and Dyer (1959) protocol. Zooplankton processing. Mesozooplankton samples were subsampled using a plankton splitter and 2 to 3 subsamples counted using a dissecting microscope. All E. affinis eggs, females (gravid and nongravid), and males were counted from each subsample (usually totaling ~100 copepods/subsample), and all gravid female prosome lengths were measured with an eyepiece micrometer. Female E. affinis C was calculated based on an equation by Kankaala and Johansson (1986) relating prosome length to C content: 2.83Female C = 6.25 Pro? , where Pro is prosome length (in mm). Egg numbers were multiplied by 0.048 to give egg C (Heinle and Flemer 1975, Hirche 1992). Clutch sizes were converted to egg production using the method of Lloyd and Roman (see Chapter 1). Briefly, the mean sizes of complete clutches were calculated for each station and divided by the sum of hatching time (HT) and interclutch duration (ID). Both of these time components are temperature dependant, with 1.0436.8HT T -= ? , and 100 1.523161.62ID T -= ? . Statistical analyses. I used JMP v. 5.0.1a (SAS Institute, Inc.) statistical software to perform linear regression and regression trees analyses on the biotic and abiotic factors affecting E. affinis clutch size. Regression trees split data at terminal nodes (inflection points between the dependent and a predictor variable) based on a single predictor variable, creating two groups that are both homogenous as possible with a minimum sum of squares (De?ath and Fabricius 2000). The data can be split as many times as there are predictor variables, but, as the number of splits increases, the less likely it is that a relationship exits between the predictor and dependent variable. I chose to use regression tree analysis over multiple regression because the independent variables were often related, which violates an assumption of multiple regression analysis. Further, regression trees are a more exploratory method for determining relationships of multiple factors on a single regressor and are often easier to interpret due to their graphical nature (Dzeroski and Drumm 2003). Due to limited output, regression trees can be limiting in their ability to provide a predictive model. They also are not capable of incorporating strong linear relationships (De?ath and Fabricius 2000), so I used the residuals of a linear regression between clutch size (dependent variable) and female prosome length (independent variable) in the analyses. Food concentration and growth calculations. Using equations derived by Huntley and Boyd (1984), I calculated both maintenance food concentrations (Cm; copepod carbon assimilation equals respiratory demands) and the critical food 101 concentration (Cc), above which copepod growth is assumed not to be food limited. These values are based solely on carbon concentrations and therefore must be analyzed cautiously, given that 1) all particulate carbon in the seston may not be available for copepod consumption and/or assimilation (Huntley and Boyd 1984), and 2) copepod reproductive and somatic growth often depends on food quality and not just the gross amount of carbon (J?nasd?ttir et al. 1995, Burdloff et al. 2000). These calculations assume that carbon (food) requirements for copepods are based on dry body mass (W, in mg) and temperature, which is incorporated into several variables (b, k, m, and n), and that copepods have an assimilation efficiency (mass of carbon passing through the gut wall into the body / total mass of ingested carbon) of 0.7 (Conover 1966). The equation for maintenance carbon concentration is: ( )m n m kWC ab - = , where k is the respiratory coefficient, W is the individual dry weight in mg, m is the respiratory exponent, n is the clearance rate exponent, a is the assimilation efficiency (0.7), and b is the clearance rate coefficient. The derivations of these variables are explained in Huntley and Boyd (1984). The food concentration at which growth, and presumably reproduction (Hopcroft and Roff 1998), is not limiting is explained by the equation: ( 1) hour c n GC abW -= , 102 where the variables are as described above (Huntley and Boyd 1984). Ghour is a variable that represents the maximum hourly weight- or carbon-specific growth rate, assuming that food is replete. I also compared our measured egg production rates with the values obtained from the Huntley and Boyd (1984) equation for egg production under food-replete conditions, with temperature (T) as the only variable: (0.110 )0.0542 T dailyG e= . Both of these approaches (calculating carbon requirements and growth under food- replete conditions) allow us to examine egg production rates in terms of food limitation. Grazing experiments Experimental design. I performed the grazing experiment in August, 2003, following the grazing experiment protocol as described by Gifford (1993). Eurytemora affinis used in grazing experiments were collected from the ETM of the Choptank River (Fig. 1; 0.1 salinity, 21.8?C), by plankton tows with a 200 m plankton net. The contents of the cod end were gently poured into a carboy containing unfiltered water from the same site. Water from the surface mixed layer was collected from both the ETM station and a site several kilometers downstream from the ETM. The water from both of these sites was poured through a 200 m sieve to be used in the grazing experiments. Grazing experiments were performed by incubating E. affinis in 250 ml bottles containing 200-m-filtered seawater from either the ETM site or the non-ETM site. In cases where copepods were incubated in the water of a different site from 103 which they were collected, the copepods were acclimated at an intermediate salinity for approximately 1 hour. Active, adult females were hand picked with a Pasteur pipette using a dissecting microscope and placed into their respective experimental bottles, with each bottle containing a total of ten individuals. There were three replicate bottles for each treatment combination, giving a total of 12 bottles (3 bottles containing E. affinis in ETM water, 3 bottles containing E. affinis in non-ETM water, 3 bottles containing A. tonsa in ETM water, 3 bottles containing A. tonsa in non-ETM water). Additionally, there were four sets of control bottles which included no copepods (i.e. ETM t0 control, ETM t24 control, non-ETM t0 control, non-ETM t24 control) with 3 replicates each. The sealed bottles were transferred to an environmental chamber at 20?C and loaded onto a plankton wheel set for 2 RPM for 24 hours. The control t0 bottles were harvested after 1 h following the start of the experiment to allow for the dissipation of any microplankton killed or damaged prior to the beginning of the experiment, while all other bottles (control and treatment) were collected after 24 hours. The contents from the treatment bottles were gently poured separately through a 64 um sieve to recover the copepods. The contents of both sets of control bottles were treated in the same manner. Water samples from all control and experimental bottles were preserved in 10% (v/v) acid Lugol?s solution for later enumeration (Parsons et al. 1984). All preserved samples were refrigerated in the dark until analysis. Copepod gut fluorescence was measured according to Bamstedt et. al (2000) after the 24 h grazing period. The copepods were filtered onto Whatman GF/F filters which were then transferred into centrifuge tubes containing 5 ml of 90% acetone. 104 The tubes were kept dark and were placed in a freezer. After 24 h, the tubes were shaken, vortexed, and centrifuged. Fluorescence was measured with a Turner fluorometer and converted to chlorophyll-a concentration using a chlorophyll-a versus fluorescence regression. Microscopic enumeration. The water samples preserved in Lugol?s solution were concentrated by settling 10 ml subsamples in Uterm?hl chambers. All samples were enumerated with an inverted microscope at 200x magnification. Microplankton between 10 mm and 200 mm were counted for subsamples from all bottles, including the t0 controls and t24 controls. Pico- and nanoplankton were not evaluated. Cells were grouped according to size (10-20 mm or >20 mm) into the following prey categories: Aloricate ciliates, dinoflagellates, diatoms, loricate ciliates, or other autotrophs (e.g. cryptophytes, colonial algae, desmids). Calculations and statistics. Prey carbon content was calculated from cell volumes and converted to carbon using the equations of Montagnes et al. (1994) and Menden-Deuer and Lessard (2000). Carbon-specific gut fluorescence was measured following the protocol of Barquero et al. (1998), and chlorophyll-a concentrations were converted to carbon using a 25:1 carbon to chlorophyll-a ratio (Steele and Baird 1961, Steele and Baird 1962, Landry et al. 1995). Microscope cell counts of samples from the control bottles and bottles with copepods allowed carbon-specific prey ingestion rates (I, day-1) to be determined by using the equations of Frost (1972). One-way ANOVA tests and multiple range tests (Tukey method) using SAS 8e 2000 software were performed to detect any significant (p<0.05) differences between the treatment combinations for mean gut 105 content, carbon-specific ingestion rates, and percent abundance changes in the control bottles. Specific prey ingestion rates were analyzed with JMP 4 software using Student?s t-test comparisons for each species by water type and each water type by species. RESULTS Seston and microplankton In 2002 (Chesapeake Bay), POC and chl-a concentrations and the chl-a/TSS ratio were greatest during the early spring and decreased throughout the year (Table I). The nutritional quotient (NQ), however, was relatively stable during the sampling period and never fell below 0.02, twice as much as the 0.01 value determined to limit E. affinis egg production (Burdloff et al. 2000). The POC values in 2003 (Choptank River) showed a similar seasonal trend as in Chesapeake Bay, but the C:N ratio remained relatively constant over the year (Table I). The chl-a/TSS ratio in the Choptank River in 2003 followed the same trend as in Chesapeake Bay in 2002 (high in spring, generally decreasing throughout year). Microplankton measurements were performed for the 2003 sampling year (Choptank River; Table II). Ciliate biomass peaked in the late spring, with large tintinnids dominating. With the exception of June, ciliate biomass remained at relatively high levels through mid-autumn. A Prorocentrum minimum bloom was evident in April and May. Besides a number of unidentified dinoflagellates, the only other dinoflagellate species of particular interest (in terms of biomass) were Gyrodinium spp. in July and Heterocapsa spp. in early spring and mid-autumn. 106 Diatom biomass peaked in early spring and dominated the microplankton community. In general, microplankton biomass was dominated by autotrophs (diatoms, dinoflagellates, and other autotrophic flagellates) in early spring, while ciliate biomass increased in late spring/early summer and remained high throughout the remainder of the year. Egg production A linear regression between clutch size (CS) and female prosome length (PL) explained 85% and 88% of clutch size variability for 2002 (Chesapeake Bay) and 2003 (Choptank River), respectively (Fig. 3). The 2002 regression tree using the residuals of CS versus PL as the regressor variable suggests that salinity was the major factor affecting CS residuals in the Chesapeake Bay ETM region that year, with all residual CS values being split at a salinity of 1.75 (Fig. 4). Although two nutritional variables are used in the second-level split of the data, chl-a/TSS (a measure of food quality) and N concentration, the mean values allotted to these nodes decrease with increasing chl-a/TSS and N values, suggesting that these variables do not affect CS. Therefore, I did not split the data further. Although salinity is also the first split of the 2003 CS residuals, the node created at salinity < 0.3 contains about 4% of the data (Fig. 5). The second split is based on a C:N of 8.64, with higher C:N values having lower mean. The third level of splits is based on a calculated nutritional quotient (NQ), which is the sum of all particulate carbohydrates, proteins, and lipids (in g), divided by TSS (in mg). Both nodes formed by the split of the ?salinity  0.3? node show a positive relationship 107 between NQ and the CS residuals, as one would expect if there is a true effect of NQ on the residuals. A regression tree for all 2002-2003 data shows a salinity value of 1.75 as the major variable describing CS residuals (Fig. 6). The ?salinity< 1.75? node, is then split in terms of C:N, with C:N values > 8.2 having a lower CS residual mean then values < 8.2. The split of the ?salinity > 1.75? node suggests for stations with an NQ  0.037, residual CS values are lower than when NQ < 0.037. Similar to the second split of the 2002 regression, this is likely a function of working with such a small variance, as one would not expect CS residuals to decrease with an increasing NQ. Prosome length was significantly related to water temperature two weeks prior to the prosome measurement (Fig. 7). Weight-specific egg production rates for E. affinis generally increased with temperature in the Chesapeake Bay ETM region (2002 but showed no relationship with in-situ temperature in the Choptank River (2003;Fig 8). Hirche (1992) developed a curve representing E. affinis egg production rates at saturating food concentrations versus temperature. The Chesapeake Bay (2002) data tended to fall along this line in the spring and fall, but in the summer the model overpredicts egg production rates, suggesting food limitation. The 2003 data is more variable compared to 2002 and was egg production was also overpredicted by the model at warmer temperatures (Fig 9). Burdloff et al. (2000) determined that the NQ limiting copepod egg production is around 0.010. Our minimum NQ was 0.016, suggesting that food quality should not limit egg production rates in the Chesapeake Bay or Choptank River ETMs. 108 Particulate organic carbon concentrations and model comparison Particulate organic carbon (POC) concentrations were usually an order of magnitude greater than those concentrations calculated to be the threshold between food-limited and non-food-limited growth (Fig. 10). Eurytemora affinis growth, and presumably reproduction, should not be limited by food quantity in the Chesapeake Bay and Choptank Rive ETM regions. Measured reproductive rates compared to Huntley and Boyd (1984) model outputs assuming food-replete conditions suggests that the measured rates (egg production) are 2.5 times greater than modeled rates (Figure 11). When only the 2002 data is used, the model explains 82% of the variance but under-estimates actual values; with both 2002 and 2003 data used, the r2 is nearly cut in half to 0.43 (Figure 11). Grazing experiment Eurytemora affinis ingestion rates for the <20 m size fraction are lower in non-ETM water than in the ETM water (Table III). The only positive ingestion rate in this size class is for the autotroph group in ETM water, which includes colonial algae, desmids, cryptophytes, and other autotrophic flagellates, and this ingestion rate in ETM water is significantly greater than in the non-ETM (Student?s t-test, p < 0.05). This group is clearly an important food item for E. affinis, as the autotroph ingestion rates are greater than protozoa in either size class. In the >20 m protozoan group, E. affinis fed more in the less turbid, non-ETM water type. Specifically, aloricate ciliates appeared to be the targeted prey type, with E. affinis individuals feeding on 24.11 cells (4.48 ng C) day-1 in the non-ETM water versus a negative grazing rate in 109 the ETM water. The only group in ETM water experiencing ingestion rates greater than in the non-ETM water were loricate ciliates, with 0.13 cells (0.40 ng C) consumed copepod-1 day-1. The copepod gut fluorescence measurements following the grazing experiments suggest that the grazing experiments can strongly underestimate carbon intake (Fig. 12). Further, there was no significant difference in gut fluorescence between the ETM and non-ETM treatments. Total C-specific grazing rates calculated from microscope counts are 5 and 3 orders-of-magnitude lower than the grazing rates determined by gut fluorescence in the ETM and non-ETM treatments, respectively (Fig. 12, Table III). DISCUSSION The strong relationship between clutch size and prosome length suggests that E. affinis egg production in both the Chesapeake Bay (r2 = 0.85) and Choptank River (r2=0.88) ETM regions is not food limited. Regressions using the residuals of these analyses versus salinity and TSS show no relationship (data not shown), meaning that copepod size is not related to either of these variables. The dynamic nature of the Chesapeake Bay and Choptank River ETM regions suggest that copepods likely experience constantly changing salinity and TSS concentrations throughout their lifetimes. Their ability to thrive in such variable environments may preclude E. affinis locating itself in a specific salinity or TSS range. A salinity of 1.75 was the first node of the 2002 regression tree. Devreker et al. (2004), working with copepods collected from the Seine Estuary, found that first- 110 and second-stage E. affinis nauplii showed lowest survival in freshwater compared to other salinities up to 35. The point at which 50% of the nauplii in their experiments survived was < 12 hours in freshwater but close to 3 days at a