ABSTRACT 
 
 
Title of dissertation:  CONTAMINANT EXPOSURE, FOOD WEB TRANSFER 
AND POTENTIAL EFFECTS ON OSPREYS (Pandion 
haliaetus) IN CHESAPEAKE BAY 
 
Rebecca Saul Lazarus, Doctor of Philosophy, 2015 
 
Dissertation directed by: Professor Mary Ann Ottinger and 
Adjunct Professor Barnett A. Rattner  
    Department of Animal and Avian Sciences 
    Marine-Estuarine Environmental Sciences Program 
 
The last large-scale ecotoxicological study of ospreys (Pandion haliaetus) in 
Chesapeake Bay was conducted in 2000-2001 and focused on the U.S. Environmental 
Protection Agency designated Regions of Concern (ROCs; Baltimore Harbor/Patapsco, 
Anacostia/middle Potomac, and Elizabeth Rivers).  From 2011-2013, ROCs, 
Susquehanna River and flats, James and Back Rivers were evaluated to determine spatial 
and temporal trends in osprey productivity and contaminants in eggs.  Concentrations of 
p,p’-DDE were below the threshold associated with eggshell thinning.  Total PCB 
concentrations in eggs from the Anacostia/middle Potomac were lower in 2011 than 
2000, but remained unchanged in Baltimore Harbor.  Polybrominated diphenyl ether 
flame retardants declined across study sites and five alternative brominated flame 
retardants were detected at low levels in osprey eggs.  Concentrations of oxidized DNA 
(biomarker of oxidative stress) were slightly elevated on the Anacostia/middle Potomac 
and Baltimore Harbor/Patapsco Rivers, but no univariate contaminant predictors 
correlated with DNA damage.  Pharmaceuticals and personal care products were also 
examined.  An integrative modeling approach was used to evaluate bioaccumulation 
potential of pharmaceuticals using hypothetical screening-level exposure scenarios.  A 
  
first-order kinetic exposure model was applied to estimate the average daily and 
cumulative 45-day dose of pharmaceuticals received by a nestling osprey.  To 
complement the exposure model, water, fish and osprey nestling plasma samples were 
analyzed for 23 pharmaceuticals and an artificial sweetener (sucralose).  Of the 18 
analytes detected in water, 8 were found in fish plasma, and 1 in osprey nestling plasma 
(antihypertensive diltiazem).  Diltiazem was detected at concentrations approximately 
21.6 times greater in fish plasma than water and 4 times greater in osprey nestling plasma 
than fish.  Diltiazem was found in all 69 osprey plasma samples (540–8,630 ng/L), with 
41% of these samples exceeding maximum concentrations found in fish.  Diltiazem 
levels in fish and osprey were below the human therapeutic plasma concentration 
(30,000 ng/L). Effect thresholds for diltiazem are unknown in ospreys at this time, and 
there was no evidence to suggest adverse effects.  Overall, findings document continued 
recovery of the osprey population, declining levels of select persistent halogenated 
compounds, and modest evidence of oxidative genetic damage in nestlings.   
 
 
 
 
 
 
   
 
 
 
 
  
 
 
 
CONTAMINANT EXPOSURE, FOOD WEB TRANSFER AND POTENTIAL 
EFFECTS ON OSPREYS (Pandion haliaetus) IN CHESAPEAKE BAY 
 
By 
Rebecca Saul Lazarus 
 
 
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 
2015 
 
 
 
 
 
 
Advisory Committee: 
Professor Mary Ann Ottinger, Chair/Co-advisor 
Adjunct Professor Barnett A. Rattner, Co-advisor 
Professor Bryan D. Watts 
Dr. Beth L. McGee 
Professor William W. Bowerman, Deans Representative 
 
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
© Copyright by  
Rebecca S. Lazarus  
2015 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ii 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Dedicated to my Mother, Father and Friends 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
iii 
 
 
ACKNOWLEDGEMENTS 
 
I would like to thank my major advisors and my doctoral committee.  Drs. Barnett 
Rattner and Mary Ann Ottinger, thank you for providing the opportunity to pursue this 
once in a lifetime project and your confidence and support in both the field and the lab.  
Thanks to the U.S. Geological Survey Patuxent Wildlife Research Center for providing 
the technical and administrative support for my project and especially to the U.S. 
Geological Survey Chesapeake Bay Program for the financial support for this study.   
Thank you to Peter McGowan, Chris Guy and Robbie Callahan of the U.S. Fish 
and Wildlife Service Chesapeake Bay Field Office for providing outstanding support and 
prioritizing my fieldwork over the years.  Thank you to Dr. Natalie Karouna-Renier, 
Sandy Schultz and Catherine Maddox for your assistance both in the lab and the office.  
Thank you to Dr. Bryan Brooks, Dr. Bowen Du and the Brooks Lab at Baylor University 
for providing your scientific and intellectual support towards this collaborative project.  
Thanks to Dan Day for providing field support, Dr. Richard Erickson for statistical 
support and to Drs. Chuck Henny and Bryan Watts for providing valuable comments on 
drafts of manuscripts.  I would also like to thank Dave Hopler and Catherine Viverette of 
the Virginia Commonwealth University for their field support on the James River.  
Thank you to all my family and friends who have been there to support me every 
step of the way.  Finally, a special thanks to my horse Raven who was there for me every 
day when I needed her most.  
 
 
 
 
 
 
 
iv 
 
TABLE OF CONTENTS 
 
Dedication                              ii 
Acknowledgements                                                  iii 
List of Tables and Figures                iv 
List of Abbreviations                vii 
 
Chapter 1: Introduction                  1 
The osprey as a sentinel                 2 
Exposure of ospreys to legacy and emerging contaminants and health effects       4 
Transfer of contaminants in the water-fish-osprey food web           13 
Oxidative DNA damage               13 
Chesapeake Bay regional waterways of concern            16 
Purpose of this research and scientific objectives            24  
 
Chapter 2: Decadal re-evaluation of contaminant exposure and productivity in ospreys                       
                 (Pandion haliaetus) nesting in Chesapeake Bay Regions of Concern          29    
Introduction                                    30 
 Materials and Methods               31 
 Results                 42 
 Discussion                 55 
Chapter 3: Chesapeake Bay fish-osprey (Pandion haliaetus) food web: evaluation of              
                  contaminant exposure and genetic damage            65
 Introduction                  66 
 Materials and Methods               68 
 Results                 79 
 Discussion                 93 
 
Chapter 4: Exposure and food web transfer of pharmaceuticals in ospreys (Pandion 
haliaetus): predictive model and empirical data          104 
 Introduction               105 
 Materials and Methods             107 
 Results               117 
 Discussion                          122 
 
Chapter 5: Overall Discussion             134 
 Summary of Findings              135 
 In Silico Tools for Assessment            140 
Examination of an Effects Hierarchy            141 
Utility of these Data              142 
Final Conclusion                                    143 
 
Compiled References                           144 
 
 
 
v 
 
LIST OF TABLES AND FIGURES 
Chapter 1 
Figure 1. Locations of Regions of Concern and study sites in Chesapeake Bay.  
Figure 2. A schematic of food web transfer matrices and overview of the major
 groups of contaminants quantified. 
 
Chapter 2 
Figure 3. Locations of sampled osprey nests in Chesapeake Bay Regions of 
Concern and Poplar Island reference site, (•) indicates a sampled nest. 
 
Figure 4. Temporal comparison of contaminants from Chesapeake Bay Regions 
of Concern in 2000-2001 (adapted from Rattner et al. 2004) and 2011-2012. 
Geometric means and individual data points are presented (•).  At each site 
historical and current datasets were compared by a t-test (α=0.05).    
 
Figure 5. Concentrations of 8-hydroxy-2’-deoxyguanosine (8-OH-dG) in osprey 
nestling whole blood samples.  Means, standard deviations and individual data 
points are presented (•).  Crosses (†) indicate a marginally significant difference 
(α=0.05-0.09) using Tukey’s HSD test. 
 
Table 1. Reproductive success of osprey nesting in Chesapeake Bay Regions of 
Concern. 
 
Table 2. Concentrations (µg/g ww) of organochlorine pesticides and metabolites 
from Chesapeake Bay Regions of Concern and Poplar Island reference sitea. 
 
Table 3. Wet weight concentrations of polychlroinated biphenyls from 
Chesapeake Bay Regions of Concern and Poplar Island reference sitea. 
 
Table 4. Concentrations (ng/g ww) of polybrominated diphenyl ethers, additional 
brominated flame retardants and methoxytriclosan from Chesapeake Bay Regions 
of Concern and Poplar Island reference sitea. 
Supplemental Table S1.  Concentrations of p,p’-DDE, total PCBs and PBDEs in 
archived eggs analyzed in 2000-2001 and re-analyzed in 2011-2012. 
 
 
 
 
vi 
 
Chapter 3 
Figure 6.  Locations of osprey nests sampled from in Chesapeake Bay and Poplar 
Island reference site, (•) indicates a sampled nest.  
 
Figure 7.  Concentrations of 8-hydroxy-2’-doxyguanosine (8-OH-dG) in osprey 
nesting whole blood. Capital letters indicate a statistically significant difference 
(p<0.05) using Tukey’s HSD test.  Means, standard deviations and individual 
values are presented (•).Data from 2011 and 2012/2013 were analyzed separately. 
 
Figure 8. Redundancy analysis (RDA) biplot of dna damage, PCBs, PBDEs, 
DDE and carbon and nitrogen stable isotopes and their relationships across study 
sites. Data are projected onto the ordination axes.  The large dots are the centroids 
and the shaded areas represent the 95% confidence ellipses.  The vectors (black 
lines) represent the contribution of different variables to each axis. 
Table 5. Reproductive success of osprey nesting in Chesapeake Bay regional
 waterways of concern. 
 
Table 6. Osprey egg concentrations (µg/g ww) of organochlorine pesticides and 
metabolites from Chesapeake Bay regional waterways and Poplar Island reference 
sitea. 
 
Table 7. Osprey egg concentrations (µg/g ww) of polychlorinated biphenyls and 
congeners from Chesapeake Bay regional waterways and Poplar Island reference 
sitea. 
 
Table 8. Osprey egg concentrations (ng/g ww) of polybrominated diphenyl ethers
 and alternative brominated flame retardants from Chesapeake Bay regional
 waterways and Poplar Island reference sitea. 
 
Table 9. Biomagnification factors (BMFs) on a wet weight (ww) and lipid weight 
 (lw) from whole fish to osprey egg contents by study site. 
 
Supplemental Figure S1. Linear contribution (eigenvalues) of predictor variables
 to each axis. 
 
Supplemental Table S2. Osprey diet composition from each study site. 
 
Supplemental Table S3. Contaminant residues in whole fish collected across all
 study sitesa,b. 
 
Supplemental Table S4. Amount of variance explained by each axis as well as
 the proportion and cumulate proposition of variance explained. 
 
 
 
vii 
 
Chapter 4 
 
Figure 9. Theoretical screening-level exposure assessment framework used to 
model the daily intake (DI) for an osprey. 
 
Figure 10. Map detailing the locations of Chesapeake Bay study sites. 
 
Table 10. Daily intake (DI) at three water concentrations and t1/2 to accumulate a 
human therapeutic dose (HTD) within 40 days for 16 pharmaceuticals compared 
to human values. 
 
Table 11. Summary of compounds detected in water and fish across study sites. 
 
Table 12. Diltiazem concentrations in water, fish and osprey on ng/L basis. 
 
Supplemental Table S5. List of 114 pharmaceutical compounds and sucralose 
for the screening-level exposure assessment, their CAS number, and 
bioconcentration factor (BCF). 
 
Supplemental Table S6. Analytes quantified in water and plasma from fish and 
ospreys and their method detection limits (MDL). 
 
Supplemental Table S7. Analytes detected in water samples. 
 
Supplemental Table S8. Concentrations of pharmaceuticals in fish plasma at 
each study site.  
 
Chapter 5 
Figure 11. Gradient illustrating the hierarchy of effects and when 
macromolecular effects may be meaningful. 
 
 
 
 
 
 
 
 
viii 
 
LIST OF ABBREVIATIONS 
 
8-OHdG = 8-hydroxy-2’-deoxyguanosine 
BCF = bioconcentration factor 
BMF = biomagnification factor  
BWt = body weight 
CAFO = concentrated animal feeding operations 
DI = daily intake 
DNA = deoxyribonucleic acid  
dw = dry weight 
e = 2.71 
FIR = food intake rate 
GSSG = glutathione disulfide (reduced glutathione) 
GSH = glutathione (oxidized glutathione)  
HTD = human therapeutic dose 
k = half-life elimination constant  
LOAEL = lowest-observed-adverse-effect-level 
lw = lipid weight 
MDA = malondialdehyde 
MDL = method detection limit 
MDG= million gallons per day 
PPCB = pharmaceuticals and personal care product 
POCIS = polar organic chemical integrative samplers 
PUFA = polyunsaturated fatty acids 
ROC = region of concern  
ROS = reactive oxygen species 
t1/2 = half-life  
TBARS = thiobarbituric acid reactive substances 
TEF= toxic equivalent factor 
TEQ = toxic equivalent 
ww = wet weight  
WQLS = water quality limited segment 
WWTP = wastewater treatment plant 
US EPA = United States Environmental Protection Agency 
 
 
 
 
1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
CHAPTER 1 
 
INTRODUCTION 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
2 
The Osprey as a Sentinel  
The Chesapeake Bay is the largest estuary in the United States and supports a 
diversity of avian species.  At the same time, increasing human population and related 
activities in this region have resulted in deterioration of both water and habitat quality 
(Ernst, 2003; Phillips, 2007).  Degradation of habitat quality from agricultural, industrial 
and urban pollution continues to impact the Chesapeake ecosystem, threatening sensitive 
wildlife species in vulnerable portions of the estuary, and adversely affecting fish and 
wildlife (e.g., Blazer et al., 2007; Rattner and McGowan, 2007; Iwanowicz et al., 2009).   
Ospreys (Pandion haliaetus) have often served as sentinels of ecosystem health 
(Furness et al., 1993; Grove et al., 2009).  The Chesapeake Bay is home to the largest 
nesting osprey population in the world (Poole, 1989; Watts et al., 2004; Watts and 
Paxton, 2007).  This population has been extensively studied in the Bay since the 1970’s 
(e.g., Wiemeyer, 1971; Henny et al., 1974; Wiemeyer et al., 1975; Watts et al., 2004).  
Golden and Rattner (2003) developed a system for ranking higher order terrestrial 
vertebrate species for their utility as a bioindicator and their vulnerability to 
environmental contaminants.  Out of 25 vertebrate species, ospreys ranked third for their 
utility in monitoring persistent organic pollutants and fourth for their vulnerability to this 
group of compounds.  Previous studies have used the osprey to monitor exposure and 
potential adverse effects of toxics on wildlife, document spatial and temporal 
ecotoxicological trends, and understand bioaccumulation and biomagnification of 
contaminants (e.g., Wiemeyer et al., 1975; Elliot et al., 1998, 2000, 2001; Henny et al., 
2003, 2009, 2010; Rattner et al., 2004, 2008).   
Ospreys provide an excellent index of ecohealth for their environment.  Grove 
and colleagues describe ospreys as a species that are tolerant to short-term disturbance for 
3 
sample collection, opportunistically nest on artificial structures, and are adaptable to 
human landscapes.  Pairs have nest-site fidelity and nests can be selected randomly for 
large-scale studies to document spatial patterns in contamination (Toschick et al., 2005).  
Nests are conspicuous and easily accessed on navigation markers, platforms, and other 
structures found in large bodies of water.  Ospreys are strictly piscivorous and feed at a 
high trophic level, which makes them ideal for monitoring bioaccumulation and 
biomagnification of lipophilic organic contaminants in the food web.  Although ospreys 
are long-lived, no relationship has been detected between age and body burdens of 
organochlorine contaminants (Ewins et al., 1999; Elliott et al., 2007).  Taken together, the 
osprey is an ideal species for large studies to address questions of climate, ecotoxicology 
risk and population monitoring.  
Contaminant accumulation in prey items directly relates to spatial variation in 
exposures at the nesting grounds.  Ospreys typically spend one month on the breeding 
grounds before laying eggs. Lipid soluble compounds may accumulate in prey items, 
where the foraging adult female deposits ingested contaminants into the lipid rich eggs, 
thereby exposing the highly sensitive developing embryo.  Nestlings are exclusively fed 
fish that the parents capture generally within 1-5 km of the nest site; however, locations 
with limited nestling nests may require birds to forage up to 15 km (Poole, 1989).  Elliot 
and coworkers (1998) reported that polychlorinated dibenzo-p-dioxins (PCDDs) 
accumulated at several orders of magnitude higher in eggs downstream of pulp and paper 
mills compared to upstream nests, reflecting patterns in localized contaminant situations. 
Other less lipophilic contaminants (e.g., metals) are also readily detected in the blood and 
nestling’s feathers (Rattner et al., 2008).   
4 
Exposure of Ospreys to Legacy and Emerging Contaminants and Health Effects 
 