salinity of 2.5. In a dynamic environment like the ETM, gravid females find themselves exposed to variable salinities throughout a reproductive cycle. If a female produces an egg clutch while in higher salinities and then is transported into freshwater, there should be no effect of low salinity on clutch size. Any extended time in low salinities may elicit a reduced reproductive output either through adaptation or increased energy consumption due to osmoregulatory energy demands. Eurytemora affinis? tolerances to low salinities may depend heavily on their natural environment, with freshwater populations reproductively outperforming estuarine populations in freshwater, even after salinity acclimation. Katona (1970), using specimens collected in brackish water, determined that E. affinis can produce viable eggs at salinities ranging from 1.5 to 20 and temperatures ranging from 2 to 23?C. Ban (1994), using E. affinis from a freshwater lake, measured some of the highest egg production rates for this species in the literature (see his Table III). Ishikawa et al. (1999) performed salinity tolerance experiments again using E. affinis from a freshwater lake and found that hatching times were not significantly different from 0 to 10 ? and slowly increased as salinity increased. Over salinities of 25, hatching success was near 0 %. Salinities of up to 15 appeared to result in the greatest hatching success and survival; this is also the salinity range that E. affinis tends to be found in Chesapeake Bay and its tributaries. The maximum salinity tolerances of these freshwater populations seem to be consistent with their estuarine 111 counterparts, but estuarine copepods salinity tolerances seem to decline rapidly below 2 (also see Roddie et al. 1984). Regression tree analysis of the Choptank River clutch size residuals suggests that food quality may have played a minor role in E. affinis clutch size in 2003. The nutritional quotient (NQ) provides the first major split in the residual data, with a value of 0.038. This value is much greater than the value of 0.01 suggested by Burdloff et al. (2000) that should limit E. affinis egg production rates, and the mean NQ for the < 0.038 group is 0.024. Given that less than 5% of the residual data is included in the < 0.038 group, NQ did not appear to significantly affect clutch size in the Choptank River ETM in 2003. The next split of the > 0.038 group is based on a CH2O concentration of 0.49 mg l-1, a measurement of food quantity. The < 0.49 mg l- 1 CH2O group is split by the a chl-a:TSS ratio of 7.8?10-5, a measure of food availability, with 70% of the data at or above the node value. The  0.49 mg l-1 CH2O group is split based on the concentration of spherical ciliates < 20 m in diameter, with 62% of data falling below the node of 2.03 ciliates ml-1 and 38% equal to or above the node. The inclusion of ciliates in the regression tree should not be unexpected, as E. affinis may actually selectively graze heterotrophic microzooplankton, especially ciliates, over chl-a (Merrell and Stoecker 1998). Compared to the 2002 regression tree analysis, the 2003 regression tree nodes are based primarily on food quantity or quality related variables. The high flow of 2003 likely increased available habitat for E. affinis in the upper Choptank River estuary compared to the low flow years (Kimmel and Roman 2004). Osmoregulatory energy demands are lower under such conditions (Kimmel and Bradley 2001) and are 112 unlikely to affect clutch sizes or egg production rates. This is supported by food limitation variables (in terms of quality and quality) providing the major splits in the 2003 regression tree analyses. In 2002, salinity described most of the differences in the clutch size residuals, with food-related variables being secondary. However, analyses of available food and food requirements suggest that food was not limiting either year. Because prosome length was shown to be the major factor affecting clutch size, the overall influence of salinity and food on clutch size is minimal for both years. Eurytemora affinis egg production does not appear to be food limited in Chesapeake Bay or the Choptank River ETM regions. Though food quality and quantity appear to play a larger role in the Choptank River in 2003 than in Chesapeake Bay in 2002, both regression trees were created using the residuals of linear regressions that explained a large majority of the variability surrounding clutch size. Therefore, only a small amount of added variability is explained by using this statistical approach. Salinity and NQ may explain a small amount of clutch size variability for 2002 and 2003, respectively, but prosome length is the major forcing on clutch size in this study and in others (Crawford and Daborn 1986, Hirche 1992, Ban 1994). It is difficult to determine the effects of temperature and prosome length on clutch size because of the strong relationship between prosome and temperature. Chow-Fraser and Maly (1991) determined that clutch size is temperature-related for Diaptomus minutus and D. oregonensis, while Williamson and Butler (1987) found no significant effect of temperature on D. pallidus clutch size. 113 Several issues exist which may confound conclusions drawn from the clutch size analysis. There may be differences between the seston measured and that which E. affinis actually consumes. Great care has been taken to 1) sample water from the location (including depth) at which E. affinis was sampled and 2) to include only particles in the appropriate size range. Other studies have taken a similar approach and have found significant results between the nutritional environment and copepod egg production (J?nassd?ttir et al. 1995, Burdloff et al. 2000). J?nassd?ttir et al. (1995) found that specific unsaturated fatty acids were predictors in describing A. hudsonica and Temora longicornis egg production rates, and this type of analysis might be necessary in conditions where C is plentiful, and micronutrients may play an important role in egg production. However, because egg production is based on copepod feeding days and sometimes weeks prior to clutch extrusion, temporal changes in seston composition could also compromise this sort of analysis. This is especially true in warmer months, when organic material turnover is expected to be greater. Female age has also been implicated in differences in clutch sizes and egg production rates (Runge and Plourde 1996). Ianora and Poulet (1993) found that egg production and hatching success decrease for A. hudsonica, and it is reasonable to assume that other estuarine copepods, like E. affinis, might experience the same reduction in reproductive output with age. The Huntley and Boyd (1984) model for growth and carbon requirements also suggest that copepods are not food limited in the Chesapeake Bay and Choptank River ETMs. During the years sampled, measured POC values were much higher than those required by E. affinis based on weight and temperature by the Huntley and 114 Boyd (1984) model. In terms of the food quantity available, these copepods should not have limited growth or reproductive output. The growth model fit the 2002 egg production data, but the 2003 data again proved to be more complicated. When both years? egg production rates were compared to the growth model outputs, the r2 dropped from 0.82 (for 2002 data only) to 0.43 (2002 and 2003 data together). Berggreen et al. (1988) suggested that under conditions where copepods were food satiated, egg production rates should be reliable proxies for somatic growth, assuming that adult females allocate carbon that would go towards somatic growth to reproduction. Coastal and estuarine environments generally contain enough food resources so that copepods are rarely, if ever, food limited (McLaren 1978, Huntley and Boyd 1984). Assuming a C-specific ingestion rate of 1.25 day-1 (= 125% body C day-1) from Figure 12 and a gross growth efficiency of 0.3 (Mauchline 1998), total C available to growth (or reproduction) is ~42% of body C day-1 (0.42 day-1). This value is well within the bounds shown for egg production values in Figure 11, further suggesting that E. affinis egg production is not food limited in the Chesapeake Bay or Choptank River ETM regions. The estimated ingestion rates used in this calculation are within the limits for all literature values of E. affinis ingestion (Mauchline 1998). Eurytemora affinis abundance often increases during years of high freshwater input (Fig. 13), but such an increase does not imply a change in food availability. Gaston and Lawton (1990) reported that population abundance generally increases with increasing habitat size. Christie et al. (2003) found that macrofaunal abundances in a kelp bed were significantly related to habitat availability and not influenced by food resources. In Chesapeake Bay, increases in freshwater input via the 115 Susquehanna River tend to lower salinities in the oligo- and mesohaline portions of the bay, thus extending the habitat available to E. affinis further down-estuary (Kimmel and Roman 2004). Increased habitat volume could account for the increase in E. affinis abundance during high-flow years. The several approaches used to measure copepod egg production in reference to available food suggest that E. affinis egg production (and juvenile somatic growth) is not food limited. Therefore, one can assume that the copepods are feeding enough to meet their respiratory and growth needs. Our grazing experiments suggest that E. affinis grazes primarily on organisms <20 m in size. Thus, E. affinis may get its carbon from autotrophic flagellates, colonial algae, and desmids in mostly the <20 m size range, with ciliates >20 m also supplementing their diet. Given the low grazing rates for these organisms, it appears likely that E. affinis in the field is obtaining carbon and other nutritional requirements from other sources. It is also possible that there were bottle effects during the grazing experiments that reduced ingestion rates below in situ values. Eurytemora affinis is known to be an omnivore, feeding on detritus, bacteria, and both autotrophic and heterotrophic Protozoa. It is possible that E. affinis gains most of its carbon from sources not measured in the experiments, namely detritus, particle-attached bacteria and protists, and 5-10 m nanoflagellates. Heinle et al. (1977) found that E. affinis consumed detritus that had been sterilized of all microbiota (protozoa and bacteria) and that which contained a microbiotic community. Survival and egg production was much higher in the non-sterilized treatment; the sterilized treatment exhibited no egg production. Copepod grazing 116 may be high on particle attached bacteria and protozoans (Crump et al. 1998), but these organisms were not counted in our grazing experiments. The grazing experiments results suggest that E. affinis can obtain enough food outside of the ETM to survive and possibly thrive. Why E. affinis remains primarily in the ETM region may have to do with physical tolerances; the elevated turbidity providing a refuge from visual predators; or possibly interactions with other zooplankton species. Mysids (Winkler and Greve 2004) and fish larvae (North and Houde 2001), both abundant in the ETM are known predators of E. affinis .The copepod Acartia tonsa, usually peaks in abundance seaward of the ETM, and thus may limit E. affinis distributions through competition and predation (see Lonsdale 1979). Eurytemora affinis may be trapped like a particle in the ETM due to convergent flow patterns (Castel and Viega 1990; Roman et al. 2001) or it may actively maintain itself in the ETM area through tidally-based vertical migrations (Hough and Naylor 1991, Simenstad et al. 1994, Morgan et al. 1997). I do not know if E. affinis is able to exploit specific food resources of the ETM better than other zooplankton species or if other organisms prevent E. affinis from establishing a population outside of the ETM region. Further studies on these biological and physical interactions should help elicit why E. affinis dominates ETM regions. 11 7 TA BL ES Year Month C (mg/l) N (mg/l) Molar C:N Chl-a (g/l) Phaeo (g/l) TSS (mg/l) Temp (?C) Sal (PSU) CH 2 O (mg/l) Prot (mg/l) Lipid (mg/l) NQ (mg/mg) Chl-a:TSS (mg/mg) 2002 March 4.17 0.65 7.37 51.79 46.77 39.02 7.62 8.39 0.97 0.04 0.44 0.045 1.99E-03 April 3.94 0.46 9.76 31.42 125.93 113.04 11.41 4.01 1.22 0.27 0.60 0.020 4.42E-04 May 2.05 0.22 10.62 4.87 23.63 42.33 15.51 7.61 0.40 1.28 0.17 0.052 1.83E-04 June 2.86 0.29 11.47 2.69 23.76 48.17 24.36 0.11 0.76 0.37 0.30 0.030 5.75E-05 July 0.94 0.17 8.58 2.51 17.45 12.27 26.08 3.46 0.19 0.16 0.15 0.042 2.08E-04 2003 March 4.44 0.70 7.88 34.12 92.92 22.38 12.65 4.18 0.73 0.07 0.23 0.044 1.52E-03 May 2.66 0.34 9.24 4.59 13.51 28.33 17.00 3.10 0.43 0.87 0.17 0.052 1.61E-04 June 1.85 0.25 8.75 2.33 6.38 35.09 22.45 0.90 0.41 1.14 0.32 0.053 6.53E-05 July 1.46 0.21 8.06 3.43 4.59 23.62 27.20 3.40 0.36 0.67 0.18 0.051 1.46E-04 Sept 1.56 0.23 7.97 10.81 52.14 23.54 23.60 0.80 0.49 0.65 0.18 0.056 4.59E-04 Oct 1.20 0.16 8.64 3.37 15.06 15.65 14.20 2.40 0.23 0.50 0.11 0.054 2.15E-04 Table 4.1. Mean monthly values for nutritional and abiotic variables (minus microplankton) measured in the Chesapeake Bay (2002) and Choptank River (2003) ETM regions. C = carbon; N = nitrogen; Phaeo = phaeophytin; TSS = total suspended solids; CH 2 O = carbohydrates; Prot = proteins; NQ = nutritional quotient (sum of CH 2 O, Prot, and Lipids divided by TSS). 11 8 Month Tin <20 m Tin >20 m Con <20 Con >20 Sph <20 Sph >20 Hetero P. min Karlo Oxy Gyro Gymn Pperi UnID Dinos Diat Myr/ Meso AFlag Total Carbon March 0.00 0.00 0.00 0.00 0.00 25.02 19.51 0.00 0.00 0.00 0.00 0.00 0.00 49.29 221.99 6.62 31.94 354.37 April 0.00 174301.71 3062.48 1.79 1.79 1.79 0.00 3957.65 0.00 0.00 0.00 0.00 0.00 0.00 14.40 20.52 7.10 181381.60 May 1645.22 176445.77 628.55 0.00 0.27 0.27 0.00 30.99 4.50 0.00 0.00 0.00 0.00 71.40 26.50 1.61 14.81 178869.90 June 0.00 0.00 1257.09 0.67 0.27 4.02 0.00 1.33 0.00 0.28 0.00 0.00 0.00 52.70 29.47 1.95 6.88 1354.68 July 4113.10 89448.09 785.68 5.18 2.02 3.52 0.00 2.11 0.00 0.00 102.95 4.13 1.58 69.92 25.43 2.87 23.46 94590.05 Sept 3290.45 83484.91 1257.09 0.27 2.69 5.76 4.43 3.44 0.00 0.00 0.28 0.00 0.28 43.99 31.56 2.94 58.99 88194.04 Oct 8226.21 71558.48 628.55 0.54 1.07 1.22 196.03 0.00 0.00 0.00 2.39 0.00 0.00 261.96 67.94 0.31 21.10 80965.80 Table 4.2. Mean monthly microplankton biomass (pg C ml -1 ) as measured in the Choptank River in 2003. Tin = tintinnids; Con = conical ciliates; Sph = spherical ciliates; Hetero = Heterocapsa spp.; P. min = Prorocentrum minimum; Karlo = Karlodinium spp.; Oxy = Oxyrrhis spp.; Gyro = Gyrodinium spp.; Gymno = Gymnodinium spp.; Pperi = Protoperidinium spp.; UnID Dino = unidentified dinoflagellates; Diat = diatoms; Myr/Meso = Myrionecta spp. or Mesodinium spp.; AFlag = autotrophic flagellates (includes cryptophytes). 119 10?20 m Loricate cil 0 -2.51?10-3 (3.95?10-5) Aloricate cil -9.63?10-5 (5.77?10-4) -7.58?10-4 (7.89?10-4) Unidentified cil -1.24?10-5 (9.53?10-5) -2.02?10-4 (1.75?10-4) Diatoms -1.96?10-5 (2.06?10-5) 0 DF 0 -1.41?105 (1.64?10-6) Autoflag 3.78?10-5 (1.23?10-4) -5.16?10-4 (2.63?10-4) Total C ingested 3.78?10-5 0 >20 m Loricate cil 1.94?10-4 (6.08?10-4) -2.22?10-3 (5.28?10-3) Aloricate cil -2.20?10-4 (8.0?10-4) 2.16?10-2 (1.72?10-3) Unidentified cil 2.04?10-5 (1.71?10-5) 5.11?10-4 (5.12?10-4) Diatoms -7.33?10-5 (5.58?10-5) 5.19?10-5 (2.66?10-4) DF 0 2.01?10-4 (5.95?10-4) Autoflag 0 0 Total C ingested 2.14?10-4 2.23?10-2 Total C ingested for both size classes 2.52?10-4 2.23?10-2 Table 4.3. Mean carbon-specific ingestion rates, I, [g C ingested (g copepod body C)-1] for E. affinis in the Estuarine Turbidity Maximum (ETM) and downstream from the ETM for each prey type, listed by size class: 10-20 mm, >20 mm; N=3. Autoflag = Autotrophic flagellates. Standard deviations are in parentheses. Total C ingested shows the sum of positive ingestion rates only. Size Class Prey Category In ETM Downstream of ETM 120 FIGURES Figure 4.1. A map of the USA east coast, with Chesapeake Bay and its tributaries enlarged. The estuarine turbidity maximum (ETM) region of the Choptank River are shown using filled stars. Chester River Sassafras River Susquehanna River Patapsco River 76 76.5 76 76.5 39 39.5 L a t i t u d e ( ? N ) Longitude (?W) Choptank River ETM 10 20 30 kilometers 121 Figure 4.2. May, 2001 contour maps of turbidity (in nephalometric turbidity units, NTU) and E. affinis abundance (# m-3) in northern Chesapeake Bay. Note the relationship between turbidity (NTU) and copepod abundance. River km is measured as the distance downstream from Havre de Grace. 10 20 30 River Km 20 10 0 D e p t h ( m ) 10 20 30 River Km 20 10 0 D e p t h ( m ) 0 3000 6000 9000 12000 15000 0 8 16 24 32 40 48 NTU E. affinis 122 Figure 4.3. Linear regressions of clutch size versus female prosome length for 2002 (Chesapeake Bay) and 2003 (Choptank River). Only completely formed clutches were used in this analysis. C l u t c h s i z e ( e g g s f e m a l e - 1 ) Prosome length (m) r2 = 0.88 r2 = 0.85 0 20 40 60 80 100 120 140 160 180 500 600 700 800 900 1000 1100 1200 2002 2003 r2 = 0.85, Clutch Size = 0.23(Prosome) ? 123.65 r2 , Clutch Size = 0.23(Prosome) ? 