Legacy contaminants 
Ospreys are sensitive to the effects of persistent lipophilic organic contaminants; 
most notably the metabolites of the organochlorine pesticide DDT (dichlorodiphenyl-
trichloroethane).  Exposure to the metabolite of DDT (p,p’-DDE, 
dichlorodiphenyldichloroethylene) resulted in population declines of osprey and other 
raptors during the 1960-70’s due to eggshell thinning (Anderson and Hickey 1972; 
Wiemeyer et al., 1975, 1988).  Following the ban of DDT in 1972, raptor populations 
across the Northeastern U.S. have shown significant recovery (e.g., Watts et al., 2004). 
 Even though concentrations of DDT, its metabolites and other organochlorine 
pesticides have declined in response to increased regulations, polychlorinated biphenyl 
(PCB) concentrations continued to remain elevated in the Chesapeake Bay (Wiemeyer et 
al., 1975, 1988; Rattner et al., 2004).  Rattner and colleagues (2004), collected osprey egg 
samples from Chesapeake Bay Regions of Concern (ROCs) and found greater total PCB 
concentrations in osprey eggs collected in 2000 from Baltimore Harbor/Patapsco River 
and the Anacostia and Potomac Rivers (7.25 and 9.28 μg/g wet weigh, ww respectively), 
compared to the South River near Annapolis, MD (4.91 μg/g ww).  These findings were 
similar to the 1972 data collected by Wiemeyer and colleagues (1975), who recorded 
mean PCB concentrations of 7.81μg/g ww, from randomly collected addled eggs along 
the Potomac River in Maryland.  Similarly, black-crown night heron (Nycticorax 
nycticorax) eggs collected from the Fort Carroll colony in Baltimore Harbor (Rattner et 
al. 2001), were compared to eggs collected 7 years earlier at the Baltimore Gas & Electric 
Colony (Rattner et al., 1993). PCB concentrations decreased only slightly (3.83 µg/g ww 
5 
in 1991 vs. 3.40 µg/g ww in 1998). Interestingly, the magnitude of the most toxic 
coplanar congeners did exhibit a decline over time.    
PCBs continue to be the principal driver of human fish consumption advisories 
for a large portion of the watershed including the ROCs.  Fish species listed under 
consumption advisories (e.g., carp, Cyprinus carpio and catfish, Ictaluridae sp.) are 
consumed by high trophic level piscivorous waterbirds including ospreys and bald eagles 
(Haliaeetus leucocephalus) (MDE, 2014; VDH, 2013). As such, the presence of PCBs is 
likely to impact piscivorous colonial waterbird populations (Tillit and Giesy, 2012). 
Furthermore, gaining an understanding of the health effects of contaminants on wildlife 
populations has important implications for human health and is critical information for 
natural resource managers given the close interlocking relationship between ecosystem 
and human health (Harris and Jones, 2008).  
PCBs are of concern for avian species, given a number of different health effects 
shown in both laboratory and field studies (Kubiak et al., 1989; Tillitt and Giesy, 2012).  
The toxicity of PCBs depends on the specific suite congeners in the mixture (Toxic 
Equivalency Quotient (TEQ) quantifying the potencies of PCB congeners to 2,3,7,8-
tetrachlorodibenzo-para-dioxin, TCDD), and their induction patterns of EROD in 
hepatocyte cultures from a suite of species.  Only a small subset of the most potent PCBs 
(77, 89, 126 and 169) can be directly embryotoxic (Kennedy et al., 1996).  Hoffman and 
coworkers (1998) orally dosed kestrel nestlings with PCB 126 at 50, 250, and 1000 ng/g 
body weight.  At the higher dosed groups, they documented decreased bone growth, 
thyroid lesions, increased EROD activity, endocrine, and immunological effects in the 
nestlings. However, if exposed to other congeners that may not be directly embryotoxic, 
6 
other subtle endocrine disruption effects (i.e., thyroid, cardiovascular or lipid impacts) 
may occur. In a similar manner, Fernie and coworkers (2003) reported developmental, 
behavioral, and endocrine-thyroid effects in maternally exposed American kestrel (Falco 
sparverius) hatchlings. Cardiovascular developmental effects have been reported in 
laboratory studies in embryonic Japanese quail (Coturnix japonica) and chickens (Gallus 
gallus domesticus) exposed increasing doses of an air-cell injected PCB congener 
mixture (Carro et al. 2013a).  Specifically, heart abnormalities occurred across all 
treatments with elongation and expansion of heart tube, improper looping and 
indentations in the emerging ventricular wall during embryo development (and even 
ventricular thinning, dilation and absence of the trabeculated layer; Carro et al., 2013a, 
2013b).  
 
Polybrominated diphenyl ethers 
Polybrominated diphenyl ether (PBDE) flame retardants have been detected in the 
eggs of avian species in both the Chesapeake and other estuaries worldwide (Chen and 
Hale, 2010). These compounds are additive (i.e., covalently bound) and chemicals can 
leach out into the environment and pose a hazard to wildlife.  Despite the recent 
discontinuation of manufacture and use of the penta-BDE and octa-BDE formations in 
the U.S., and an agreement to phase out BDE-209 (the deca- formulation) by the end of 
2012, PBDEs will continue to enter the environment from in-service and discarded 
products. 
Rattner and colleagues (2004) measured eight PBDE congeners (BDE 28, 47, 49, 
99, 100, 153, 154 and 183) in osprey eggs collected in 2000-2001. The Anacostia/middle 
7 
Potomac River study sites had significantly higher total PBDE concentrations (geometric 
mean 0.725 μg/g ww) compared to the South River (geometric mean 0.176 μg/g ww).  
The BDE-47 congener accounted for 65% (0.114 μg/g out of 0.176 μg/g) of the total 
found in the South River and 66% (0.480 μg/g out of 0.725 μg/g) in Baltimore 
Harbor/Patapsco River.  On a spatial scale, high PBDE levels in bird eggs have been 
measured in industrialized regions and large urbanized compared to remote rural areas 
(Norstrom et al., 2002; Chen et al., 2008; Potter et al., 2009). Similar patterns in congener 
accumulation have been described for several additional piscivorous waterbird species, 
including the brown pelicans (Pelecanus occidentalis), and double-crested cormorants 
(Phalacrocorax auritus).  
Total PBDE concentrations in herring gull (Larus argentatus) eggs in the Great 
Lakes increased exponentially over 17 years (1983-2000), with an estimated doubling 
time of 2.2-3.1 years (Norstrom et al., 2002). Park and colleagues (2009) found levels of 
PBDEs in California peregrine falcon (Falco peregrinus) eggs tripled with each decade 
from 1986-2007. This suggests that, PBDE levels could be increasing in Chesapeake Bay 
waterbird populations.  In contrast, others have reported that PBDE concentrations are 
beginning to level off.  Gauthier and coworkers (2008) examined data from herring gull 
eggs in the Great Lakes and found residues of PBDEs in eggs were beginning to level off 
post-2000.  
In order to interpret the potential effects of PBDEs found in Chesapeake Bay 
osprey eggs, McKernan and coworkers (2009) conducted an egg injection study in which 
a suite of endpoints were examined across several different avian species including 
chickens, mallards (Anas platyrhynchos), and American kestrels.  American kestrels were 
8 
found to be the most sensitive, with decreased pipping and hatching success reported at 
egg concentrations of 10 and 20 μg DE-71/g egg.  Measurement of the amount of the 
injected PBDE that entered the egg contents from the injection site in the air cell 
indicated a lowest-observed-adverse-effect-level (LOAEL) of 1.8 µg/g ww (~32 µg/g 
lipid weight, lw; McKernan et al., 2009, 2010).  Other studies also documented increased 
oxidative stress, delay in time to hatch, altered glutathione levels, changes in Vitamin A, 
and decreased thyroxine concentrations in nestling kestrels as well as altered courtship 
behavior in adults (Fernie et al., 2005, 2006, 2008; Rattner et al., 2013).  Comparative 
toxicity studies indicated that mallard embryos and common tern (Sterna hirundo) appear 
relatively insensitive to PBDEs (McKernan et al., 2009; Rattner et al., 2013). 
The LOAEL of 1.8 µg DE-71/g ww was approached by PBDE concentrations in 
eggs of several wild raptor species (Rattner et al., 2004, Chen et al., 2008, She et al., 
2008, Henny et al., 2009).  Henny and colleagues (2009) documented reduced 
reproductive performance in osprey pairs exposed to PBDEs.  The number of young per 
occupied nest was reduced to 0.83 per nest with eggs containing over 1.0 μg/g PBDE ww 
compared to 1.43 young/occupied nest for eggs containing <250 ng/g PBDE ww.  
Although similar to the LOAEL reported by McKernan and coworkers, a more recent 
study (Henny et al. 2011), suggests ospreys can tolerate higher levels of PBDEs.  Birds 
laying eggs containing >1.0 µg total PBDE/g ww did not show a significant relationship 
between PBDEs and reproductive success.  Henny and colleagues (2011) suggested that 
additional monitoring is warranted to understand effects of PBDEs on reproductive 
success.  There is no current evidence for large-scale population level effects from 
9 
PBDEs in wild birds (Chen and Hale, 2010) as have been documented with other 
contaminants, including DDT, methyl mercury and PCBs.   
 