130.65 123 Figure 4.4. Regression tree analysis using the residuals from the 2002 (Chesapeake Bay) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See Table I caption for abbreviation information. -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 0 40 80 120 160 Salinity < 1.75 Salinity  1.75 Chl/TSS < 1.28?10-4 Chl/TSS  1.28?10-4 Chl/TSS  8.41?10-4 Chl/TSS  8.41?10-4 Means of Residuals C o u n t All data First split Second split 124 Figure 4.5. Regression tree analysis using the residuals from the 2003 (Choptank River) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See the captions for Tables I and II for abbreviation information. C o u n t Means of Residuals NQ < 3.8?10-3 NQ  3.8?10-3 CH2O < 0.49 CH2O  0.49 Chl/TSS < 7.8?10-5 Sph. cil. < 2.03 Sph. cil.  2.03 -22 -6 -4 -2 0 2 4 6 8 10 12 14 0 20 40 60 80 100 All data First split Second split Third split 125 Figure 4.6. Regression tree analysis using the residuals from the 2002 (Chesapeake Bay) and 2003 (Choptank River) clutch size versus prosome length linear regression against the environmental variables measured (see Table I for a full list). ?Count? refers to the number of data points falling at that mean value. See the captions for Tables I and II for abbreviation information. -11-10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 50 100 150 200 250 Salinity < 1.75 Salinity  1.75 C:N  8.2 C:N < 8.2 NQ  0.04 NQ < 0.04 Temp < 22?C Temp  22?C C o u n t Means of Residuals All data First split Second split Third split 126 Figure 4.7. Linear regressions of female prosome length and estimated mean temperature during development for 2002 (Chesapeake Bay) and 2003 (Choptank River). P r o s o m e l e n g t h (  m ) r2 = 0.71 r2 = 0.77 500 600 700 800 900 1000 1100 1200 5 10 15 20 25 30 2002 2003 Temperature (?C) , Prosome = -15.29(Temp) + 1012.3 , Prosome = -23.09(Temp) + 1148.9 127 Figure 4.8. Carbon-specific egg production rates plotted versus temperature. Linear regression is for 2002 (Chesapeake Bay) data and is significant at p < 0.01. C - o r W t . - s p e c i f i c g r o w t h ( d - 1 ) Temperature (?C) r2 = 0.60 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 10 20 30 2002 2003 128 Figure 4.9. Comparisons of C-specific egg production rates from this and other studies to the Hirche (1992) model of E. affinis egg production under food replete conditions. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 2.5 7.5 12.5 17.5 22.5 27.5 Hirche 1992 Irigoien et al. 1993 Burdloff et al. 2000 Present Study 2002 Present Study 2003 Temperature (?C) C - o r W t . - s p e c i f i c g r o w t h ( d - 1 ) 129 0.01 0.1 1 10 7 12 17 22 27 2002 2003 Figure 4.10. Calculations for particulate carbon concentrations (POC) concentrations required for copepod maintenance (Cm; copepod carbon assimilation = respiratory demands, stippled line) and non-food limited copepod growth (Cc, solid line). Actual POC concentrations for 2002 and 2003 are also plotted. Temperature (?C) P O C ( m g l - 1 ) Cm Cc 130 Figure 4.11. Eurytemora affinis egg production rates versus modeled rates from Huntley and Boyd (1984). The modeled rates are based on temperature and copepod weight and assume no food limitation. The r2 for the linear regression is 0.43 (p < 0.05). 0 0.2 0.4 0.6 0.8 1 1.2 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 2002 2003 Model growth values (d-1) E g g p r o d u c t i o n r a t e s ( d - 1 ) 131 Figure 4.12. Results of gut fluorescence measurements following grazing experiment. Chlorophyll-a has been converted to carbon units in order to C-specific ingestion on autotrophic prey. Eury chl-spec Ing, Expt 1 0.00 0.50 1.00 1.50 ETM non ETM In ETM Downstream of ETM I [  g C (  g b o d y C ) - 1 d a y - 1 ] 132 Figure 4.13. Means of E. affinis abundance at Chesapeake Bay Program sampling stations from 1986-1995. 1990 and 1993 are considered years of enhanced Susquehanna River freshwater input (?wet? years). 1986, 1987, 1988, 1989, and 1992 are considered ?normal? and ?dry? years. The years following wet years (1991, 1994, and 1995) were not included in this analysis because E. affinis abundance those years was highly influenced by resting eggs laid the previous year (Kimmel and Roman 2004). 0 3000 6000 9000 12000 15000 0 60 120 180 240 Normal/Dry Years Wet Years Km down-estuary from Susquehanna River mouth E . a f f i n i s a b u n d a n c e ( i n d i v i d u a l s m - 3 ) 133 LITERATURE CITED Ambler, J. W. 1986. Effect of food quantity and quality on egg production of Acartia tonsa Dana from East Lagoon, Galveston, Texas. Estuarine, Coastal and Shelf Science 23(2):183-196. Ban, S. H. 1994. Effect of Temperature and Food Concentration On Postembryonic Development, Egg-Production and Adult Body-Size of Calanoid Copepod Eurytemora-Affinis. Journal of Plankton Research 16(6):721-735. Barquero, S., J. A. Cabal, R. Anadon, E. Fernandez, M. Varela and A. Bode. 1998. Ingestion rates of phytoplankton by copepod size fractions on a bloom ssociated with an off-shelf front off NW Spain. Journal of Plankton Research 20(5):957-972. Berggreen, U., B. Hansen and T. Kioerboe. 1988. Food size spectra, ingestion and rowth of the copepod Acartia tonsa during development: Implications for determination of copepod production. Marine Biology 99(3):341-352. Berk, S. G., D. C. Brownlee, D. R. Heinle, H. J. Kling and R. R. Colwell. 1978. Ciliates as a food source for marine planktonic copepods. Microbial Ecology 4(1):27-40. Bligh, E. G. and W. J. Dyer. 1959. A Rapid Method of Total Lipid Extraction and Purification. Canadian Journal of Biochemistry and Physiology 37(8):911- 917. Boak, A. and R. Goulder. 1983. Bacterioplankton in the diet of the calanoid copepod Eurytemora sp. in the Humber Estuary. Marine Biology 73:139-149. Burdloff, D., S. Gasparini, B. Sautour, H. Etcheber and J. Castel. 2000. Is the copepod egg production in a highly turbid estuary (the Gironde, France) a function of the biochemical composition of seston? Aquatic Ecology 34(2):165-175. Burdloff, D., S. Gasparini, F. Villate, I. Uriarte, U. Cotano, B. Sautour and H. Etcheber. 2002. Egg production of the copepod Acartia bifilosa in two contrasting European estuaries in relation to seston composition. Journal of Experimental Marine Biology and Ecology 274(1):1-17. Chow-Fraser, P. and E. J. Maly. 1991. Factors governing clutch size in two species of Diaptomus (Copepoda: Calanoida). Canadian Journal of Fisheries and Aquatic Sciences 48(3):364-370. Christie, H., N. M. Jorgensen, K. M. Norderhaug and E. Waage-Nielsen. 2003. Species distribution and habitat exploitation of fauna associated with kelp 134 (Laminaria hyperborea) along the Norwegian coast. Journal of the Marine Biological Association of the United Kingdom 83(4):687-699. Conover, R. J. 1966. Assimilation of organic matter zooplankton. Limnology and Oceanography 11(3):338-345. Crawford, P. and G. R. Daborn. 1986. Seasonal variations in body size and fecundity in a copepod of turbid estuaries. Estuaries 9(2):133-141. Crump, B. C., J. A. Baross and C. A. Simenstad. 1998. Dominance of particle- attached bacteria in the Columbia River estuary, USA. Aquatic Microbial Ecology 14(1):7-18. De'ath, G. and K. E. Fabricius. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology 81(11):3178-3192. Devreker, D., S. Souissi and L. Seuront. 2004. Development and mortality of the first naupliar stages of Eurytemora affinis (Copepoda, Calanoida) under different conditions of salinity and temperature. Journal of Experimental Marine Biology and Ecology 303(1):31-46. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith. 1956. Colorimetric Method for Determination of Sugars and Related Substances. Analytical Chemistry 28(3):350-356. Dzeroski, S. and D. Drumm. 2003. Using regression trees to identify the habitat preference of the sea cucumber (Holothuria leucospilota) on Rarotonga, Cook Islands. Ecological Modelling 170(2-3):219-226. Fisher, T. R., L. W. Harding, Jr., D. W. Stanley and L. G. Ward. 1988. Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson estuaries. Estuarine, Coastal and Shelf Science 27(1):61-93. Frost, B. W. 1972. Effects of size and concentration of food particles on the feeding behavior of the marine planktonic copepod Calanus pacificus. Limnology and Oceanography 17(6):805-815. Gasparini, S. and J. Castel. 1997. Autotrophic and heterotrophic nanoplankton in the diet of the estuarine copepods Eurytemora affinis and Acartia bifilosa. Journal of Plankton Research 19(7):877-890. Gasparini, S., J. Castel and X. Irigoien. 1999. Impact of suspended particulate matter on egg production of the estuarine copepod, Eurytemora affinis. Journal of Marine Systems 22(2-3):195-205. 135 Gaston, K. J. and J. H. Lawton. 1990. Effects of scale and habitat on the relationship between regional distribution and local abundance. Oikos 58(3):329-335. Gifford, D. J. 1993. Planktonic protozoa in copepod diet: Examples from the North Atlantic and the North Pacific. ICES 1(1): Heinle, D. R. and D. A. Flemer. 1975. Carbon requirements of a population of the estuarine copepod Eurytemora affinis. Marine Biology 31(3):235-247. Heinle, D. R., R.P. Harris, J.F. Ustach, and D.A. Flemer. 1977. Detritus as food for estuarine copepods. Marine Biology 40:341-353. Hirche, H.-J. 1992. Egg Production of Eurytemora affinis - effect of k-strategy. Estuarine, Coastal, and Shelf Science 35:395-407. Hopcroft, R. R. and J. C. Roff. 1998. Zooplankton growth rates: the influence of female size and resources on egg production of tropical marine copepods. Marine Biology 132:79-86. Hough, A. R. and E. Naylor. 1992. Endogenous Rhythms of Circatidal Swimming Activity in the Estuarine Copepod Eurytemora-Affinis (Poppe). Journal of Experimental Marine Biology and Ecology 161(1):27-32. Huntley, M. and C. Boyd. 1984. Food-limited growth of marine zooplankton. The American Naturalist 124(4):455-478. Ianora, A. and S. A. Poulet. 1615. Egg viability in the copepod Temora stylifera. Limnology and Oceanography 38(8):1615-1626. Irigoien, X. and J. Castel. 1997. Light limitation and distribution of chlorophyll pigments in a highly turbid estuary: The Gironde (SW France). Estuarine, Coastal and Shelf Science 44(4):507-517. Ishikawa, A., S. Ban and N. Shiga. 1999. Effects of salinity on survival and embryonic and postembryonic development of Eurytemora affinis from a freshwater lake. Plankton Biology and Ecology 46:113-119. Jonasdottir, S. H. 1994. Effects of food quality on the reproductive success of Acartia tonsa and Acartia hudsonica: Laboratory observations. Marine Biology 121(1):67-81. Jonasdottir, S. H., D. Fields, and S. Pantoja. 1995. Copepod egg production in Long Island Sound, USA, as a function of the chemical compostion of seston. Marine Ecology Progress Series 119:87-98. 136 Kankaala, P. and S. Johansson. 1986. The influence of individual variation on length- biomass regressions in three crustacean zooplankton species. Journal of Plankton Research 8(6):1027-1038. Katona, S. K. 1970. Growth characteristics of the copepods Eurytemora affinis and E. herdmani in laboratory cultures. Helgolander wiss. Meeresunters. 20:373-384. Kemp, W. M. and W. R. Boynton. 1984. Spatial and temporal coupling of nutrient inputs to estuarine primary production: The role of particulate transport and decomposition. Bulletin of Marine Science 35:522-535. Kimmel, D. G. and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Kimmel, D. G. and B. P. Bradley. 2001. Specific protein responses in the calanoid copepod Eurytemora affinis (Poppe, 1880) to salinity and temperature variation. Journal of Experimental Marine Biology and Ecology 266(2):135- 149. Landry, M. R., W. K. Peterson and C. J. Lorenzen. 1995. Zooplankton Grazing, Phytoplankton Growth, and Export Flux - Inferences from Chlorophyll Tracer Methods. Ices Journal of Marine Science 52(3-4):337-345. Lonsdale, D. J., D. R. Heinle and C. Siegfried. 1979. Carnivorous feeding behavior of the adult calanoid copepod Acartia tonsa Dana. J. Exp. Mar. Biol. Ecol. 36(3):235-248. Mauchline, J. 1998. The biology of calanoid copepods. Academic Press, San Diego. 402 pp. McLaren, I. A. 1978. Generation lengths of some temperate marine copepods: estimation, prediction, and implications. J. Fish. Res. Board Can. 35(10):1330-1342. Menden-Deuer, S. and E. J. Lessard. 2000. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnology and Oceanography 45(3):569-579. Merrell, J. R. and D. K. Stoecker. 1998. Differential grazing on protozoan microplankton by developmental stages of the calanoid copepod Eurytemora affinis Poppe. Journal of Plankton Research 20(2):289-304. Montagnes, D. J. S., J. A. Berges, P. J. Harrison and F. J. R. Taylor. 1994. Estimating carbon, nitrogen, protein, and chlorophyll a from volume in marine phytoplankton. Limnology and Oceanography 39(5):1044-1060. 137 North, E. W. and E. D. Houde. 2001. Retention of White Perch and Striped Bass Larvae: Biological-Physical Interactions in Chesapeake Bay Estuarine Turbidity Maximum. Estuaries 24(5):756-769. Nguyen, R. T. and H. R. Harvey. 1994. A rapid micro-scale method for the extraction and analysis of protein in marine samples. Marine Chemistry 45:1-2. Parsons, T. R. and J. D. H. Strickland. 1963. Discussion of Spectrophotometric Determination of Marine-Plant Pigments, with Revised Equations for Ascertaining Chlorophylls and Carotenoids. Journal of Marine Research 21(3):155-163. Parsons, T. R., Y. Maita and C. M. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford. 187 pp. Richman, S., D.R. Heinle, and R. Huff. 1977. Grazing by adult estuarine copepods of the Chesapeake Bay. Marine Biology 42:69-84. Roddie, B. D., R. J. G. Leakey and A. J. Berry. 1984. Salinity-temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth Estuary. Journal of Experimental Marine Biology and Ecology 77(2):191-211. Roman, M. R., D. V. Holliday and L. P. Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. Runge, J. A. and S. Plourde. 1996. Fecundity characteristics of Calanus finmarchicus in coastal waters of eastern Canada. Ophelia 44:1-3. Smith, P.K., et al. (1985). Measurement of protein using bicinchoninic acid. Analytical Biochemistry. 150:76-85. Steele, J.H. and I.E. Baird (1961) Relations between primary production, chlorophyll and particulate carbon. Limnology and Oceanography 6:68-78. Steele, J.H. and I.E. Baird (1962) Further relations between primary production, chlorophyll, and particulate carbon. Limnology and Oceanography 7:42-47. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10(4):317-340. White, J. R. and M. R. Roman. 1992. Egg production by the calanoid copepod Acartia tonsa in the mesohaline Chesapeake Bay: The importance of food resources and temperature. Marine Ecology Progress Series 86(3):239-249. 138 Williamson, C. E. and N. M. Butler. 1987. Temperature, food and mate limitation of copepod reproductive rates: Separating the effects of multiple hypotheses. Journal of Plankton Research 9(5):821-836. Winkler, G. and W. Greve. 2004. Trophodynamics of two interacting species of estuarine mysids, Praunus flexuosus and Neomysis integer, and their predation on the calanoid copepod Eurytemora affinis. Journal of Experimental Marine Biology and Ecology 308(1):127-146. Wofsy, S. C. 1983. A simple model to predict extinction coefficients and phytoplankton biomass in eutrophic waters. Limnology and Oceanography 28(6):1144-1155. 139 CHAPTER V: Tidally-timed vertical migration of zooplankton in the Chesapeake Bay estuarine turbidity maximum ABSTRACT Estuarine zooplankton often use some behavioral means of position maintenance, such as tidally-timed vertical migration, to maintain themselves in distinct salinity ranges. To determine if zooplankton, especially the copepod Eurytemora affinis, vertically migrate to maintain their position within the Chesapeake Bay estuarine turbidity maximum (ETM) region, I conducted time-series measurements of salinity, temperature, fluorescence, total suspended solids (TSS), current velocities and acoustically determined zooplankton biovolume (BV). Zooplankton BV was significantly related to TSS near the bottom and middle depths, but there was no relationship higher into the water column. Whereas TSS concentrations in the water column appeared to be primarily the result of tidal current scouring along the bottom, the lack of concordance between zooplankton BV and TSS higher in the water column suggests active vertical movement by zooplankton. It is possible that zooplankton in the Chesapeake Bay ETM vertically migrate in response to tides, but the data are ambiguous. The data do suggest that zooplankton in the up-estuary regions of the ETM might vertically migrate upward during ebb tides. In the down-estuary region, zooplankton are concentrated near the bottom during flood tides. Such movements could position zooplankton at depths with the greatest horizontal movement, creating a conveyor belt of zooplankton movement through the ETM region. Additionally, the Susquehanna River may provide a 140 freshwater source to sweep zooplankton into the ETM. Such a migration pattern would retain zooplankton in their optimal salinity range while possibly allowing them to be maintained in, or find food-rich regions. It is clear that, if zooplankton migrate, they do not migrate solely in response to tidal phase (based on BV and TSS versus current direction relationships), and that any migration patterns do not minimize horizontal displacement. INTRODUCTION Position maintenance within particular regions of an estuary can require behavioral mechanisms that reduce advection (Kimmerer et al. 1998; Forward and Tankersley 2001). Zooplankton vertical migration patterns are often attributed to the diel cycle of solar irradiance (e.g. Cushing 1951, Ohman 1990). During light-induced diel vertical migration, zooplankton migrate to the surface at night to feed and back to depth as sunrise approaches to avoid predation (i.e, Huntley and Brooks 1982, Frost 1988, Verheye and Field 1992). This phenomenon is especially prevalent in oceans and lakes, where potential advection from a particular habitat is less than it is in estuaries (Ohman 1990; Ringelberg 1995). However, due to changes in tidal current velocities and direction with depth and over time, vertical migrations cued to tidal cycles are common in the estuarine larvae of meroplanktonic organisms such as crabs (Cronin and Forward 1983, Tankersley and Forward 1994, Zeng et al. 1999), fish (Dauvin and Dodson 1990, Joyeux 1999), and molluscs (Roberts et al. 1989) and in certain holoplankton such as the calanoid copepod Eurytemora affinis (Hough and Naylor 1991). 141 The estuarine turbidity maximum (ETM) region of most partially-mixed estuaries is located near the landward limit of salt water intrusion and contains, by definition, a higher concentration of suspended solids than waters up- or down- estuary. Such high turbidity may reduce diel migration patterns by increasing light attenuation (Kimmerer et al. 1998; Roman et al. 2001). Eurytemora affinis is an epibenthic copepod whose abundance usually peaks in low-salinity ETM regions worldwide (Simenstad et al. 1990; Hough and Naylor 1992; Sautour and Castel 1995; Roman et al. 2001). In order for a population maximum of E. affinis to occur in an ETM region with a net flow out of that region (seaward) there must be 1) active movement by individuals to transport themselves into areas that will promote their retention, 2) physical processes that resuspend and settle organisms in a way that promotes retention, and/or 3) higher reproductive output within the region than up-or down-estuary (Kimmerer et al. 1998). Eurytemora affinis has been found to utilize tidally-timed vertical migration to maintain its position within the ETM in some estuaries (Hough and Naylor 1991, Simenstad et al. 1994, Morgan et al. 1997). In these studies, E. affinis was generally dominant higher in the water column during flood tide and deeper during ebb, thus being associated with the lowest velocity regions within the water column. Morgan et al. (1997) found that the principal lunar tidal component (M2) described nearly half of all variability in E. affinis abundance, further supporting the idea of a tidally-timed vertical migration. Other studies have demonstrated that E. affinis populations may be retained within estuaries by passive means. Castel and Viega (1990) determined that the 142 swimming abilities of E. affinis are ineffective in retaining the population within the turbid Gironde estuary. This finding and the positive relationship between E. affinis abundance and suspended matter concentration led them to conclude that the copepods utilize hydrodynamic processes to maintain themselves within the Gironde estuary; that is, they behave as passive particles. Kimmerer et al. (1998) found that macrozooplankton (mysids and amphipods) exhibited tidally-induced vertical migrations that maintained their positions within San Francisco Bay (higher in the water column during flood, lower during ebb), but the copepods E. affinis and Pseudodiaptomus forbesi did not. However, both copepod species were found higher in the water column than expected, probably due to the high turbidity of northern San Francisco Bay which would provide a predation refuge. Pseudodiaptomus forbesi, a brooding copepod like E. affinis, has been known to use setae on its antennules to attach itself to detrital particles, which 1) could reduce mortality by visual predators and 2) allow the copepod to utilize the particle trapping dynamics of the ETM to maintain its position within the ETM. Eurytemora affinis and other planktonic ETM fauna may adopt a similar strategy. Eurytemora affinis may also exercise plasticity in its migratory behaviors. Roman et al. (2001) used acoustical methods to measure zooplankton biovolume (BV) and concluded that, under conditions of high freshwater input, zooplankton in the Chesapeake Bay ETM are aggregated and resuspended by the same physical processes that affect sediments. They proposed a conceptual model in which zooplankton and sediments are resuspended into the water column landward of the salt front during an ebb tide. During flood conditions, zooplankton and sediments 143 below the salt front are also readily resuspended but are limited in their vertical suspension by the pycnocline. Resuspension in this model is based on the spatial location where maximum bottom currents scour the seafloor (i.e., above the salt front during ebb flow and below the salt front during a flood). In this study, I examined zooplankton biovolume in the Chesapeake Bay ETM using a larger dataset and longer time series under conditions of freshwater input that were much reduced compared to those observed by Roman et al. (2001). Our goals were to determine if zooplankton in the Chesapeake Bay ETM region (dominated by E. affinis) actively regulate their positions via tidally-timed vertical migrations and to compare the results with the Roman et al. (2001) model. METHODS Field Sampling I conducted two surveys in May, 2002, in the channel of the Chesapeake Bay ETM region (Fig. 1). Tucker trawl samples during this period show that E. affinis was the most abundant zooplankton species in the ETM region (Chapter 1). Our surveys consisted of sampling 3 to 7 fixed stations sequentially over a 22 to 27 hour time period following a hydrographic survey to locate the ETM. The first May survey (May 8-9) included thirteen axial (north to south) transects through the ETM region; the second survey (May 14-15) included eleven. At each station, a vertical profile was taken using a Sea Bird CTD rosette equipped with sensors to measure salinity, temperature, fluorescence, light transmission, and oxygen concentration. In addition, a shipboard Acoustic Doppler Current Profiler (ADCP) was used to measure 144 current velocities at 0.5 m vertical bins. All data was later averaged into 1 m vertical bins. Fluorescence measurements were converted to chlorophyll-a (chl-a; g l-1) concentration by collecting water samples, extracting chl-a in a 90% acetone solution, and then performing a regression between chl-a and fluorescence. Total suspended solid (TSS; mg l-1) concentration was determined by filtering water samples onto pre- weighed Whatman GF/F filters, drying to constant weight, and subtracting the initial filter weight. These TSS measurements were then regressed against transmissiometer readings, allowing all transmissiometer readings to be converted to TSS. Zooplankton biovolume (BV) was quantified using a Tracor Acoustic Profiling System (TAPS). This device is capable of taking fine-scale vertical (< 1 m) measurements of zooplankton BV concentration (mm3 m-3) by measuring acoustical backscattering of frequencies ranging from 300 kHz to 3 MHz. Several BV measurements are made each meter and are later averaged into 1 m bins. Details on TAPS theories, operation, and data analysis can be found in Holliday and Pieper (1995) and Roman et al. (2001). TAPS measurements of zooplankton biomass and abundance are highly correlated with traditional net and pump estimates (Fig. 2) (Barans et al. 1997; Pieper et al. 2001; Roman et al. 2001). The minimum linear size of detection is around 200 m, making TAPS ideal for measuring zooplankton ranging from nauplii to mysids while excluding most sediments. Even flocculated material that may exceed 200 m equivalent spherical diameter is not usually detected by acoustical methods, which instead detect flocs as their constituent grains and not as single particles (Fugate and Friedrichs 2002). 145 Data Analyses Visual exploratory analyses of temporal and spatial patterns were performed using the data contouring program Surfer (Golden Software, v. 8). Single-station time series were plotted for stations centered on river km (distance from Havre de Grace, Maryland) 30, where zooplankton counts from Tucker trawl surveys collected just prior to the present surveys indicated that E. affinis was the most abundant zooplankton species (Chapter 1). Contour plots of each transect of the 25-hour surveys were also created. Statistical analyses were chosen based on the assumption that zooplankton tidally-timed migration is indicated by 1) the depth of zooplankton BV maxima with respect to tidal direction and velocity and 2) relationships between BV and TSS. Because zooplankton may respond to current velocity (magnitude and direction) based on their location within the ETM region, BV and TSS values for stations up- estuary from river km 28 (located at approximately river km 26, 22, 18, and 16), which was near the center point of all transects, were averaged in 1 m depth bins for each transect. The same process was repeated for down-estuary stations (located at approximately river km 30, 34, and 38). The resulting dataset was composed of depth-specific, transect-averaged BV and TSS values for both landward (up-estuary) and seaward (down-estuary) regions. Depths of 0-4 m were grouped as ?surface? depths, 4-8 m were grouped as ?middle? depths, and 8-12 m were grouped as ?bottom? depths. Linear regressions between both BV and TSS versus current velocity (by depth) were then performed. 146 Pairwise correlation analyses were also used to determine relationships between BV and TSS in surface, middle, and bottom waters. RESULTS Zooplankton BV aggregations are centered in the upper and lower water column, with significantly less BV being measured at mid-depths (Table I). Surface BV is greater than the other depths and both surface and bottom BV values are significantly higher than middle BV measurements. Zooplankton BV is positively related to TSS concentrations in middle and bottom waters for both cruises (Table II). There are no significant relationships between the zooplankton BV depth maxima and time-of-day, indicating no diel vertical migration (data not shown). At river km 30, both BV and TSS become elevated in bottom waters during flood tides in the first survey, with TSS resuspension occurring after zooplankton resuspension (Fig. 3). Increases in zooplankton biomass correspond to a peak in flood currents. Peaks in surface ebb current flow match surface maxima in zooplankton biomass (Fig. 3). In contrast, TSS and BV are significantly greater during flood currents at the bottom and surface for the second cruise (Fig 4). Displacement calculations for the May and July cruises show that following complete tidal cycles, water in the upper water column is displaced down-estuary and water near the bottom was displaced up-estuary (Table III). The average displacement calculated over 3 to 4 complete tidal cycles was -3.38 km (negative denotes a down-estuary direction) in surface waters, -0.36 km in middle waters, and 147 +1.32 km in bottom waters. These displacements are consistent with the estuarine circulation expected at a location downstream of the salt front. The relationships between both zooplankton BV and TSS concentrations versus current direction are variable in contour plots of the multiple fixed-station surveys. In general, zooplankton BV during the first May survey tend to be concentrated near the surface in transects e through j (Fig. 5), which covered both flood and ebb tides (Fig. 3). In the other 5 surveys conducted during both flood and ebb, zooplankton are concentrated equally near the surface and the bottom. Most TSS resuspension occurred during flood tides and was isolated to bottom and middle depths. TSS maxima were also limited to the lower water column during the second May survey (Fig. 6). Zooplankton BV maxima do not appear to follow a trend in water column location versus tidal stage during this survey. In general, BV and TSS maxima are more intense during flood tides, especially along the bottom (Fig. 6, frames a, j, and k). It should be noted that contour plots of the second survey only extend from river km 22 to 40, while some transects extend to river km 18 and 16. Regression analyses of BV and TSS versus tidal velocity suggest that zooplankton BV location within the water column is related to current velocity in different ways with regard to location within the estuary (up-estuary or down-estuary) (Figs. 7 and 8). In the up-estuary region of the ETM, ebb currents (seaward) are related to higher BV concentrations in middle and surface depths. Bottom BV concentrations increase as both ebb and flood currents increase. Although some regressions show a positive, parabolic relationship between TSS and current velocity, TSS concentrations tend to increase at all depths as current velocity in the flood 148 direction increases (Figs. 7). Seaward of the ETM, BV increases as flood tide velocities increase in both bottom and middle depths. Concentrations of TSS increase with flood velocities at all depths (Fig. 8). There is no significant relationship between BV and current velocity in surface waters seaward of the ETM. DISCUSSION Freshwater input into Chesapeake Bay in 2001 was below the 50-year average (http://md.water.usgs.gov/monthly/bay.html) and has been shown to positively influence E. affinis location and abundance in Chesapeake Bay (Kimmel and Roman 2004) and other estuaries (Peitsch et al. 2000; Kimmerer 2002). Further, spring E. affinis abundance in northern Chesapeake Bay is affected by freshwater input the previous autumn, with greater Susquehanna River input in the autumn leading to a greater abundance of E. affinis the following spring (Kimmel and Roman 2004). Below-average freshwater input may explain the relatively low copepod abundances found during our ETM sampling program (Chapter 1) compared to other years (Roman et al. 2001; Kimmel and Roman 2004). Roman et al (2001) conducted their ETM field work in 1996, which was a year of exceptionally high Susquehanna River flow, and E. affinis was abundant throughout the year (Kimmel and Roman 2004). Freshwater input was less in 2002 than in 1996, producing a seasonal shift in zooplankton abundance in the ETM, with the copepods Acartia tonsa and E. affinis exhibiting similar abundances at down-estuary stations (maximum values of 200-300 ind l-1, determined by May Tucker trawl surveys). May acoustic surveys were analyzed for evidence of zooplankton vertical migration under the presumption that 149 data primarily represented E. affinis, which was the dominant zooplankton species at this time. In spite of this dominance by E. affinis, A. tonsa BV is possibly incorporated in the down-estuary measurements of total zooplankton BV. There is also the possibility that cladoceran BV is included in the BV measurements. North and Houde (2003), working in the Chesapeake Bay ETM in May 1998 and 1999, reported abundances of the cladoceran Bosmina longirostris peaking in the ETM region and up-estuary of the ETM. Lloyd (Chapter 1) also suggests that maximum abundances of cladocerans, Daphnia spp. and especially B. longirostris, occur up-estuary of the ETM, usually in freshwater. While E. affinis and B. longirostris distributions overlap to some extent in the Chesapeake Bay ETM, abundance maxima of E. affinis (adults and copepodites) are often an order-of- magnitude greater than B. longirostris density (Fig. 9; North and Houde 2003). Contour plots and calculations from a Tucker trawl survey conducted just days apart from both 25-hour surveys show that average and maximum cladoceran (B. longirostris and Daphnia spp.) densities in the ETM in May, 2002, are 10% and 32% of average and maximum E. affinis densities, respectively. Assuming 1) an average cladoceran size of 750 m (the largest individuals of the more abundant Bosmina longirostris are ~500 m (Zaret and Kerfoot 1975), and the largest Daphnia spp. individuals collected in the ETM are slightly > 1000 m), 2) the average size of an E. affinis individual is ~1000 m, and 3) a given length-to-BV ratio, the average and maximum calculated BV of cladocerans in the ETM are 8% and 14% of E. affinis BV, respectively. Therefore, while cladoceran BV is likely incorporated into 150 estimates of zooplankton BV, the cladoceran signal is likely small relative to copepod BV. The single-station contour plots ( Figs. 3 and 4) show a general trend of zooplankton BV and TSS resuspension in bottom waters during a flood tide that is either timed with, or precedes, a surface BV peak. Roman et al. (2001) noted that zooplankton BV resuspension tends to lead TSS as they both track with tidal currents. This same phenomenon appears to exist in the current study. The phasing between current speed and the resuspension of sediments and zooplankton may be due to the cohesive nature (i.e. stickiness) of sediments which must be disrupted by current energy before sediments can be eroded (Sanford et al. 2001). Thus, increasing tidal currents would first resuspend zooplankton, which are aggregated near the bottom, followed by sediment resuspension (Roman et al. 2001). The two time series centered on the station near river km 30 (Figs. 3 and 4) demonstrate noticeable differences in water column zooplankton BV and TSS concentrations relative to current direction. Zooplankton concentrations along the bottom and at the surface increase with flood tides, and TSS increases in the lower water column with flood tides. Ebb currents appear to create zooplankton surface maxima but generally lack enhanced near-bottom zooplankton or TSS values. Zooplankton biomass maxima along the bottom coincide with TSS peaks, but surface BV peaks do not