Additional brominated flame retardants 
With the phase out of traditional PBDEs, the use of alternative brominated flame 
retardants (Alt-BFRs) has increased.  These compounds include, α, β and γ hexabromo-
cyclododecane (HBCD), 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE), di(2-
ethylhexyl)-2,3,4,5-tetrabromophthalate (TBPH), 2-ethylhexyl 2,3,4,5-tetrabromo-
benzoate (TBB) and decabromodiphenyl ether (DBDPE).  As with PBDEs, these BFRs 
are additive flame retardants that can leach into the environment and have been detected 
in many matrices including sewage (La Guardia et al., 2010).  Measurable concentrations 
are also found in breast milk, food and household furniture (Roosens et al., 2009; 
Stapleton et al., 2009) as well as in peregrine falcon eggs (Guerra et al., 2012).  HBCD is 
produced in high volumes for use in polystyrene foams, upholstered furniture and 
textiles, and electrical appliances (US EPA, 2010).  HBCD ranks third in production to 
TBBPA and deca-BDE with 16,700 metric tons produced per year (Morose, 2006).   
Among the replacement flame-retardants, the commercial HBCD formulation was 
tested for effects in reproductively active American kestrels (e.g., Marteinson et al., 2011, 
2012).  Birds received a dietary dose of 0.51 µg HBCD per g kestrel for 75 days; 
resulting in the dosed birds having enlarged testes, elongated spermatids, a moderate 
increase in plasma testosterone, reduced free and total T4, and reductions in courtship 
behaviors of male and female dosed kestrels compared to the controls.  Other flame 
retardants including BTBPE (a replacement for octa-BDE) and DBDPE have been 
10 
detected in eggs from glaucous gulls (Larus hyperboreus) in the Arctic (Verreault et al., 
2007) and in some peregrine falcon eggs (Guerra et al., 2012).  
 
Metals 
Many metals are found in the Chesapeake Bay region, often below the thresholds 
for adverse effects for waterbird populations.  These levels of metals have not been 
associated with population declines in ospreys and bald eagles (Rattner and McGowan, 
2007; Rattner et al., 2008). Golden and coworkers (2003) found higher concentrations of 
lead (Pb) in the feathers of black-crowned night-herons in samples from Baltimore 
Harbor compared to herons nesting at a remote island site (Holland Island, Maryland) in 
Chesapeake Bay.  Nestling ospreys showed a similar pattern in their feather 
concentrations of Pb (Rattner et al., 2008).   Interestingly, Pb concentrations were slightly 
elevated in Baltimore Harbor, compared to the South, West and Rhode River reference 
sites; whereas significantly greater concentrations in blood and feathers occurred in the 
Elizabeth River.  
Cadmium (Cd) is another metal of potential concern as it has been detected in 
livers of waterfowl including mute swans (Cygnus olor) and long-tailed ducks (Clangula 
hyemalis) at concentrations up to 94 μg/g and 81 μg/g dry weight respectively (Beyer et 
al., 1998). However, the threshold of toxicity for Cd has not been established for 
waterbirds; additional information would be needed to assess any adverse effects. Total 
Maximum Daily Loads (TMDLs) have been established for metals, but have yet to be 
related to effects on wildlife.  In summary, heavy metals (e.g., Cd, Hg, Pb) have been 
detected in osprey nestling blood in ROCs; however, concentrations were well below 
11 
toxicity thresholds and, as such, did not adversely affect nestling survival, and therefore 
may not be a primary concern for fish eating birds.  
 
 
Pharmaceuticals and personal care products 
 
The application of newer analytical technologies that allow the detection of 
picogram quantities of contaminants has revealed the presence of over 600 
pharmaceuticals and personal care products (PPCPs) compounds in a variety of 
environmental matrices including sediments, sewage sludge, biosolids, drinking water 
and fish tissue (Brooks et al., 2005; Alavarez et al., 2009; Iwanowicz et al., 2009; 
Ramirez et al., 2009; Küster and Adler, 2014). The term “PPCP” encompasses veterinary, 
over the counter and illicit human drugs, and personal care products including cosmetics, 
food supplements and fragrances, as well as their respective metabolites (Daughton and 
Ternes, 1999; US EPA, 2009). With this in mind, it is quite possible that these 
compounds have been continuously entering the environment for many years without 
notice via somewhat unregulated sources (e.g., WWTP effluent, improper disposal of 
expired drugs, flooding events and veterinary drugs in concentrated animal feeding 
operation (CAFO) run-offs) resulting in potential chronic exposure and accumulation 
across a range of organisms (Nikolaou et al., 2007). Exposure to a mixture of PPCPs 
could result in synergistic impacts, potentially compromising the health of fish and 
wildlife (Bean et al., 2014).  
A recent study investigating the occurrence of PPCPs in fish and liver samples 
from five streams near sewage treatment plant effluent discharge sites revealed 
measurable concentrations of antidepressants including fluoxetine (Prozac®), its 
12 
metabolite norfluoxetine, and sertraline (Zoloft®) at 3.2-4.0 ng/g in fish and 19-70 ng/g in 
liver, respectively (Ramirez et al., 2009). Other pharmaceuticals detected in fish fillet and 
liver samples included the antihistamine diphenhydramine used in Benadryl®, the 
antihypertension medication diltiazem, the antiseizure medication carbamazepine and the 
anitilipemic gemfibrozil (Ramirez et al., 2009).  Although pharmaceuticals have been 
detected in water bodies at low levels, Kolpin and coworkers (2002) suggest that these 
compounds are not removed by traditional wastewater treatment and are resistant to 
degradation, with the potential of degrading into more persistent compounds.   
Of particular concern is exposure of non-target organisms to these 
pharmaceuticals with unknown effects from low-level chronic exposure.  Much of the 
focus on aquatic organisms (e.g., Brodin et al., 2014; Du et al., 2014), has recently shifted 
to wildlife species and bioaccumulation across trophic levels in higher order vertebrates 
(Arnold et al., 2013; Shore et al., 2014).   Current questions include the behavioral effects 
of these pharmaceuticals in birds.  Most well-known and widely studied is the case study 
of Asian vultures that were poisoned after feeding on carcasses of cattle that had been fed 
diclofenac (Oaks and Watson, 2011).  This is the only well-known instance for a 
veterinary pharmaceutical having a population effect on wildlife.  Little is known about 
the occurrence of these compounds in wildlife in the Chesapeake Bay.  Their presence, 
adverse effect levels and potential for food web transfer in wildlife have yet to be 
established. 
 