correspond with surface turbidity peaks (see Figs. 3 and 4). If zooplankton were simply responding passively to the same forces acting on sediments, TSS and BV concentrations should overlap at all depths. Suspended sediment concentrations, however, rarely increase above 30 mg l-1 above a depth of 151 6m for both May surveys, which is near the background TSS concentration for Chesapeake Bay (Sanford et al. 2001). It appears that both zooplankton and TSS are resuspended by current-induced scouring along the bottom, with resuspension of zooplankton preceding the resuspension of sediments. However with decreased current speed, zooplankton remain suspended, perhaps by actively swimming, in the water column longer than sediments, which apparently settle. Sanford et al. (2001), working in the Chesapeake Bay ETM in 1996, showed that the average settling speed for the smallest, most abundant particles (in concentrations over 40 mg l-1) in the ETM was less than 0.02 mm s-1. The largest size class of particles (at concentrations less than 3 mg l-1) settled at approximately 1.3 mm s-1, and the mean settling speed for all particle size classes was around 0.3 mm s- 1. Morgan et al. (1997) determined the average sinking speed of E. affinis to be 3.3 mm s-1. Based on these results, it seems unlikely that, once suspended, sediments would sink faster than E. affinis if this copepod behaves as a passive particle (i.e., is not swimming). It is also known that sediment flocculation can change the density, mass and size characteristics of particles in the Chesapeake Bay ETM and other water bodies. Manning and Dyer (1999), using a laboratory flume, developed a power function predicting settling velocity (Ws) versus floc diameter (Dx) which had an r2 of 0.85: Ws = 0.0116 (Dx)0.8446. Flocs sizes may range up to 300 m in the Chesapeake Bay ETM (L. Sanford, pers. comm.), giving a calculated sinking speed of 1.43 mm s-1, still below that calculated 152 for E. affinis. Therefore, after factoring in flocculation processes, it still seems unlikely that sediment settling rate would be greater than E. affinis sinking velocities. A fixed, single-station analysis is suitable for analyzing changes in TSS and BV at a point of interest but does not provide details about the spatial complexity of sediments and zooplankton in the larger ETM region. It is possible that the separation between bottom and surface TSS and zooplankton BV peaks only occurs due to the time between sampling events ? the two peaks may be part of the same anomaly. Both time series have ?arms? of lower BV concentrations at mid-depths that seemingly connect well-defined surface and bottom peaks. Behavioral mechanisms, such as swimming and sinking, would serve to maintain these maxima. Assuming a sinking rate of 4 m for every 20 min (Morgan et al. 1997), a copepod could passively sink the equivalent depth of the Chesapeake Bay ETM in just over 1 hour. The above description of the time series at river km 30 assumes that the ETM channel is a closed system. Another possible explanation for the disjoined TSS and BV peaks is that tidal currents scouring adjacent shoals cause copepods to be swept into the surface waters of the channel. The synchronous timing of the surface and bottom peaks would correspond to the same process (tides) acting on different regions of the ETM (channel bed and shoal bed). A shoal effect such as this would also be expected to result in surface TSS peaks, which are not present. Unlike the Roman et al. (2001) results in 1996, a year of exceptionally high Susquehanna River input, the 2002 single fixed-station surveys did not reveal large resuspensions of TSS and zooplankton BV high into the water column during ebb tides. I did observe surface zooplankton BV increases during both flood and ebb 153 flows that may correspond to a resuspension of zooplankton from below. However, unlike the 1996 survey, there were no TSS maxima extending into the upper water column during either flood or ebb tides. Because of the differences in freshwater input between 1996 and 2002, comparisons of the physical conditions between those years should be undertaken with caution. The conclusions drawn from single-station analyses are limited because data from these stations could represent the vertical movements of either a single zooplankton population or the movements of multiple populations from down- or up- estuary through the station. The 25-hour multiple fixed-station axial surveys (Figs. 5 and 6) allow for greater spatial analysis of BV and TSS movements. However, as stated above, a pattern of zooplankton abundance by depth with respect to tidal stage is not apparent in contour plots of the multiple fixed-station surveys. Again, the general trend for the first survey is that zooplankton abundance is greatest near the surface while TSS peaks in bottom waters. During the second survey, TSS is also highest near the bottom, especially during flood tides, and BV peaks throughout the water column. Quite often, TSS and BV maxima in one transect seem to be horizontally transported to a new river km in plots of the next transect. It is likely that BV variations during the surveys were related to zooplankton being tidally transported, i.e., displaced horizontally with water movements, rather than to only vertical movements. The contour plots of the 25-hour surveys suggest that a different pattern of zooplankton resuspension and vertical transport was present in May surveys of 2002 and 1996, but the 2002 pattern is ambiguous based on the present data. Although 154 Tucker trawl data collected during the same time period as the acoustic surveys suggest that E. affinis is the most abundant zooplankton species in the ETM in 2002, it is possible that cladocerans in the up-estuary region and A. tonsa in the down- estuary region contaminate the acoustic survey data. Cladocerans are primarily found in greatest abundance near the surface in freshwater (North and Houde 2003; Lloyd, Chapter 1). High BV measurements at the surface during ebb tides may indicate cladocerans being transported down-estuary into the study area. Down-estuary of the ETM, variable BV measurements could indicate the presence of A. tonsa, which is usually found in the mesohaline regions of Chesapeake Bay. Migration behaviors of this copepod are primarily diel in nature and therefore may mask tidally-timed migration patterns of other taxa, namely E. affinis. The apparent inconsistency in zooplankton location relative to tidal phase between the two 2002 surveys as suggested by the multiple-station contour plots (Figs. 5 and 6) probably results from 1) incomplete sampling of the water column and 2) low tidal displacements relative to the size of the ETM region. The CTD and its associated instruments, including TAPS, incompletely sample the top 1-2 m of surface waters and bottom 1-2 m along the seabed. Eurytemora affinis is an epibenthic copepod, and incomplete bottom sampling may therefore miss a large proportion of its population(s) within the ETM, while partial surface sampling would create an incomplete dataset with regard to vertical migration patterns. The average axial tidal displacements for each depth class (surface, middle, and bottom) were a maximum of a few kilometers, while habitat in the 0-5 salinity range usually extended more than 10 km along the bay axis (depending on depth and tidal phase). Kimmerer 155 et al. (1998) suggested that salinity cues may be involved in vertical migration behaviors. Thus, any migration pattern may be masked by the fact that, over a single tidal cycle, ETM zooplankton may not be transported out of the optimal salinity range and will therefore not display a vertical migration pattern. Separating the ETM into up- and down-estuary regions allowed zooplankton within particular salinity ranges to be isolated, making migration patterns easier to statistically clarify. Assuming that most BV measured during these surveys belongs to E. affinis, one possible conclusion from the contour plots and regression analyses is that zooplankton in the ETM possess the ability to regulate their vertical positions in the water column and do so in response to tidal and salinity cues. The general pattern appears to be tidal suspension and active vertical migration toward the surface during ebb tides at up-estuary (freshwater/low salinity) locations and active migration and/or passive sinking toward the seafloor during flood tides in down-estuary (more saline) regions. In 2002, TSS and BV do not coincide in surface waters, and there is a different response of TSS and BV versus current magnitude relative to estuarine location (Figs. 7 and 8). The general relationship between tidal currents and TSS and zooplankton BV derived from regression analyses suggest that, in the up-estuary region of the ETM, zooplankton along the bottom are resuspended regardless of current direction. The parabolic relationship between both BV and TSS versus current direction suggests that zooplankton are passively resuspended as current velocity increases. Zooplankton BV concentrations in middle and surface waters increase with increasing ebb (seaward) tidal currents, especially at surface depths. This pattern would position copepods in locations with the greatest down-estuary 156 displacement. In the more saline, down-estuary area of the ETM, a near-opposite pattern exists. Zooplankton BV concentrations in the bottom and middle water column increase with increasing current velocities in the flood direction, placing them at depths with the greatest landward displacement. Based on this pattern, zooplankton are primarily found at depths where the likelihood of advective transport is maximized. Previous studies of tidally-timed vertical migration have concluded that vertical movement in which zooplankton are higher in the water column during flood and deeper during ebb should serve to minimize seaward flux and enhance position maintenance (Hough and Naylor 1992; Morgan et al. 1997; Kimmerer et al. 1998). The pattern described for the 2002 BV data would place zooplankton at depths where horizontal transport is most likely, moving zooplankton along a ?conveyor belt? of water movement ? up-estuary transport occurs along the bottom, and down-estuary transport occurs near the surface (Fig. 10). Therefore, zooplankton position would be maintained within the ETM over a larger area than if zooplankton vertically migrated to depths with the least horizontal transport. Roddie et al. (1984) showed that E. affinis survival was greatest in salinities between 0 and 5 at May water temperatures (15-19?C). During both May cruises, zooplankton BV was significantly more concentrated in lower salinities and appeared to be especially dense between salinities of 0 and 5 (Fig. 11). Although the vertical migration pattern described for the May cruises probably results in greater advection than the pattern described by Morgan et al (1997) and Kimmerer et al. (1998), it would appear to keep zooplankton in the low salinity ETM region. This ?conveyor 157 belt? migration pattern may also serve to keep E. affinis in areas of high food (phytoplankton, protists, and detritus with its associated biota) concentrations or may enhance its search for food. A variation on the ?conveyor belt? hypothesis described above is that the up- estuary region is a source of zooplankton that are tidally transported into the ETM. As they are transported down-estuary into more saline waters, the zooplankton may passively sink and/or actively migrate downward. Ensuing flood currents resuspend zooplankton from the bottom and transport them back up-estuary. Here, they may be resuspended, especially by ebb tides, and again transported down-estuary along with zooplankton recently advected from up-estuary. The 2002 survey contour plots appear to show a different pattern of zooplankton and TSS resuspension than plots of the 1996 data. As stated above, 1996 was one of the wettest years on record in the Chesapeake Bay watershed, resulting in high Susquehanna River freshwater input into Chesapeake Bay. In October of 1996, bottom ebb current magnitudes (driven by freshwater input) were greater than flood, while the ?average? October is usually marked by a minimum in freshwater input (see Fig. 9 in Roman et al. 2001). High Susquehanna River input pushes the ETM region further down-estuary (possibly beyond the survey area) and increases turbidity, presumably by increasing TSS resuspension (North and Houde 2001; Sanford et al. 2001). Enhanced ebb tides due to high freshwater input may have caused the high water column concentrations of TSS and zooplankton during ebb flow. This phenomenon was not present during the 2002 surveys. In 2002, zooplankton BV was unrelated to TSS concentrations in surface waters. The possibility of zooplankton 158 advection to and from adjacent shoals is improbable because surface TSS values do not increase or decrease with BV fluxes. Due to the ambiguity of these data, it is unclear as to whether or not zooplankton in the Chesapeake Bay ETM actively regulate their vertical position in the water column and do so in response to tidal and salinity cues. If they do, the general pattern appears to be vertical migration toward the surface during ebb tides at up-estuary locations and toward the seafloor during flood tides in down-estuary regions, with the Susquehanna River possibly supplying zooplankton to surface waters up-estuary of the ETM. Vertical migration behaviors might be more apparent during periods of reduced freshwater input via the Susquehanna River (i.e., 2002), presumably because of reduced habitat and a greater tidal influence in the upper Chesapeake Bay (Kimmel and Roman 2004). Enhanced freshwater inputs like those in 1996 create a larger-than-average ETM region that is located further down-estuary than the 2002 ETM (compare isohaline and maximum TSS locations between Roman et al. (2001) Figs. 2, 3, and 4 versus Lloyd et al. (Chapter 1) Figs. 5, 8, and 11, respectively), making comparisons between these two years difficult. The possible differences between the 1996 and 2002 ETM surveys should encourage future research on tidally-timed migration in the Chesapeake Bay ETM, which should ideally incorporate both acoustic and traditional (net, pump) sampling methods. While the 2002 data are inconclusive regarding a tidal migration pattern zooplankton, the parabolic relationship between near-bottom BV and current velocity up-estuary of the ETM suggests that if zooplankton do migrate, they do not migrate solely in response to tidal phase (Fig. 7, ?Bottom? frames). Further, if there is active, 159 tidally-timed migration in the Chesapeake Bay ETM, it does not serve to minimize advection, unlike migration patterns observed in other estuaries. 160 TABLES Table 5.1. Median, minimum, and maximum values for Total Suspended Solids (TSS, mg l-1) and Zooplankton Biovolume (BV, mm3 m-3) grouped for both surveys. Surface = 3-6m, Middle = 6-9m, Bottom = 9-12m. Depth Median Min Max Median Min Max Surface 10.4 6.4 25.0 920 12 88948 Middle 14.0 6.6 61.7 281 9 3379 Bottom 33.8 9.3 177.9 1096 107 25863 Total Suspended Solids Zooplankton BV 161 Table 5.2. Correlation coefficients by depth of pairwise correlation analyses between TSS and BV for both cruises. Surface = 0-4 m; Middle = 4-8 m, Bottom = 8-12 m. Values with an asterisk are statistically significant (p < 0.001). May 8-9 May 14-15 Surface -0.10 0 Middle 0.48* 0.33* Bottom 0.61* 0.57* 162 Cruise Tidal phase 3.5m 7.5m 11.5m May 8-9 Ebb -13.87 -9.47 -4.25 Flood 6.02 7.24 7.89 Ebb -10.97 -6.28 -4.73 Flood 9.74 10.04 6.65 Total -4.54 0.77 2.78 May 14-15 Tidal phase Ebb -4.06 -3.44 -3.04 Flood 1.85 3.33 2.97 Ebb -13.34 -4.80 -3.11 Flood 11.09 Total -2.23 -1.48 -0.13 Table 5.3. Calculated water displacements (km) for both surveys at three depths (3.5m, 7.5m, and 11.5m). Total displacement is the net average displacement over two tidal cycles except for 7.5 and 11.5m depths for the May 14-15 survey, which are calculated over one tidal cycle. Negative values indicate displacement down-estuary (ebb direction), positive values indicate displacement up-estuary (flood direction). 163 FIGURES Figure 5.1. Map of the United States east coast with the Chesapeake Bay ETM region magnified. Solid circles and the star represent the approximate locations of all stations of the 25-hour surveys. The star represents the location of the station at river km 30. The dotted line perpendicular to the bay axis represents the dividing line between up-estuary and down-estuary stations and is at approximately river km 28. Chester River Sassafras River Susquehanna River Patapsco River 76 76.5 76 76.5 39 39.5 L a t i t u d e ( ? N ) Longitude (?W) 10 20 30 kilometers 164 Figure 5.2. Comparison of TAPS Biovolume (BV) measurements with zooplankton counts from pump samples. 0 1000 2000 3000 4000 5000 0 100 200 300 400 500 600 May July October Pump (no. individuals m-3 ? 