 
 
 
 
 
13 
Transfer of Contaminants in the Water-Fish-Osprey Food Web 
 
Ospreys are unique in that they feed at a high trophic level, and represent the fate 
of bioaccumulated and biomagnified persistent lipophilic organic contaminants in the 
aquatic ecosystem.  Biomagnification studies integrate the source of prey (fish) with the 
predator (osprey) and allow us to follow a contaminant (or its metabolites) though the 
food web.  By identifying individual prey species specific to salinity and geographic 
location, we can closely reconstruct the fish-osprey food web.  Biomagnification factors 
(BMFs) can help predict exposure of high risk or endangered species to contaminants 
(Henny et al., 2003).  Henny and coworkers calculated BMFs from whole fish to osprey 
eggs in 1993 and 2001 (Henny et al., 2003, 2009) using the fish osprey food web on the 
Columbia River (total PCBs BMF: 12-13 lw, organochlorine pesticides (e.g., p, p’- DDE 
BMF: 112-113 lw), and mercury (Hg) BMF: 0.60 dw).  BMFs were also calculated for 
PCDDs (e.g., BMF: 20 lw) and polychlorinated dibenzofurans (PCDFs) (e.g., BMF: 18 
lw).  In 2009, Chen and coworkers (2009) calculated BMFs for total PCB congeners 
(BMF 23.9 lw), total PBDEs (BMF: 25.1 lw) and p,p’-DDE (BMF: 18 lw) in the James 
River fish-osprey food web. Chen and coworkers (2010) noted that a BMF greater than 5 
indicates that there is a potential for the contaminant to biomagnify up the food web.    
 
Oxidative DNA Damage 
In the toxicological literature, there are several prominent classes of xenobiotics 
that increase reactive oxygen species (ROS) in cells.  These compounds have damaging 
effects, impair cellular defense mechanisms (Mitchelmore and Chipman, 1998) and 
exacerbate the production of free oxyradicals.  Electrophilic oxyradicals can bind to 
nucleophilic sites on molecules including DNA, lipids and proteins leading to production 
14 
of harmful products including intermediates that can bind to the DNA to form adducts. 
Polyunsaturated fatty acids (PUFAs) are most susceptible to ROS since H atoms can 
easily be abstracted from C-H bonds.  As with redox cycling compounds, attack of lipids 
by ROS initiates a chain reaction. As the lipid hydroperoxides (polyunsaturated fatty 
acids) break down, they form malondialdehyde (MDA) (Hoffman et al., 2011), which is 
carcinogenic in its own right.  It can react with the DNA bases deoxyguanosine and 
deoxyadenosine (Marnett, 1999). In Oakes and Van Der Kraak (2003), increases of 
thiobarbituric acid (TBARs) activity was measured in white suckers (Catostomus 
commersoni) residing downstream from pump mill effluent discharge, indicating that 
redox active compounds were present in this toxic chemical mixture.   
Oxidation of guanosine (8-hydroxy-2’-deoxyguanosine, 8-OHdG) can be used as 
a sensitive biomarker of oxidative injury to DNA bases and occurrence of DNA damage.  
In the literature, one widely used assay to detect DNA strand breaks and damage across a 
variety of taxa is the comet assay (i.e., single-cell gel electrophoresis) described by Singh 
and coworkers (1988).  When DNA is damaged, the disruption results in the appearance 
of a “tail” similar to a comet trailing in the gel. The head of the comet represents intact 
and undamaged DNA.  A 1998, a spill releasing toxic mine waste containing high levels 
of heavy metals (Ar, Cu, Pb, Zn) occurred in western Spain.  A series of studies 
conducted by Pastor and coworkers (2004) documented genotoxic effects of metals on 
local bird populations near the spill.  Blood samples from white storks (Ciconia ciconia) 
and black kites (Milvus migrans) collected over four years showed that birds from 
contaminated sites had consistently higher tail moments compared to controls.  An 
increasing pattern of genotoxic damage was observed throughout the four subsequent 
15 
years of monitoring.  High genotoxic damage was also observed in nestlings in follow up 
studies indicating potential multi-generational long-term consequences and possibly 
heritable damage to DNA (Pastor et al., 2004). 
In the Chesapeake Bay watershed, Pinkney and coworkers (2004a) described fish, 
particularly the brown bullhead (Ameiurus nebulosus) living in highly polluted areas (i.e., 
in proximity to combined sewer overflows) on the Anacostia River, as having large skin 
tumors containing high levels of PAHs.  Further DNA adducts were observed from 
isolated DNA from fish livers through a 32P post-labeling assay.  PBDEs have an 
oxidative mechanism of action and evoke mild oxidative stress in American kestrels 
(Fernie et al., 2005), observed as DNA damage in common tern and kestrel hatchlings 
(Rattner et al., 2013).  Consistent with other endpoints examined and previous work (e.g., 
McKernan et al., 2009), DE-71 treated kestrel hatchlings had significantly higher levels 
of hepatic concentrations of oxidized glutathione (GSSG) and also 8-OH-dG; however, 
instead of depleting reduced glutathione (GSH), as was observed in other studies (e.g., 
Fernie et al., 2005), a slight increase was observed in GSH (Rattner et al., 2013).  To 
complement the assays for glutathione status, hepatic thiols, which protect against the 
effects of oxidative stress and lipid peroxidation, a manifestation of oxidative damage, 
were measured (Hoffman et al., 2005). Although no statistically significant differences 
were observed, total sulfhydryl, protein bound sulfhydryl, and TBARS concentrations 
were slightly elevated in treated kestrels.  This supports the contention that exposure 
conditions induced mild oxidative stress in kestrels and may evoke similar effects in 
other species.  Other studies have reported DNA damage after exposure to the BDE-47 
16 
and BDE-99 congeners in rats (e.g., He et al., 2008; Albina et al., 2010; Pellacani et al., 
2012) which may lead to epigenetic effects in later populations.  
 
 
Chesapeake Bay Regional Waterways of Concern 
 
Anacostia/middle Potomac River 
 
The Potomac River watershed encompasses four states and drains 37,995 km2.  
Stretches of the Potomac in District of Columbia, VA and MD have segments listed on 
the 303d list as impaired waters listed under the Clean Water Act.  In these listed water 
bodies, the current pollution controls are insufficient to maintain water quality standards 
(US EPA, 2012a, 2012b).  Due to the listing, a TMDL (total maximum daily load) for 
toxics such as PCBs have been established for most of the Anacostia and middle Potomac 
Rivers (Haywood and Buchanan, 2007). The TMDL sets a maximum loading amount for 
PCBs on the river and includes inputs from both point and non-point sources in its 
allocations. On the Potomac, PCBs enter the river through a combination of both point 
and non-point sources, with most of the inputs coming from direct drainage from areas 
adjacent to the river (Haywood and Buchanan, 2007).  Inputs of PCBs can be attributed 
to direct discharge from WWTPs (e.g., Blue Plains) and power plants, as well as non-
point sources flowing in from the Anacostia’s combined sewer overflows, landfills, and 
re-mobilization events where contaminated sediments are transported further down river.  
These latter non-point sources are most difficult to regulate and may offer a continued 
input of sediment bound toxics into the food web.  As the river flows out of the District, 
land use changes to primarily suburban uses (Interstate Commission on the Potomac 
River Basin 1997). Consequences of chemical input can be reflected in fish health in this 
17 
watershed. Reproductive data for fish indicates reports of intersex in male smallmouth 
bass (Micropterus dolomieu) near wastewater treatment effluent point sources (Blazer et 
al., 2007; Iwanowicz et al., 2009).  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Prevalence of tumors in brown bullheads, increased cytochrome P450 activity, 
increased bile PAH metabolites and elevated concentrations of organochlorine 
contaminants have been documented in the tidal portions of the Potomac River, including 
Quantico, and the Anacostia River (Pinkney et al., 2001, 2004a). These effects indicate 
Anacostia/middle 
Potomac Rivers
Patapsco River/
Baltimore Harbor
Back River 
Susquehanna River and Flats
Poplar Island
Elizabeth River 
Upper James River 
Figure 1. Map of study sites in Chesapeake Bay                   
(including Chesapeake Bay Regions of Concern). 
VA 
DE 
MD 
18 
the presence of a suite of redox active contaminants that produce ROS, which affect fish 
populations.  Polar organic chemical integrative samplers (POCIS) detected galaxolide 
(955 ng/POCIS), tonalide (515 ng/POCIS) and other emerging contaminants at the Blue 
Plains and Conococheague downstream study sites (e.g., Iwanowicz et al., 2009).  
Collaborator Vicki Blazer and coworkers have conducted a significant amount of 
research on fish health on the Potomac, particularly near Blue Plains WWTP; however, 
the extent to which the pharmaceuticals are bioaccumulating in the food web and their 
effects on wildlife are unknown.  Legacy contaminants continue to drive fish 
consumption advisories on the Potomac.  Subsequently, the investigation of the transfer 
of compounds in the fish osprey food web and their impact on osprey productivity and 
health is warranted (MDE, 2009).  
Since last reported in 2000-2001, levels of total PBDE congeners in the Anacostia 
and middle Potomac River were approaching observed effect thresholds (Henny et al., 
2009, McKernan et al., 2009).  The highest residues of total PBDEs in osprey eggs on 
this tributary were 928 ng/g ww.  This value approaches the LOAEL for pipping and 
hatching success in American kestrels reported by McKernan and coworkers (2009) of 
1.8 µg/g ww.  As mentioned earlier, other adverse effects of PBDEs in kestrels (the 
laboratory raptor model) include changes in reproductive behavior in adults, increased 
oxidative stress, and decreased thyroxine concentrations (Fernie et al., 2005, 2006, 2008).  
Evidence of increased sensitivity of kestrels to PBDEs has also been documented in 
McKernan et al., 2009, where reduced pipping and hatching success as observed at the 
highest injection dose and in Rattner et al. (2013), in which significant effects were 
19 
reported for increased oxidative stress and DNA damage endpoints in American kestrel 
embryos.  
 