1000) T A P S B V ( m m 3 m - 3 ) r2 = 0.61 165 Figure 5.3. Salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1), and current velocity (bottom plot; positive values indicate flood currents, negative values indicate ebb currents) for the first survey (May 8-9). The solid line in the current plot represents currents at 11m depth, the stippled line represents currents at 3 m depth. The black bar above the top frame marks stations that were sampled at night. -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0 5 10 15 20 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 -10 -5 1 11 21 31 41 51 61 71 81 91 Time from start of survey (hours) D e p t h ( m ) C u r r e n t ( m s - 1 ) Flood Ebb B i o v o l u m e ( m m 3 m - 3 ) T S S ( m g l - 1 ) 166 Figure 5.4. Salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1), and current velocity (bottom plot; positive values indicate flood currents, negative values indicate ebb currents) for the second survey (May 14-15). The solid line in the current plot represents currents at 11m depth, the stippled line represents currents at 3 m depth. The black bar above the top frame marks stations that were sampled at night. -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 -10 -5 1 11 21 31 41 51 61 71 81 91 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 0 5 10 15 20 25 Time from start of survey (hours) D e p t h ( m ) C u r r e n t ( m s - 1 ) Flood Ebb T S S ( m g l - 1 ) B i o v o l u m e ( m m 3 m - 3 ) 167 Figure 5.5. Multiple fixed-station contour plots of salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1) versus river km (distance from Havre de Grace, Maryland) for the first survey (May 8-9). Transects were surveyed chronologically by letter (a-l). Tidal stage is given for each transect. The bold line represents the 1 PSU isohaline, each isohaline to the right increases by 2 PSU. 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 ebb to slack flood flood flood flood to slack ebb ebb slack to flood flood flood ebb ebb River km B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) a b c d e f g h i j k l 168 Figure 5.6. Multiple fixed-station contour plots of salinity (line contours; PSU), zooplankton biovolume (top contour plot; mm3 m-3), total suspended solids (middle contour plot; mg l-1) versus river km (distance from Havre de Grace, Maryland) for the second survey (May 14-15). Transects were surveyed chronologically by letter (a-k). Tidal stage is given for each transect. The bold line represents the 1 PSU isohaline, each isohaline to the right increases by 2 PSU. River km B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) B V ( m m 3 m - 3 ) T S S ( m g l - 1 ) a b c d e f g h i j k 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 25 30 35 40 -20 -15 -10 -5 1 501 1001 1501 2001 2501 3001 3501 4001 4501 25 30 35 40 -20 -15 -10 -5 1 7 13 19 25 31 37 43 49 55 ebb ebb late ebb to early flood flood ebb ebb ebb to slack flood flood flood early ebb 16 9 0 2000 4000 6000 8000 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 0 25 50 75 100 -1 -0.5 0 0.5 1 BV = 7020.5 (Vel) 2 + 315.4 (Vel) + 238.3 r 2 = 0.29 p < 0.001 TSS = 12.9 (Vel) + 27.4 r 2 = 0.11 p < 0.01 BV = -647.0 (Vel) + 286.6 r 2 = 0.05 p < 0.05 TSS = 3.1 (Vel) + 13.6 r 2 = 0.08 p < 0.01 BV = -2593.2 (Vel) + 1872.7 r 2 = 0.26 p < 0.001 TSS = 1.4 (Vel) + 10.6 r 2 = 0.15 p < 0.01 Bottom Middle Surface TS S (m g l -1 ) BV (m m 3 m -3 ) Current Velocity (m s -1 ) TSS 16.4 (Vel) 2 + 4.7 (Vel) + 10.6 r 2 = .27 p < . = 16.4 ( el) 2 4.7 ( el) 10.6 r 2 0.27 . TSS = 47.6 (Vel) 2 + 17.2 (Vel) + 20.7 r 2 = 0.19 p < 0.01 Figure 5.7. Regression analyses between both zooplankton biovolume (BV) and total suspended solids (TSS) versus current velocity (Vel) for transect stations up-estuary from river km 28. Bottom = 8-12 m depth; Middle = 4-8 m depth; Surface = 0-4 m depth. 17 0 0 2000 4000 6000 8000 -1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 0 25 50 75 100 -1 -0.5 0 0.5 1 BV = 2225.7 (Vel) + 1113.3 r 2 = 0.27 p < 0.001 TSS = 25.6 (Vel) + 31.0 r 2 = 0.29 p < 0.001 BV = 299.8 (Vel) + 522.5 r 2 = 0.07 p < 0.05 TSS = 5.4 (Vel) + 16.6 r 2 = 0.16 p < 0.001 p = 0.57, not significant TSS = 1.67 (Vel) + 12.2 r 2 = 0.11 p < 0.05 Bottom Middle Surface TS S (m g l -1 ) BV (m m 3 m -3 ) Current Velocity (m s -1 ) Figure 5.8. Regression analyses between both zooplankton biovolume (BV) and total suspended solids (TSS) versus current velocity (Vel) for transect stations down-estuary from river km 28. Bottom = 8-12 m depth; Middle = 4-8 m depth; Surface = 0-4 m depth. 171 Figure 5.9. Contour plots of Tucker trawl data from Chesapeake Bay ETM region, May, 2002. Line contours ? salinity (PSU); Color contours ? Eurytemora affinis and cladoceran (Bosmina longirostris and Daphnia spp.) abundance (individuals m-3); X- axis is distance down-estuary from Havre de Grace, Maryland. Abundance (individuals m-3) Eurytemora affinis Cladocera 172 28 km Flood Ebb seaward landward Figure 5.10. A hypothetical snapshot of zooplankton biovolume (BV; each represents approx. 500 mm-3 of BV) and total suspended solid (TSS; relative intensity indicated by shading) concentrations during maximum flood and ebb conditions in the Chesapeake Bay ETM region in May, 2002. The bold arrows in both frames represent relative current velocities in surface (1-4 m depth), middle (4-8 m depth) and bottom (8-12 m depth) portions of the water column, and the stippled line shows the mean isohaline structure. On the seaward side of the mid-point between all transects (28 km from Havre de Grace), BV concentrations in the bottom and middle portions of the water column are significantly greater during flood than ebb. On the landward side, BV concentrations are significantly greater in surface and middle waters during ebb than flood. During flood, bottom scouring resuspends both TSS and epibenthic zooplankton in both landward and seaward regions, especially in bottom waters. During the slack before ebb, TSS concentrations decrease as sediments settle out of the water column. Zooplankton swimming behaviors and zooplankton resuspension during the following ebb may result in significantly greater BV in surface and middle waters than during flood. Additionally, zooplankton swept down-stream from the Susquehanna River might add to zooplankton BV in surface waters, especially during ebb tides. The locations of BV maxima during the tidal cycle suggest that zooplankton tend to be at the depths of maximum current velocities, resulting in a ?conveyor belt? of zooplankton movement. TSS concentrations are generally greater in both landward and seaward regions during flood than ebb at all depths, probably due to greater bottom stress during flood. Zooplankton from the Susquehanna River 173 Figure 5.11. Linear regression of zooplankton biovolume (BV) versus salinity of all May data. 0 20000 40000 60000 80000 100000 0 5 10 15 Salinity B V ( m m 3 m - 3 ) BV = -133.3 (S) + 2006.0 p < 0.01 174 LITERATURE CITED Barans, C. A., B. W. Stender, D. V. Holliday and C. F. Greenlaw. 1997. Variation in the vertical distribution of zooplankton and fine particles in an estuarine inlet of South Carolina. Estuaries 20(3):467-482. Castel, J. and J. Veiga. 1990. Distribution and retention of the copepod Eurytemora affinis hirundoides in a turbid estuary. Marine Biology 107(1):119-128. Cronin, T. W., and R. B. Forward, Jr. 1983. Vertical migration rhythms of newly hatched larvae of the estuarine crab, Rhithropanopeus harrisii. Biological Bulletin 165:139-153. Cushing, D. H. 1951. The Vertical Migration of Planktonic Crustacea. Biological Reviews of the Cambridge Philosophical Society 26(2):158-192. Dauvin, J.-C. and J. Dodson. 1990. Relationship between feeding incidence and vertical and longitudinal distribution of rainbow smelt larvae (Osmerus mordax) in a turbid well-mixed estuary. Marine Ecology Progress Series 60:1- 12. Forward, R.B., Jr. and R.A. Tankersley. 2001. Selective tidal-stream transport of marine animals. In R.N. Gibson, M. Barnes, R.J.A. Atkinson, (eds.). Oceanography and marine biology, an annual review. London: Taylor Francis. 305-353. Frost, B. W. 1988. Variability and possible adaptive significance of diel vertical migration in Calanus pacificus, a planktonic marine copepod. Bulletin of Marine Science 43(3):675-694. Fugate, D. C. and C. T. Friedrichs. 2002. Determining concentration and fall velocity of estuarine particle populations using ADV, OBS, LISST. Continental Shelf Research 22:1867-1886. Holliday, D. V. and R. E. Pieper. 1995. Bioacoustical oceanography at high frequencies. ICES Journal of Marine Science 52:3-4. Hough, A. R. and E. Naylor. 1991. Field studies on retention of the planktonic copepod Eurytemora affinis in a mixed estuary. Marine ecology progress Series. Oldendorf. 76(2):115-122. 175 Hough, A. R. and E. Naylor. 1992. Endogenous Rhythms of Circatidal Swimming Activity in the Estuarine Copepod Eurytemora-Affinis (Poppe). Journal of Experimental Marine Biology and Ecology 161(1):27-32. Huntley, M. and E. R. Brooks. 1982. Effects of age and food availability on diel vertical migration of Calanus pacificus. Marine Biology 71(1):23-31. Joyeux, J. C. 1999. The abundance of fish larvae in estuaries: Within-tide variability at inlet and immigration. Estuaries 22:889-904. Kimmel, D. G. and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Kimmerer, W. J., J. R. Burau and W. A. Bennett. 1998. Tidally oriented vertical migration and position maintenance of zooplankton in a temperate estuary. Limnology and Oceanography 43(7):1697-1709. Kimmerer, W. J. 2002. Physical, Biological, and Management Responses to Variable Freshwater Flow into the San Francisco Estuary. Estuaries 25(6B):1275-1290. Manning, A. J. and K. R. Dyer. 1999. A laboratory examination of floc characteristics with regard to turbulent shearing. Marine Geology 160:147-170. Morgan, C., JR Cordell, and CA Simenstad. 1997. Sink or swim? Copepod population maintenance in the Columbia River estuarine turbidity maxima region. Marine Biology 129:309-317. North, E. W. and E. D. Houde. 2001. Retention of White Perch and Striped Bass Larvae: Biological-Physical Interactions in Chesapeake Bay Estuarine Turbidity Maximum. Estuaries 24(5):756-769. North, E. W. and E. D. Houde. 2003. Linking ETM physics, zooplankton prey, and fish early-life histories to striped bass Morone saxatilis and white perch M. Americana recruitment. Marine Ecology Progress Series 260:219-236. Ohman, M. D. 1990. The demographic benefits of diel vertical migration by zooplankton. Ecological Monographs 60(3):257-281. Peitsch, A., B. Koepcke and N. Bernat. 2000. Long-term investigation of the distribution of Eurytemora affinis (Calanoida; Copepoda) in the Elbe Estuary. Limnologica Jena 30(2):175-182. 176 Pieper, R. E., D. E. McGehee, C. F. Greenlaw and D. V. Holliday. 2001. Acoustically measured seasonal patterns of Zooplankton in the Arabian Sea. Deep Sea Research 48:6-7. Ringelberg, J. 1995. Changes in light intensity and diel vertical migration: A comparison of marine and freshwater environments. Journal of the Marine Biological Association of the United Kingdom 75(1):15-25. Roberts, D., D. Rittschof, D. J. Gerhart, A. R. Schmidt, and L. G. Hill. 1989. Vertical migration of the clam Mercenaria mercenaria (L.) (Mollusca: Bivalvia). Environmental correlates and ecological significance. Journal of Experimental Marine Biology and Ecology 126:271-280. Roddie, B. D., R. J. G. Leakey and A. J. Berry. 1984. Salinity-temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth Estuary. Journal of Experimental Marine Biology and Ecology 77(2):191-211. Roman, M. R., D. V. Holliday and L. P. Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. Sanford, L. P., S. E. Suttles and J. P. Halka. 2001. Reconsidering the physics of the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):655-669. Sautour, B. and J. Castel. 1995. Comparative Spring Distribution of Zooplankton in 3 Macrotidal European Estuaries. Hydrobiologia 311(1-3):139-151. Simenstad, C. A., L. F. Small, and C. D. McIntire. 1990. Consumption processes and food web structure in the Columbia River estuary. Progress in Oceanography 25:271-297. Simenstad, C. A., C. A. Morgan, J. R. Cordell, and J. A. Baross. 1994. Flux, passive retention, and active residence of zooplankton in Columbia River estuarine turbidity maxima. Olsen & Olsen, Fredensborg, Denmark. Tankersley, R. A., and R. B. Forward, Jr. 1994. Endogenous swimming rhythms in estuarine crab megalopae: Implications for flood-tide transport. Marine Biology 118:415-423. Verheye, H. M. and J. G. Field. 1992. Vertical distribution and diel vertical migration of Calanoides carinatus (Kroeyer, 1849) developmental stages in the southern Benguela upwelling region. Journal of Experimental Marine Biology and Ecology 158(1):123-140. 177 Zeng, C., P. Abello and E. Naylor. 1999. Endogenous tidal and semilunar moulting rhythms in early juvenile shore crabs Carcinus maenas: Implications for adaptation to a high intertidal habitat. Marine Ecology Progress Series 191:257-266. 178 CHAPTER VI: General Conclusions SUMMARY AND CONCLUSIONS Although primary production is low in the Chesapeake Bay ETM (Fisher et al. 1988), this region is a dynamic, often flourishing, ecosystem. Phytoplankton and detritus washed into this region from upstream, the bacteria and protozoa organisms associated with them, and free-living microzooplankton all likely combine to form the base of the food web in the ETM. The calanoid copepod Eurytemora affinis is especially suited to occupy such a habitat given its 1) preference for low salinity and 2) ability to select particles based on nutritional value, which is enough to create a chlorophyll-a (chl-a) ?hole? in this region. This copepod appears to feed more efficiently outside of the particle milieu of the ETM but is limited to the ETM region due to greater predation pressures down-estuary. Chapter 1 The constraints (physical versus biological) on E. affinis distribution appear to be primarily based on the strength of Susquehanna River input into Chesapeake Bay. During periods of low flow (below the annual mean), the size of the ETM is reduced, and its location is pushed up-estuary (Kimmel and Roman 2004). Such conditions make biological interactions more discernible. For example, Susquehanna River flow generally peaks in the springtime (March, April, May) and is at a minimum around August; July and October tend to have similar values. Early-to-mid autumn flow may also increase if tropical cyclones pass over the Chesapeake Bay region. In 2001 and 2002, years of average or slightly below average freshwater input, Tucker trawl data 179 show that E. affinis and Acartia tonsa distributions often overlapped in the spring but were spatially partitioned in the summer and fall. Laboratory experiments demonstrate that E. affinis is able to survive and reproduce at salinities and temperatures above those in which it was found at these times, suggesting that interactions with A. tonsa and/or mysids limited its down-stream abundance. Mysids and A. tonsa may also limit downstream E. affinis abundance during high flow periods, but, because such conditions increase available habitat, interactions between E. affinis and mesohaline zooplankton would affect a smaller proportion of the population. More research concerning interactions between E. affinis and A. tonsa, mysids, ctenophores, and andadromous fish larvae should be performed for several reasons. Identifying the effects of these groups on E. affinis? abundance and reproductive rates will help to further elucidate why E. affinis is constrained primarily to the ETM. As these groups all directly interact, it would also be helpful to know, for example, what fraction of mysid prey consists of E. affinis and the proportions in which fish larvae prey on these groups. Because the ETM appears to be such a unique ecosystem, tracking carbon through the food web is interesting both academically and economically (i.e. fisheries). Chapter 2 Egg production rates have often been used to estimate somatic growth due to the general ease of determining EP. This approach assumes that any resources that would go into somatic growth for juveniles would be allocated to reproductive growth in adults. The fact that 2002 EP estimates by the clutch size method fit all three 180 growth models, especially the Hirst-Sheader model, supports the use of EP estimates to estimate juvenile somatic growth under food-replete conditions. Caution should be taken when estimating EP with the clutch size method under low food conditions, as laboratory estimates of ID may not be representative of field ID. Further, it must be noted that because of the poor fit of the 2003 data to growth models, indirect methods of estimating copepod egg production, namely the egg-ratio method and clutch size method, may not always be appropriate, and incubations may provide better estimates. Two issues with the use of the clutch size method for estimate copepod egg production rates are that 1) the experiments to determine ID are time intensive, and 2) ID/temperature relationships for one copepod species cannot be extrapolated to another. Further complicating such indirect methods of estimating egg production rates, E. affinis may actually represent a species complex (Lee 1999) where particular rates or tolerances appear to differ between populations. This is suggested by differences in E. affinis ID/temperature relationships between Vijverberg (1980) and Lloyd (Chapter 2). Acclimating E. affinis captured in low salinity waters to freshwater conditions has proven difficult (pers. obs.), though many E. affinis populations are found in the freshwater regions of river and in lakes (Ban 1994; Lee 1999). There is also the possibility that copepod ID is dependent on the size of the female. Eurytemora affinis females used in the ID experiments were collected from waters around 10?C and incubated at temperatures ranging from 5 to 27.5?C. While these females were about the same size, the same incubation temperatures in the field 181 would be associated with adult copepods of different sizes due to