Baltimore Harbor/Patapsco River and Back River 
During the DDT era, there were no ospreys nesting in Baltimore Harbor/Patapsco 
River and few pairs in 2000-2001 (Henny et al., 1974, Rattner et al., 2004).  The United 
States Environmental Protection Agency (US EPA) categorized this tributary as an ROC 
due to poor water quality conditions and elevated concentrations of pesticides, PCBs and 
metals (Chesapeake Bay Program, 1994). With the banning of DDT and other 
compounds, osprey populations rebounded and now have moved into some of the most 
highly polluted areas of the Bay including Baltimore Harbor (Rattner et al., 2004, Watts 
et al., 2004). Egg samples showed that p,p’-DDE and other organochlorine pesticide 
levels were decreasing in these areas, while PCB levels remained unchanged.  Fish 
consumption advisories recommend avoiding several major fish species due to PCB 
contamination including channel and white catfish (Ictalurus punctatus and Ictalurus 
catus), carp, and white perch (Morone americana) (MDE, 2009).  These high 
concentrations of PCBs may be indicative of a suite of potential adverse effects 
mentioned in the earlier section including endocrine-thyroid, behavioral, developmental 
and cardiovascular effects in developing embryos and nestlings that solely feed on fish 
caught in highly contaminated areas in Baltimore Harbor.  The sediments in Sparrow’s 
Point/Bear Creek area in Baltimore Harbor and the Inner Harbor itself remain heavily 
contaminated (McGee et al., 1999).  Due to the large increase in the osprey population, 
birds are now nesting in these areas.  Surveys conducted in 2011 in Baltimore 
20 
Harbor/Patapsco River revealed that at least three osprey pairs are now nesting in Bear 
Creek where only one pair was observed in the 2000-2001 study (Rattner et al., 2004).   
In nearby Back River, elevated sediment PBDE concentrations were detected at 
Cox’s Point.  Back River receives effluent from the state’s largest WWTP.  At Cox Point, 
total BDE levels were 4,735 ng total BDE/g dw and the BDE-209 congener made up 93% 
of the total (Klosterhaus, 2007).  Few ospreys have been reported nesting in Back River, 
but the pairs that do may be at risk of exposure to high levels especially since many of the 
PBDE congeners can biomagnify at all levels in the food web (Klosterhaus, 2007; Chen 
et al., 2009).  As seen on the Potomac River, values of PBDE congeners were high near 
the Blue Plains WWTP at levels that may have  the potential to exert adverse effects, not 
just on avian survival, but also induce sublethal effects including oxidative DNA damage 
and behavioral effects (e.g., Fernie et al., 2006, 2008, 2009; Rattner et al., 2013).   
 
Elizabeth River 
The Elizabeth River was listed as an US EPA Chesapeake Bay ROC in the 1990’s 
due to issues with toxic contamination and poor water quality and currently is identified 
in the Chesapeake Bay Executive Order as an area of concern (Chesapeake Bay 
Executive Order, 2009).  In 1973, when the osprey population was surveyed at 1,450 
pairs of nesting birds, only two pairs were located in the James River and nearby 
tributaries (Henny et al., 1974; Rattner et al., 2004).  The Elizabeth is the location of the 
largest naval shipyard in the United States and also many manufacturing plants; thus, the 
Elizabeth faces significant pollution and sewage discharge issues.  In fact, as far back as 
1983, the US EPA singled out the Elizabeth as the most highly polluted river in the Bay 
21 
Watershed (Elizabeth River Project, 2008).  Its highly contaminated toxic sediment has 
resulted from major creosote spills (Atlantic Wood Industries Inc.).  In addition, it also 
contains high concentrations of PAHs, metals and PCBs (US EPA, 1994).  Both 
histopathological lesions and the increased presence of DNA adducts have been reported 
in mummichog (Fundulus heteroclitus) populations in this area (Rose et al., 2000).  DNA 
adducts are a characteristic effect of PAH exposure and are indicative of a high cancer 
rate that was found in fish (e.g., identified by the presence of skin ulcerations and 
preneoplastic lesions).   Fish consumption advisories remain in effect for PCBs and 
dioxins.  These advisories warn against eating a suite of fish species including gizzard 
shad, carp, and several catfish species (VDH, 2009).   
Many of these species are readily consumed by of the osprey (Glass and Watts, 
2009) and the birds nesting in these areas are consuming PCB contaminated fish on a 
regular basis.  As discussed above, there is a wide range of many sublethal effects of 
PCBs and developing embryos and nestlings in this area may be at risk.  Other species in 
the area exhibit increased sublethal effects such as the DNA damage in mummichung.  
Many efforts have been underway to clean up the Elizabeth River and restore some of the 
most highly contaminated areas.  Recently, the Elizabeth River Project has initiated 
efforts in areas on the South Branch, such as Money Point, where sediments have been 
dredged and wetlands and oyster reefs are now restored (Elizabeth River Project, 2008).  
Interestingly, a decade ago there were only a few pairs of osprey on the South Branch, a 
sign of the severe ecosystem degradation.  The presence of successful nesting pairs, their 
contaminant burden and health of ospreys nesting in these degraded areas will serve as 
sentinel of the health of the entire ecosystem.  
22 
James River 
 
The James River flows entirely through the state of VA and serves as a major 
tributary to Chesapeake Bay.  The Chesapeake Bay Executive Order called for ongoing 
monitoring efforts to understand the extent of contaminants on the James River.  
Historically, chlorine was responsible for two major fish kills in 1973 and 1974.  
Subsequently, the larger James River and Boat Harbor Plants were identified as potential 
sources of wastewater contaminants along with the smaller Fort Eustis and Williamsburg 
plants (Bellanca and Bailey, 1977) further downstream.  Upriver, there are inputs from 
the Richmond WWTP, Falling Creek WWTP, Proctor’s Creek WWTP and Hopewell 
WWTP.  The James River has a large osprey population that has been well-studied 
(Watts et al., 2004).  Osprey nests can become very dense and plentiful especially in the 
high productivity areas near Hopewell and further upriver near Presquile National 
Wildlife Refuge.  The suite of WWTPs on the James River has the potential to be 
hotspots for PBDEs (Chen and Hale, 2010) to enter the food web and exert potential 
adverse effects on wildlife as discussed for the Anacostia/middle Potomac River. 
Along with inputs of nitrogen and phosphorous, the James River health was 
compromised by Kepone® contamination from its manufacturing source in Hopewell as 
early as the mid 1960’s and was followed by the decline of nesting ospreys in this area 
(Kennedy, 1977).  Luellen and coworkers (2006) reported overall concentrations of the 
insecticide Kepone® were decreasing in fish, but samples still contained detectable levels. 
Sediment burial continues to be the primary source of removal of Kepone® from the 
environment; however, turbidity and sediment re-suspension could continue to result in 
hotspots of contamination.   In 2001, osprey egg samples collected off of Craney Island 
23 
in the Elizabeth River were analyzed for Kepone®, but this insecticide was not detected 
(Rattner et al., 2004).   
Although Kepone® advisories are still in effect, the fish consumption advisories 
are primarily driven by PCBs (Harris and Jones, 2008; VDH, 2009).  The highest levels 
of PCBs were detected in carp and blue catfish in Richmond, Hopewell and Bailey’s 
Creek, VA, all exceeding thresholds for suitable fish consumption set by the US EPA. A 
number of sewage treatment plants are clustered around the major cities (Richmond, 
Falling and Proctor’s Creek, Hopewell, Williamsburg and Norfolk). At Richmond, the 
river is highly industrialized and narrow, until below Hopewell where the predominant 
land use is for agriculture. Much of this area is listed as a priority agriculture watershed 
for inputs of both nitrogen and phosphorous (Wolf, 2009).  Although the osprey 
population has been studied extensively on the James River (e.g., Glass and Watts, 2009; 
Watts et al., 2004), there is limited contaminant information on legacy contaminants, 
replacement brominated flame-retardants (e.g., HBCD, TBBH, and TBB), and no 
information the occurrence of pharmaceuticals in wildlife. 
 