the effect of temperature on development rate (lower temperatures = larger adults, higher temperatures = smaller adults). While temperature is negatively related to female size, female size is positively related to clutch size, and a larger clutch (from a larger female) may take longer to create than a smaller clutch (from a smaller female). Therefore, if the goal of ID duration experiments is to estimate egg production rates of copepods in the field, the experiments should be conducted on females whose size is representative of the temperature at which they are incubated. It would be interesting to compare the results of such an experiment to the results of Chapter 2. Chapter 3 Increases in freshwater input cause an increase in particulate input (Sanford et al. 2001) and, as stated above, an increase in the nutritional quality of particulates. Research during 2002 and 2003 in the Chesapeake Bay and Choptank River ETMs on food limitation suggests that, even during dry years, E. affinis is not food limited in terms of food quantity. There may be a slight increase food quality during wet years, but the increase in available habitat seems to be the greatest benefit to copepods. Optimally, the research for this chapter would involve measuring E. affinis egg production in the same system (either Chesapeake Bay or the Choptank River) over multiple years. Also, microplankton (64-200 m) counts for both years would have helped to better characterize the Chesapeake Bay and Choptank River ETMs. Although the effects of freshwater input on the size and intensity of the ETMs in both systems seems to be the same, there is no guarantee that both E. affinis populations respond the same way to particular variables (see the Chapter 2 summary 182 above). However, the conclusion that E. affinis is not food limited in the ETM is supported by both the clutch size/prosome regression and the results of the grazing experiment. Chapter 4 Freshwater input also influences E. affinis vertical distributions. Dry periods (months or years) allow for copepod responses to tidal current/salinity changes to affect their position within the ETM. Specifically, zooplankton migrate toward the surface during flood tides and downward during ebb, thus staying at depths with the lowest horizontal current velocities. When freshwater input is high, zooplankton may be swept out of the upper portion of the ETM due the enhanced seaward flow, thus masking any vertical migration patterns. Because the ETM usually extends further down-estuary during these conditions, zooplankton that are transported seaward may not necessarily be transported out of the ETM habitat. If I were to carry out the vertical migration project again, I would collect data in April, optimally during a period of average freshwater input. Under such conditions, E. affinis abundance tends to be high while Acartia tonsa abundance is low and is greatest down-estuary of the ETM. The sampling protocol ? 25-hour, multiple-station transects ? is ideal for analyzing vertical movements that also have a tidal (horizontal) component, and the use of TAPS to collect BV data allowed for rapid data collection not possible through traditional net or pump sampling. The data might have been improved by binning different zooplankton size ranges, specifically to differentiate between E. affinis nauplii and copepodite stages. Also, tidal displacement could have been directly measured using an inert tracer. 183 The Chesapeake Bay and Choptank River turbidity maxima are unique, dynamic ecosystems. Eurytemora affinis, the most abundant zooplankton species in these regions for much of the year, is well adapted to ETM salinity, current, and food conditions. It is possible that competitive and predatory interactions with other zooplankton (A. tonsa, mysids, larval fish) exclude E. affinis from down-estuary regions that it may otherwise successfully colonize. Its ability to utilize the resources in highly turbid regions, however, allow for population abundances that are often greater than other species? abundances further up- or down-estuary of ETMs. 184 LITERATURE CITED Devreker, D., S. Souissi and L. Seuront. 2004. Development and mortality of the first naupliar stages of Eurytemora affinis (Copepoda, Calanoida) under different conditions of salinity and temperature. Journal of Experimental Marine Biology and Ecology 303(1):31-46. Fisher, T. R., L. W. Harding, Jr., D. W. Stanley and L. G. Ward. 1988. Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson estuaries. Estuarine, Coastal and Shelf Science 27(1):61-93. Ishikawa, A., S. Ban and N. Shiga. 1999. Effects of salinity on survival and embryonic and postembryonic development of Eurytemora affinis from a freshwater lake. Plankton Biology and Ecology 46:113-119. Katona, S. K. 1970. Growth characteristics of the copepods Eurytemora affinis and E. herdmani in laboratory cultures. Helgolander wiss. Meeresunters. 20:373-384. Kimmel, D. G. and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Lee, C. E. 1999. Rapid and repeated invasions of fresh water by the copepod Eurytemora affinis. Evolution 53(5):1423-1434. Sanford, L. P., S. E. Suttles and J. P. Halka. 2001. Reconsidering the physics of the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):655-669. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10(4):317-340. 185 COMPLETE LITERATURE CITED Ambler, J. W. 1985. Seasonal factors affecting egg production and viability of eggs of Acartia tonsa Dana from East Lagoon, Galveston, Texas. Estuarine, Coastal, and Shelf Science 20:743-760. Ambler, J. W. 1986. Effect of food quantity and quality on egg production of Acartia tonsa Dana from East Lagoon, Galveston, Texas. Estuarine, Coastal and Shelf Science 23(2):183-196. Andersen, C. M., and T. G. Nielsen. 1997. Hatching rate of the egg-carrying estuarine copepod Eurytemora affinis. Marine Ecology Progress Series 160:283-289. Azeiteiro, U. M. M., L. Jesus and J. C. Marques. 1999. Distribution, population dynamics, and production of the suprabenthic mysid Mesopodopsis slabberi in the Mondego Estuary, Portugal. Journal of Crustacean Biology 19(3):498-509. Ban, S. H. 1994. Effect of temperature and food concentration on postembryonic development, egg-production and adult body-size of the calanoid copepod Eurytemora-affinis. Journal of Plankton Research 16:721-735. Ban, S., and T. Minoda. 1992. Hatching of diapause eggs of Eurytemora-affinis (Copepoda, Calanoida) collected from lake-bottom sediments. Journal of Crustacean Biology 12:51-56. Barans, C. A., B. W. Stender, D. V. Holliday and C. F. Greenlaw. 1997. Variation in the vertical distribution of zooplankton and fine particles in an estuarine inlet of South Carolina. Estuaries 20(3):467-482. Barquero, S., J. A. Cabal, R. Anadon, E. Fernandez, M. Varela and A. Bode. 1998. Ingestion rates of phytoplankton by copepod size fractions on a bloom ssociated with an off-shelf front off NW Spain. Journal of Plankton Research 20(5):957-972. Blehr?dek, J. 1935. Temperature and living matter. Protoplasma, Monograph 8:1- 277. Berggreen, U., B. Hansen and T. Kioerboe. 1988. Food size spectra, ingestion and rowth of the copepod Acartia tonsa during development: Implications for determination of copepod production. Marine Biology 99(3):341-352. Berk, S. G., D. C. Brownlee, D. R. Heinle, H. J. Kling and R. R. Colwell. 1978. Ciliates as a food source for marine planktonic copepods. Microbial Ecology 4(1):27-40. 186 Bligh, E. G. and W. J. Dyer. 1959. A Rapid Method of Total Lipid Extraction and Purification. Canadian Journal of Biochemistry and Physiology 37(8):911- 917. Boak, A. C. and R. Goulder 1983. Bacterioplankton in the diet of the calanoid copepod Eurytemora sp. in the Humber Estuary. Marine Biology. 73:139-149. Bollens, G. C. R. and D. L. Penry. 2003. Feeding dynamics of Acartia spp. copepods in a large, temperate estuary (San Francisco Bay, CA). Marine Ecology Progress Series 257:139-158. B?ttger, R., and D. Schnack. 1986. On the effect of formaldehyde fixation on the dry weight of copepods. Meeresforschung/Rep. Mar. Res. 31:141-152. Burdloff, D., S. Gasparini, B. Sautour, H. Etcheber, and J. Castel. 1999. Is the copepod egg production in a highly turbid estuary (the Gironde, France) a function of the biochemical composition of seston? Aquatic Ecology. 34:165- 175. Burdloff, D., S. Gasparini, F. Villate, I. Uriarte, U. Cotano, B. Sautour and H. Etcheber. 2002. Egg production of the copepod Acartia bifilosa in two contrasting European estuaries in relation to seston composition. Journal of Experimental Marine Biology and Ecology 274(1):1-17. Burkill, P. H., and T. F. Kendall. 1982. Production of the copepod Eurytemora affinis in the Bristol Channel. Marine Ecology Progress Series 7:21-31. Castel, J and J Viega. 1990. Distribution and retention of the copepod Eurytemora- affinis-hirudinoides in a turbid estuary. Marine Biology 107(1):119-128. Castonguay, M. and G. J. FitzGerald. 1990. The ecology of the calanoid copepod Eurytemora affinis in salt marsh tide pools. Hydrobiologia 202(3):125-133. Cervetto, G., R. Gaudy and M. Pagano. 1999. Influence of salinity on the distribution of Acartia tonsa (Copepoda, Calanoida). Journal of Experimental Marine Biology and Ecology 239(1):33-45. Chigbu, P. 2004. Assessment of the potential impact of the mysid shrimp, Neomysis mercedis, on Daphnia. Journal of Plankton Research 26(3):295-306. Chow-Fraser, P. and E. J. Maly. 1991. Factors governing clutch size in two species of Diaptomus (Copepoda: Calanoida). Canadian Journal of Fisheries and Aquatic Sciences 48(3):364-370. Christie, H., N. M. Jorgensen, K. M. Norderhaug and E. Waage-Nielsen. 2003. Species distribution and habitat exploitation of fauna associated with kelp 187 (Laminaria hyperborea) along the Norwegian coast. Journal of the Marine Biological Association of the United Kingdom 83(4):687-699. Conover, R. 1978. Transformations of organic matter. In Kinne, O. (ed.), Marine Ecology, vol. 4, Dynamics. Wiley, Chichester, pp. 221-499. Conover, R. J. 1966. Assimilation of organic matter zooplankton. Limnology and Oceanography 11(3):338-345. Crawford, P. and G. R. Daborn. 1986. Seasonal variations in body size and fecundity in a copepod of turbid estuaries. Estuaries 9(2):133-141. Cronin, T. W., and R. B. Forward, Jr. 1983. Vertical migration rhythms of newly hatched larvae of the estuarine crab, Rhithropanopeus harrisii. Biological Bulletin 165:139-153. Crump, B. C., J. A. Baross and C. A. Simenstad. 1998. Dominance of particle- attached bacteria in the Columbia River estuary, USA. Aquatic Microbial Ecology 14(1):7-18. Cuker, B. E. and M. A. Watson. 2002. Diel Vertical Migration of Zooplankton in Contrasting Habitats of the Chesapeake Bay. Estuaries 25(2):296-307. Cushing, D. H. 1951. The Vertical Migration of Planktonic Crustacea. Biological Reviews of the Cambridge Philosophical Society 26(2):158-192. Dauvin, J.-C. and J. Dodson. 1990. Relationship between feeding incidence and vertical and longitudinal distribution of rainbow smelt larvae (Osmerus mordax) in a turbid well-mixed estuary. Marine Ecology Progress Series 60:1- 12. De'ath, G. and K. E. Fabricius. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology 81(11):3178-3192. DeMott, W. R. 1998. Utilization of a cyanobacterium and a phosphorus-deficient green alga as complementary resources by daphnids. Ecology 79:2463-2481. Devreker, D., S. Souissi and L. Seuront. 2004. Development and mortality of the first naupliar stages of Eurytemora affinis (Copepoda, Calanoida) under different conditions of salinity and temperature. Journal of Experimental Marine Biology and Ecology 303(1):31-46. Dubois, M., K. A. Gilles, J. K. Hamilton, P. A. Rebers and F. Smith. 1956. Colorimetric Method for Determination of Sugars and Related Substances. Analytical Chemistry 28(3):350-356. 188 Dzeroski, S. and D. Drumm. 2003. Using regression trees to identify the habitat preference of the sea cucumber (Holothuria leucospilota) on Rarotonga, Cook Islands. Ecological Modelling 170(2-3):219-226. Edmondson, W. T. 1968. A graphical model for evaluating the use of the egg ratio for measuring birth and death rates. Oecologia 1:1-37. Edmondson, W. T., G. W. Comita, and G. C. Anderson. 1962. Reproductive rate of copepods in nature and its relation to phytoplankton population. Ecology 43:625-634. Elser, J. J., and J. Urabe. 1999. The stoichiometry of consumer-driven nutrient recycling: Theory, observations, and consequences. Ecology 80:735-751. Fisher, T. R., L. W. Harding, Jr., D. W. Stanley and L. G. Ward. 1988. Phytoplankton, nutrients, and turbidity in the Chesapeake, Delaware, and Hudson estuaries. Estuarine, Coastal and Shelf Science 27(1):61-93. Fockedey, N. and J. Mees. 1999. Feeding of the hyperbenthic mysid Neomysis integer in the maximum turbidity zone of the Elbe, Westerschelde and Gironde estuaries. Journal of Marine Systems. 22:2-3. Forward, R.B., Jr. and R.A. Tankersley. 2001. Selective tidal-stream transport of marine animals. In R.N. Gibson, M. Barnes, R.J.A. Atkinson, (eds.). Oceanography and marine biology, an annual review. London: Taylor Francis. 305-353. Froelich, P. N. 1988. Kinetic control of dissolved phosphate in natural rivers and estuaries: A primer on the phosphate buffer mechanism. Limnology and Oceanography 33:649-668. Frost, B. W. 1972. Effects of size and concentration of food particles on the feeding behavior of the marine planktonic copepod Calanus pacificus. Limnology and Oceanography 17(6):805-815. Frost, B. W. 1988. Variability and possible adaptive significance of diel vertical migration in Calanus pacificus, a planktonic marine copepod. Bulletin of Marine Science 43(3):675-694. Fugate, D. C. and C. T. Friedrichs. 2002. Determining concentration and fall velocity of estuarine particle populations using ADV, OBS, LISST. Continental Shelf Research 22:1867-1886. Gasparini, S. and J. Castel. 1997. Autotrophic and heterotrophic nanoplankton in the diet of the estuarine copepods Eurytemora affinis and Acartia bifilosa. Journal of Plankton Research 19(7):877-890. 189 Gasparini, S., J. Castel and X. Irigoien. 1999. Impact of suspended particulate matter on egg production of the estuarine copepod, Eurytemora affinis. Journal of Marine Systems 22(2-3):195-205. Gaston, K. J. and J. H. Lawton. 1990. Effects of scale and habitat on the relationship between regional distribution and local abundance. Oikos 58(3):329-335. Gifford, D. J. 1993. Planktonic protozoa in copepod diet: Examples from the North Atlantic and the North Pacific. ICES 1(1): Gilmurray, M. C. and G. R. Daborn. 1981. Feeding relations of the Atlantic silverside Menidia menidia in the Minas Basin, Bay of Fundy. Marine Ecology Progress Series. 6(3):231-235. Gismervik, I. 1997a. Stoichiometry of some marine planktonic crustaceans. Journal of Plankton Research 19:279-285. Gismervik, I. 1997b. Implications of zooplankton stoichiometry on distribution of N and P among planktonic size fractions. Journal of Plankton Research 19:343- 356. Gliwicz, Z. M. 1980. Filtering rates, food size selection, and feeding rates in cladocerans - another aspect of interspecific competition in filter-feeding zooplankton. In W. C. Kerfoot (ed.), Evolution and ecology of zooplankton communities. Goldman, J. C. 1979. Temperature effects on steady-state growth, phosphorus uptake, and the chemical composition of a marine phytoplankter. Microbial Ecology 5:153-166. Goolish, E. M., and R. S. Burton. 1989. Energetics of osmoregulation in an intertidal copepod: Effects of anoxia and lipid reserves on the pattern of free amino acid accumulation. Functional Ecology 3:81-89. Graneli, E. and J. T. Turner. 2002. Top-down regulation in ctenophore-copepod- ciliate- diatom-phytoflagellate communities in coastal waters: A mesocosm study. Marine ecology progress series 239:57-68. Hasle, G. R. 1978. The inverted microscope method. in A. Sournia, editor. Phytoplankton manual. UNESCO, Paris. Heinle, D. R. and D. A. Flemer. 1975. Carbon requirements of a population of the estuarine copepod Eurytemora affinis. Marine biology 31:235-247. Heinle, D. R., R.P. Harris, J.F. Ustach, and D.A. Flemer. 1977. Detritus as food for estuarine copepods. Marine Biology 40:341-353. 190 Hirche, H. J. 1992. Egg-production of Eurytemora-affinis - effect of K-strategy. Estuarine Coastal and Shelf Science 35:395-407. Hirst, A. G. and A. D. McKinnon 2001. Does copepod egg production represent adult female copepod growth? A call to account for body weight changes. Marine Ecology Progress Series 223:179-199. Hirst, A. G., and M. Sheader. 1997. Are in situ weight-specific growth rates body-size independent in marine planktonic copepods? A re-analysis of the global syntheses and a new empirical model. Marine Ecology Progress Series. 154:155-165. Hoffmeyer, M. S. 2004. Decadal change in zooplankton seasonal succession in the Bahia Blanca estuary, Argentina, following introduction of two zooplankton species. Journal of Plankton Research 26(2):181-189. Holliday, D. V. and R. E. Pieper. 1995. Bioacoustical oceanography at high frequencies. ICES Journal of Marine Science 52:3-4. Hopcroft, R. R. and J. C. Roff. 1998. Zooplankton growth rates: the influence of female size and resources on egg production of tropical marine copepods. Marine Biology 132:79-86. Hough, A. R. and E. Naylor. 1991. Field studies on retention of the planktonic copepod Eurytemora affinis in a mixed estuary. Marine ecology progress Series. Oldendorf. 76(2):115-122. Hough, A. R. and E. Naylor. 