 
Lower Susquehanna River and flats 
 
As the largest tributary to the Chesapeake, the Susquehanna empties into the 
northern Bay and provides 50% of the freshwater input. The river flows down from New 
York (NY), Pennsylvania (PA) and Maryland (MD), encompassing a drainage basin of 
71,200 km2 (Susquehanna River Basin Commission, 2006). Human activities continue to 
contribute toward habitat loss and degradation.  The basin yields some of the most 
productive and heavily used agricultural lands, and transports 66% of the nitrogen and 
24 
40% of phosphorous non-point nutrient inputs into the Bay on average per year, which 
directly affect habitat and water quality for fish and wildlife (Langland, 1998).  In 
addition, flooding, storm water run-off and effluents from urban, agricultural and 
industrial sources contribute 25% of sediment loads to the Bay.  In 1996, the lower river 
basin was placed on Maryland’s 303d list for nutrients, sediments, Cd and PCBs (MDE, 
2005).  Although there are historic data related to health, contaminant exposure and 
associated hazards for fish in Chesapeake Bay (Pinkney et al., 2001, 2004a, 2004b; 
Blazer et al., 2007), considerably less is known about waterbirds particularly in northern 
tributaries (Rattner and McGowan, 2007).  Due to the distribution of osprey nests on the 
Susquehanna in 2013, sampling efforts were concentrated at the mouth of the 
Susquehanna and the Susquehanna flats.  The Susquehanna flats are located at the 
northernmost region of the Bay (Havre de Grace, Maryland). 
 
Purpose of this Research and Scientific Objectives   
In 2000, the Chesapeake Executive Council Toxics Strategy highlighted the 
ongoing toxic contamination issues US EPA ROCs and called for ongoing efforts for 
monitoring, clean up and remediation by 2010 (Chesapeake Bay Program, 2000).  
Though progress has been made, goals have not been met.  Therefore in May 2009, the 
Chesapeake Bay Executive Order 13508 (Chesapeake Bay Executive Order, 2009) was 
signed into action, and placed an expanded emphasis on the restoration and protection of 
the Bay and its biota.  Specifically, contaminants are addressed under goals developed to 
restore clean water in an effort to reduce contaminant exposure to fish and wildlife.  
Under this Executive Order goal, the US EPA, Department of the Interior (DOI) and the 
National Oceanic and Atmospheric Administration (NOAA) are called on to work with 
25 
state and local governments in understanding the extent of the contaminant issues in the 
Bay and also develop reduction goals by 2013 along with strategies for reducing 
contaminants by 2015. New threats will also be addressed including emerging 
contaminates in 2011 and 2012.  In order to address goals and outcomes laid out in the 
Executive Order, the U.S. Geological Survey (USGS) in collaboration with other 
organizations has expanded its research and monitoring activities to assist Federal and 
State management agencies, and nongovernment organizations to improve the health and 
quality of the Bay.   
This project will increase our scientific knowledge of the occurrence and effects 
of contaminants in fish and wildlife populations in two major tributaries of concern 
identified by the Executive Order (Potomac and James Rivers) with continued monitoring 
of the Chesapeake Bay ROCs (Figure 1).  Summary data from the Chesapeake Bay 
Program indicate that waters in these tributaries are impaired by PCBs, metals and 
unidentified toxics (Weinberg, 2010).  Sampling will also include the Susquehanna River 
and flats, at the mouth of the largest tributary to the Chesapeake Bay, where limited 
contaminant exposure data on wildlife has been collected (Figure 1).  Although fish-
eating birds have been used in previous Chesapeake Bay ecotoxicological studies, there 
are limited contaminant exposure records for Chesapeake Bay in the “Contaminant 
Exposure and Effects – Terrestrial Vertebrates” (CEE-TV) database for these locations 
since 2000-2001 and no information on the occurrence of pharmaceuticals and personal 
care products (Rattner and McGowan, 2007).  A decade has elapsed since the last large-
scale ecotoxicological study of Chesapeake Bay wildlife was undertaken.  
Bioaccumulative replacement flame-retardants are being produced in large-volumes and 
26 
have shown potential reproductive effects in American kestrels (Martinienson et al., 
2011, 2012). They have also shown up in a suite of biological matrices (Covaci et al., 
2010).  Pharmaceuticals and personal care products have been detected in water and fish 
samples from wastewater effluent dominated streams including the Potomac River 
(Iwanowicz et al. 2009); however, it is unknown to what extent they bioaccumulate 
through the food web and are bioavailable to piscivorous species.  
In this study ospreys were used as a sentinel to monitor a suite of regional 
waterways and tributaries to Chesapeake Bay where there are limited ecotoxicological 
data for wildlife, or locations that are hotspots for contaminants and have been 
historically impaired and identified by the US EPA as ROCs.  Ospreys have been well 
studied in the Chesapeake Bay, and this work will help us understand patterns in 
contaminants over time across a large spatial scale and help managers identify new 
contaminants that may be entering the food web which are important from both a human 
and wildlife health perspective.  Our studies focused on the spatial and temporal trends of 
contaminant exposure and effects on ospreys in Chesapeake Bay and placed both 
historical and emerging contaminants into a food web perspective.  
 
 
 
 
 
 
 
27 
 
 
 
 
 
 
 
 
 
1. The first study, “Decadal re-evaluation of contaminant exposure and productivity 
in ospreys nesting in Chesapeake Bay Regions of Concern” had two main 
objectives: 
 
a) Measured contaminants in osprey egg samples, and conducted morphological 
(culmen length and nestling body weight), reproductive endpoints to assess 
their relationship to contaminant exposure in Chesapeake Bay ROCs.  
 
b) Looked at spatial and temporal trends in osprey productivity and contaminant 
exposure in Chesapeake Bay ROCs.  
 
2. The second study, “Chesapeake Bay fish-osprey food web: contaminant exposure 
and potential genetic damage” had four objectives: 
 
a) Reconstructed osprey dietary preferences to account for variation in 
dietary exposure in order to relate contaminant concentrations of 
organochlorine pesticides PCBs, and PBDEs in whole fish to osprey eggs.  
This was done by calculation of biomagnification factors (Figure 2) and a 
comparison of the calculated factors to others found in the literature.  
 
b) Measured oxidative DNA damage in osprey nestlings across all study sites 
as a biomarker of contaminant effects. 
 
c) Conducted an analysis investigating the relationships between 
contaminants and stable isotopes by site to DNA damage. 
 
3. The third study, “Exposure and food web transfer of pharmaceuticals in ospreys: 
predictive model and empirical data” had two main objectives:  
Figure 2. A schematic of food web transfer matrices and overview of the 
major groups of contaminants quantified in the following studies. 
28 
 
a) Conducted a screening-level exposure assessment for 117 pharmaceuticals in 
ospreys. 
 
b) Analyzed 23 pharmaceuticals and an artificial sweetener in the water-fish-
osprey food web (Figure 2).  
 
The following chapters describe a series of three studies conducted to evaluate 
contaminant exposure and effects in ospreys nesting in Chesapeake Bay.  The first study 
described in Chapter 2 consisted of a decadal re-evaluation of contaminants in 
Chesapeake Bay ROCs, spatial and temporal patterns in contaminant exposure and 
potential effects at the genetic, individual and population levels.  The next chapter 
(Chapter 3) described a study investigating contaminant exposure from a food web 
perspective.  Here we conducted an integrative sampling scheme and calculated 
biomagnification factors from fish to osprey for both legacy and emerging contaminants.  
This study also evaluated effects at a genetic, individual and population level (looking at 
changes in DNA damage, body weight, culmen length and overall osprey productivity).  
The final study (Chapter 4) evaluated a suite of 23 pharmaceuticals and an artificial 
sweetener (sucralose) in the water, fish osprey food web.  A screening-level risk 
assessment was conducted to predict those compounds, which may reach the human 
therapeutic dose in a single day or over a 40-45 day nestling period.  This assessment was 
complemented by empirical measurements of pharmaceuticals in water, fish and osprey 
nestling plasma samples from several tributaries in the Bay.   
 