1992. Endogenous Rhythms of Circatidal Swimming Activity in the Estuarine Copepod Eurytemora-Affinis (Poppe). Journal of Experimental Marine Biology and Ecology 161(1):27-32. Howarth, R. W., H. S. Jensen, R. Marino, and H. Postma. 1995. Transport To and Processing Of P in Near-shore and Oceanic Waters. in H. Tiessen, editor. Phophorus in the Global Environment: Transfers, Cycles, and Management. John Wiley and Sons, West Sussex, England. Huntley, M. and C. Boyd. 1984. Food-limited growth of marine zooplankton. The American Naturalist 124(4):455-478. Huntley, M. and E. R. Brooks. 1982. Effects of age and food availability on diel vertical migration of Calanus pacificus. Marine Biology 71(1):23-31. Huntley, M. and M. D. G. Lopez. 1992. Temperature-dependent production of marine copepods: a global synthesis. The American Naturalist 140:201-242. 191 Ianora, A. and S. A. Poulet. 1615. Egg viability in the copepod Temora stylifera. Limnology and Oceanography 38(8):1615-1626. Ikeda, T., and S. Motoda. 1978. Estimated zooplankton production and their ammonia excretion in the Kuroshio and adjacent seas. Fish. Bull. 76:357-367. Irigoien, X. and J. Castel. 1997. Light limitation and distribution of chlorophyll pigments in a highly turbid estuary: The Gironde (SW France). Estuarine, Coastal and Shelf Science 44(4):507-517. Irigoien, X., J. Castel, and B. Sautour. 1993. In-situ grazing activity of planktonic copepods in the Gironde Estuary. Cahiers De Biologie Marine 34:225-237. Ishikawa, A., S. Ban and N. Shiga. 1999. Effects of salinity on survival and embryonic and postembryonic development of Eurytemora affinis from a freshwater lake. Plankton Biology and Ecology 46:113-119. Jacobs, F., W. Burton, and I. Moss. 1985. Patterns of zooplankton abundance, species composition, and biomass in upper Chesapeake Bay. Estuaries 8:78A. Jonasdottir, S. H. 1994. Effects of food quality on the reproductive success of Acartia tonsa and Acartia hudsonica: Laboratory observations. Marine Biology 121(1):67-81. Jonasdottir, S. H., D. Fields, and S. Pantoja. 1995. Copepod egg production in Long Island Sound, USA, as a function of the chemical compostion of seston. Marine Ecology Progress Series 119:87-98. Joyeux, J. C. 1999. The abundance of fish larvae in estuaries: Within-tide variability at inlet and immigration. Estuaries 22:889-904. Kankaala, P. and S. Johansson. 1986. The influence of individual variation on length- biomass regressions in three crustacean zooplankton species. Journal of Plankton Research 8(6):1027-1038. Katona, S. K. 1970. Growth characteristics of the copepods Eurytemora affinis and E. herdmani in laboratory cultures. Helgol?nder wiss. Meeresunters. 20:373-384. Katona, S. K. 1975. Copulation in the copepod Eurytemora affinis (Poppe, 1880). Crustaceana 28:89-95. Kemp, W. M. and W. R. Boynton. 1984. Spatial and temporal coupling of nutrient inputs to estuarine primary production: The role of particulate transport and decomposition. Bulletin of Marine Science 35:522-535. 192 Kimmel, D. G. and B. P. Bradley. 2001. Specific protein responses in the calanoid copepod Eurytemora affinis (Poppe, 1880) to salinity and temperature variation. Journal of Experimental Marine Biology and Ecology 266(2):135- 149. Kimmel, D. G. and B. P. Bradley. 2001. Specific protein responses in the calanoid copepod Eurytemora affinis (Poppe, 1880) to salinity and temperature variation. Journal of Experimental Marine Biology and Ecology 266(2):135- 149. Kimmel, D. G. and M. R. Roman. 2004. Long-term trends in mesozooplankton abundance in Chesapeake Bay, USA: influence of freshwater input. Marine Ecology Progress Series 267:71-83. Kimmerer, W. J. 2002. Physical, Biological, and Management Responses to Variable Freshwater Flow into the San Francisco Estuary. Estuaries 25(6B):1275-1290. Kimmerer, W. J., J. R. Burau and W. A. Bennett. 1998. Tidally oriented vertical migration and position maintenance of zooplankton in a temperate estuary. Limnology and Oceanography 43(7):1697-1709. Koski, M. 1999. Carbon:nitrogen ratios of Baltic Sea copepods-indication of mineral limitation? Journal of Plankton Research 21:1565-1573. Landry, M. R., W. K. Peterson and C. J. Lorenzen. 1995. Zooplankton Grazing, Phytoplankton Growth, and Export Flux - Inferences from Chlorophyll Tracer Methods. Ices Journal of Marine Science 52(3-4):337-345. Laprise, R., J.J. Dodson. 1994. Environmental variability as a factor controlling spatial patterns in distribution and species-diversity of zooplankton in the St. Lawrence Estuary. Marine Ecology Progress Series 107(1-2):67-81. Lee, C. E. 1999. Rapid and repeated invasions of fresh water by the copepod Eurytemora affinis. Evolution 53(5):1423-1434. Lonsdale, D. J., D. R. Heinle and C. Siegfried. 1979. Carnivorous feeding behavior of the adult calanoid copepod Acartia tonsa Dana. J. Exp. Mar. Biol. Ecol. 36(3):235-248. Manning, A. J. and K. R. Dyer. 1999. A laboratory examination of floc characteristics with regard to turbulent shearing. Marine Geology 160:147-170. Mauchline, J. 1998. The biology of calanoid copepods. Academic Press, San Diego. 193 McLaren, I. A. 1966. Predicting development rate of copepod eggs. Biological Bulletin 131:457-469. McLaren, I. A. 1978. Generation lengths of some temperate marine copepods: estimation, prediction, and implications. J. Fish. Res. Board Can. 35(10):1330-1342. Menden-Deuer, S. and E. J. Lessard. 2000. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnology and Oceanography 45(3):569-579. Merrell, J. R. and D. K. Stoecker. 1998. Differential grazing on protozoan microplankton by developmental stages of the calanoid copepod Eurytemora affinis Poppe. Journal of Plankton Research 20(2):289-304. Montagnes, D. J. S., J. A. Berges, P. J. Harrison and F. J. R. Taylor. 1994. Estimating carbon, nitrogen, protein, and chlorophyll a from volume in marine phytoplankton. Limnology and Oceanography 39(5):1044-1060. Morgan, C., JR Cordell, and CA Simenstad. 1997. Sink or swim? Copepod population maintenance in the Columbia River estuarine turbidity maxima region. Marine Biology 129:309-317. Nagaraj, M. 1992. Combined effects of temperature and salinity on the development of the copepod Eurytemora-affinis. Aquaculture 103:65-71. Nguyen, R. T. and H. R. Harvey. 1994. A rapid micro-scale method for the extraction and analysis of protein in marine samples. Marine Chemistry 45:1-2. North, E. W. and E. D. Houde. 2001. Retention of White Perch and Striped Bass Larvae: Biological-Physical Interactions in Chesapeake Bay Estuarine Turbidity Maximum. Estuaries 24(5):756-769. North, E. W. and E. D. Houde. 2003. Linking ETM physics, zooplankton prey, and fish early-life histories to striped bass Morone saxatilis and white perch M. americana recruitment. Marine Ecology Progress Series 260:219-236. Ohman, M. D. 1990. The demographic benefits of diel vertical migration by zooplankton. Ecological Monographs 60(3):257-281. Omori, M. and W. M. Hamner. 1982. Patchy distribution of zooplankton: Behavior, population assessment and sampling problems. Marine Biology. 72(2):193- 200. Parsons, T. R. and J. D. H. Strickland. 1963. Discussion of Spectrophotometric Determination of Marine-Plant Pigments, with Revised Equations for 194 Ascertaining Chlorophylls and Carotenoids. Journal of Marine Research 21(3):155-163. Parsons, T. R., Y. Maita and C. M. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford. 187 pp. Parsons, T. R., Y. Maita, and C. M. Lalli. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, Oxford. 187pp. Peitsch, A., B. Koepcke and N. Bernat. 2000. Long-term investigation of the distribution of Eurytemora affinis (Calanoida; Copepoda) in the Elbe Estuary. Limnologica Jena 30(2):175-182. Pieper, R. E., D. E. McGehee, C. F. Greenlaw and D. V. Holliday. 2001. Acoustically measured seasonal patterns of Zooplankton in the Arabian Sea. Deep Sea Research 48:6-7. Porter, J., D. Deere, M. Hardman, C. Edwards, and R. Pickup. 1997. Go with the flow - use of flow cytometry in environmental microbiology. FEMS Microbiology Ecology 24:93-101. Poulet, S. A. 1983. Factors controlling utilization of non-algal diets by particle- grazing copepods: A review. Oceanologica acta 6:221-234. Purcell, J. E., J. R. White and M. R. Roman. 1994. Predation by gelatinous zooplankton and resource limitation as potential controls of Acartia tonsa copepod populations in Chesapeake Bay. Limnology and Oceanography 39(2):263-278. Purcell, J. E., T. A. Shiganova, M. B. Decker and E. D. Houde. 2001. The ctenophore Mnemiopsis in native and exotic habitats: U.S. estuaries versus the Black Sea basin. Hydrobiologia 451:1-3. Purcell, J. E. and M. B. Decker. 2005. Effects of climate on relative predation by scyphomedusae and ctenophores on copepods in Chesapeake Bay during 1987-2000. Limnology and Oceanography. 50(1):376-387. Redfield, A. C. 1958. The biological control of chemical factors in the environment. American Scientist 46:206-221. Richman, S., D.R. Heinle, and R. Huff. 1977. Grazing by adult estuarine copepods of the Chesapeake Bay. Marine Biology. 42:69-84. Ringelberg, J. 1995. Changes in light intensity and diel vertical migration: A comparison of marine and freshwater environments. Journal of the Marine Biological Association of the United Kingdom 75(1):15-25. 195 Roberts, D., D. Rittschof, D. J. Gerhart, A. R. Schmidt, and L. G. Hill. 1989. Vertical migration of the clam Mercenaria mercenaria (L.) (Mollusca: Bivalvia). Environmental correlates and ecological significance. Journal of Experimental Marine Biology and Ecology 126:271-280. Roddie, B. D., R. J. G. Leakey and A. J. Berry. 1984. Salinity-temperature tolerance and osmoregulation in Eurytemora affinis (Poppe) (Copepoda: Calanoida) in relation to its distribution in the zooplankton of the upper reaches of the Forth Estuary. Journal of Experimental Marine Biology and Ecology 77(2):191-211. Roman, M. R., D. V. Holliday and L. P. Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. Roman, M., Anne Gauzens, WK Rhinehart, and JR White. 1993. Effects of low oxygen waters on Chesapeake Bay zooplankton. Limnology & Oceanography 38(8):1603-1614. Roman, MR, DV Holliday, and LP Sanford. 2001. Temporal and spatial patterns of zooplankton in the Chesapeake Bay turbidity maximum. Marine Ecology Progress Series 213:215-227. Runge, J. A. and J. C. Roff. 2000. The measurement of growth and reproductive rates. Pages 401-444 in R. P. Harris, P.H. Wiebe, J. Lenz, H.R. Skjokdal, and M. Huntley, editors. ICES Zooplankton Methodology Manual. Academic Press, San Diego. Runge, J. A. and S. Plourde. 1996. Fecundity characteristics of Calanus finmarchicus in coastal waters of eastern Canada. Ophelia 44:1-3. Sabatini, M., and T. Kiorboe. 1994. Egg-production, growth and development of the cyclopoid copepod Oithona-similis. Journal of Plankton Research 16:1329- 1351. Sanford, L. P., S. E. Suttles and J. P. Halka. 2001. Reconsidering the physics of the Chesapeake Bay estuarine turbidity maximum. Estuaries 24(5):655-669. Sautour, B. and J. Castel. 1995. Comparative Spring Distribution of Zooplankton in 3 Macrotidal European Estuaries. Hydrobiologia 311(1-3):139-151. Schubel, J. R. 1968. Turbidity maximum of the northern Chesapeake Bay. Science 161:1013-1015. Schubel, J. R. and D. W. Pritchard. 1986. Responses of Upper Chesapeake Bay to Variations in Discharge of the Susquehanna River. Estuaries 9(4A):236-249. 196 Sharp, J. H., C. H. Culberson, and T. M. Church. 1982. The chemistry of the Delaware Estuary. General considerations. Limnology and Oceanography 27:1015-1028. Shoji, J., E. W. North and E. D. Houde. 2005. The feeding ecology of Morone americana larvae in the Chesapeake Bay estuarine turbidity maximum: the influence of physical conditions and prey concentrations. Journal of Fish Biology 66:1328-1341. Simenstad, C. A., C. A. Morgan, J. R. Cordell, and J. A. Baross. 1994. Flux, passive retention, and active residence of zooplankton in Columbia River estuarine turbidity maxima. Olsen & Olsen, Fredensborg, Denmark. Simenstad, C. A., L. F. Small and C. D. McIntire. 1990. Consumption processes and food web structure in the Columbia River estuary. Progress in Oceanography 25(1-4):271-297. Smith, P.K., et al. (1985). Measurement of protein using bicinchoninic acid. Analytical Biochemistry. 150:76-85. Sokal, R. R. and F. J. Rohlf. 1994. Biometry, 3rd Edition. W. H. Freeman: New York. Solorzano, L., and J. H. Sharp. 1980. Determination of total dissolved phosphorus and particulate phosphorus in natural waters. Limnology and Oceanography 25:754-758. Soltanpou-Gargari, A. and S. Wellershaus. 1985. Eurytemora affinis -- one year study of abundance and environmental factors. Veroeff. Inst. Meeresforsch. Bremerh. 20(2):183-198. Steele, J.H. and I.E. Baird (1961) Relations between primary production, chlorophyll nd particulate carbon. Limnology and Oceanography 6:68-78. Steele, J.H. and I.E. Baird (1962) Further relations between primary production, chlorophyll, and particulate carbon. Limnology and Oceanography 7:42-47. Sterner, R. W. 1990. The ratio of nitrogen to phosphorus resupplied by herbivores: Zooplankton and the algal competitive arena. American Naturalist 136:209- 229. Strom, S. L., R. Benner, S. Ziegler, and M. J. Dagg. 1997. Planktonic grazers are a potentially important source of marine dissolved organic carbon. Limnology and Oceanography 42:1364-1374. Tackx, M. L. M., P. J. M. Herman, S. Gasparini, X. Irigoien, R. Billiones and M. H. Daro. 2003. Selective feeding of Eurytemora affinis (Copepoda, Calanoida) in 197 temperate estuaries: model and field observations. Estuarine, Coastal and Shelf Science 56(2):305-311. Tafe, D. J. 1990. Zooplankton and Salinity in the Rufiji River Delta, Tanzania. Hydrobiologia 208(1-2):123-130. Tankersley, R. A., and R. B. Forward, Jr. 1994. Endogenous swimming rhythms in estuarine crab megalopae: Implications for flood-tide transport. Marine biology 118:415-423. Touratier, F., L. Legendre, and A. Vezina. 1999. Model of copepod growth influenced by the food carbon:nitrogen ratio and concentration, under the hypothesis of strict homeostasis. Journal of Plankton Research 21:1111-1132. Uitto, A. 1996. Summertime herbivory of coastal mesozooplankton and metazoan microplankton in the northern Baltic. Marine Ecology-Progress Series 132(1- 3):47-56. Vaupel-Klein, J. and R. E. Weber. 1975. Distribution of Eurytemora affinis (Copepoda: Calanoida) in relation to salinity: field and laboratory observations. Netherlands Journal of Sea Research 9:3-4. Verheye, H. M. and J. G. Field. 1992. Vertical distribution and diel vertical migration of Calanoides carinatus (Kroeyer, 1849) developmental stages in the southern Benguela upwelling region. Journal of Experimental Marine Biology and Ecology 158(1):123-140. Vieira, L., U. Azeiteiro, P. Re, R. Pastorinho, J. C. Marques, and F. Morgado. 2003. Zooplankton distribution in a temperate estuary (Mondego estuary southern arm: Western Portugal). Acta Oecologica-International Journal of Ecology 24:163-173. Viherluoto, M. and M. Viitasalo. 2001. Effect of light on the feeding rates of pelagic and littoral mysid shrimps: a trade-off between feeding success and predation avoidance. Journal of Experimental Marine Biology and Ecology 261(2):237- 244. Viitasalo, M. and M. Rautio. 1998. Zooplanktivory by Praunus flexuosus (Crustacea : Mysidacea): functional responses and prey selection in relation to prey escape responses. Marine Ecology-Progress Series 174:77-87. Vijverberg, J. 1980. Effect of temperature in laboratory studies on development and growth of Cladocera and Copepoda from Tjeukemeer, The Netherlands. Freshwater Biology 10:317-340. 198 Vuorinen, I. 1987. Vertical migration of Eurytemora (Crustacea, Copepoda): A compromise between the risks of predation and decreased fecundity. Journal of Plankton Research 9(6):1037-1046. Walve, J., and U. Larsson. 1999. Carbon, nitrogen and phosphorus stoichiometry of crustacean zooplankton in the Baltic Sea: implications for nutrient recycling. Journal of Plankton Research 21:2309-2321. White, J. R. and M. R. Roman. 1992. Egg production by the calanoid copepod Acartia tonsa in the mesohaline Chesapeake Bay: The importance of food resources and temperature. Marine Ecology Progress Series 86(3):239-249. White, J. R. and M. R. Roman. 1992. Seasonal Study of Grazing By Metazoan Zooplankton in the Mesohaline Chesapeake Bay. Marine Ecology-Progress Series 86(3):251-261. Williamson, C. E. and N. M. Butler. 1987. Temperature, food and mate limitation of copepod reproductive rates: Separating the effects of multiple hypotheses. Journal of Plankton Research 9(5):821-836. Winkler, G. and W. Greve. 2004. Trophodynamics of two interacting species of estuarine mysids, Praunus flexuosus and Neomysis integer, and their predation on the calanoid copepod Eurytemora affinis. Journal of Experimental Marine Biology and Ecology 308(1):127-146. Winkler, G., J. J. Dodson, N. Bertrand, D. Thivierge and W. F. Vincent. 2003. Trophic coupling across the St. Lawrence River estuarine transition zone. Marine Ecology Progress Series 251(59-73. Wofsy, S. C. 1983. A simple model to predict extinction coefficients and phytoplankton biomass in eutrophic waters. Limnology and Oceanography 28:1144-1155. Zeng, C., P. Abello and E. Naylor. 1999. Endogenous tidal and semilunar moulting rhythms in early juvenile shore crabs Carcinus maenas: Implications for adaptation to a high intertidal habitat. Marine Ecology Progress Series 191:257-266. Zimmerman, R., R. Gibson and J. Harrinton. 1979. Herbivory and detritivory among gammaridean amphipods from a Florida seagrass community. Marine Biology 54(1):41-47.