 
 
 
29 
 
 
 
 
 
 
 
 
 
 
CHAPTER 2 
A DECADAL RE-EVALUATION OF CONTAMINANT EXPOSURE AND EFFECTS 
ON OSPREYS (Pandion haliaetus) NESTING IN CHESAPEAKE BAY REGIONS OF 
CONCERN 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
30 
Introduction 
The Chesapeake Bay is the largest estuary in the United States and supports a 
diversity of avian species. Degradation of habitat quality by a mixture of agricultural, 
industrial and urban pollution continues to threaten the most vulnerable portions of the 
estuary, and jeopardize fish and wildlife health.  Globally, the largest nesting osprey 
(Pandion haliaetus) population is found in the Chesapeake Bay, which has been 
nicknamed the “osprey garden of the world” (Poole, 1989). Ospreys are used as 
ecological sentinels of environmental health due to their high trophic level, widespread 
distribution and nest-site fidelity (Grove et al., 2009; Henny et al., 2010). 
The Chesapeake Bay osprey population has been extensively studied since the 
1970s. In 1973, the osprey population was estimated to be only 1450 nesting pairs, and 
only seven pairs were observed north of the Chesapeake Bay Bridge (near Annapolis, 
MD) to the Susquehanna River (Henny et al., 1974). With the banning of the pesticide 
DDT in 1972, the osprey population rebounded both numerically and spatially. A bay-
wide survey in 1995–1996 estimated 3500 nesting pairs, and population growth was rapid 
in the tidal freshwater tributaries and areas north of the Chesapeake Bay Bridge (Watts 
et al., 2004). 
For decades, impaired water quality and toxic chemicals in sediment, water and 
biota have been found in highly industrialized and urbanized areas of Chesapeake Bay 
including Baltimore Harbor and the Anacostia and Elizabeth Rivers. These three sites 
have been designated by the U.S. Environmental Protection Agency (US EPA) as 
Regions of Concern (ROCs) (US EPA, 1994). Rattner et al. (2004) reported that osprey 
productivity in these regions was marginally adequate to sustain local populations. 
31 
Although concentrations of p,p’-DDE and other organochlorine pesticides declined in 
eggs, PCB levels remained elevated especially in Baltimore Harbor compared to 
reference sites. Polybrominated diphenyl ethers (PBDEs) were detected in osprey eggs 
from all sites, with values being some of the greatest found in North America.  All ROCs 
continue to have human health advisories on the consumption of several fish and shellfish 
species due to contamination with PCBs and other pesticides (VDH, 2009; MDE, 2014). 
In May 2009, the Chesapeake Bay Executive Order 13508 was signed, which 
placed emphasis on continued monitoring of ROCs and the restoration and protection of 
the Bay (Chesapeake Bay Executive Order, 2009).  Just over a decade has elapsed since 
the last large-scale ecotoxicological monitoring study of Chesapeake Bay wildlife, during 
which time there were limited exposure data for Bay avifauna (Rattner and McGowan, 
2007; Chen et al., 2010; US EPA, 2012a).  As part of a larger study examining 
contaminant exposure, food web transfer, and potential effects on ospreys in Chesapeake 
Bay, the three historical ROCs were re-visited to examine temporal and spatial changes in 
osprey productivity and concentrations of legacy and more contemporary pollutants.  
 
Materials and Methods 
Study sites 
Ospreys nesting on navigational markers, platforms, duck blinds, and other 
accessible structures were sampled during their nesting seasons from March through July 
of 2011 and 2012. Sampling was conducted in Chesapeake ROCs including (i) Baltimore 
Harbor/Patapsco River in MD (2011, n = 7), (ii) the Anacostia/middle Potomac Rivers in 
parts of Washington, D.C., Maryland, and Virginia (2011, n = 9), and (iii) the Elizabeth 
32 
River in Virginia (2012, n = 6) (Figure 3). These sites encompassed urban and industrial 
gradients along a 20–25 km stretch of each river. The Paul S. Sarbanes Ecosystem 
Restoration Project at Poplar Island, MD (hereafter, Poplar Island), a remote mid-Bay 
location, was used as a reference site (2011–2012, n = 4 eggs/year). Notably, results from 
a 2010 common tern (Sterna hirundo) egg collection at Poplar Island indicated low levels 
of organochlorine pesticides, non-coplanar PCBs, and PBDEs (Rattner et al., 2013), 
making it a more suitable reference site than the South, West and Rhode Rivers that were 
used in 2000–2001 (Rattner et al., 2004). In 2011–2012, nests in ROCs were strategically 
selected near those nests previously studied. All procedures involving ospreys were 
conducted under approval of the Institutional Animal Care and Use Committees of the 
USGS-Patuxent Wildlife Research Center (USGS-PWRC) and the University of 
Maryland, and with appropriate Federal and State scientific collection permits. 
 
Egg sample collection 
Using the sample egg technique described in Blus (1984), osprey eggs were 
collected for residue analysis (n = 30 total). After each clutch was complete (three or 
more eggs), a random egg was sampled within a week for subsequent contaminant 
analysis. Eggs were transported to the USGS-PWRC, cleaned with distilled water, 
weighed and their length and breadth were measured to the nearest 0.01 mm. A 2.4-mm 
hole was drilled into the blunt end of the egg (Dremel® MultiPro® 7.2 V, Model 770, Mt. 
Prospect, IL, USA). 
 
 
 
33 
 
 
 
 
 
 
Moisture loss during incubation concentrates contaminants in the egg.  Distilled water 
was injected into the air cell to return moisture content and contaminant concentrations to 
those in the egg when it was freshly laid (Heinz et al., 2009a). Eggs were weighed, 
opened to determine fertility and developmental stage, and contents transferred into a 
chemically clean jar (I-CHEM, VWR Scientific, Radnor, PA, USA), stored at −80 °C, 
and eventually transported to the Virginia Institute of Marine Science (VIMS), 
Figure 3. Locations of sample nests in Chesapeake Bay Regions of Concern 
and Poplar Island reference site, (●) indicates a sampled nest.  
34 
Gloucester, VA, USA. Eggshells were dried for 3–4 months at room temperature and 
measured for thickness at three sites on the equator using a micrometer (Model 1010M; 
L.S. Starrett Co., Athol, MA, USA) and averaged. 
 
Osprey reproductive success 
Osprey productivity was monitored following the definitions and traditional 
methods described by Postupalsky (1977) and more recently Steenhof and Newton 
(2007). Initial nest visits were made in March to locate study nests and determine 
breeding status. Additional nests were monitored to have back-up locations for sampled 
nests if the targeted nests failed. Nests were visited at 7–10 day intervals to determine the 
fate of the nest including the numbers of eggs laid, eggs hatched, and young present at 
advanced age (>45 days) to fledging. Other observations including evidence of predation 
or disturbance were noted. 
 
Osprey nestling blood samples and morphological endpoints 
A blood sample was collected from a 40–45-day-old nestling at each nest from 
which a sample egg was collected. In the event the target nest failed (8/30 instances), the 
sample was drawn from a nestling residing in a nearby nest (within 2.5 km). Specifically, 
before fledging, one nestling per nest (n = 30) was removed from the nest for about 
10 min. Following physical examination, culmen length and body weight were measured 
and the crop was palpated to determine level of crop filling. A 5–7-mL brachial blood 
sample was drawn using a 23-gauge 1-inch needle into a heparinized syringe (Sarstedt 
International, Newton, NC, USA). About 100 μL of fresh osprey nestling blood was 
35 
transferred to a microcentrifuge tube, frozen on dry ice, and stored at −80 °C for the 
DNA damage assay. The remainder of the blood sample was saved for a concurrent study 
(Lazarus et al., 2015b). 
 
Analysis of contaminant residues in osprey eggs 
Osprey egg analyses for 11 PBDE congeners, 5 alternative brominated flame 
retardants [(alt-BFRs: α, β, γ hexabromocyclododecane (HBCD), 1,2-bis (2,4,6-
tribromophenoxy)ethane (BTBPE), di(2-ethylhexyl)-2,3,4,5-tetrabromophthalate 
(TBPH), 2-ethylhexyl 2,3,4,5-tetrabromobenzoate (TBB), and decabromodiphenyl ether 
(DBDPE)], 129 PCB congeners, 44 organochlorine pesticides and methoxytriclosan were 
conducted based on the methods of Chen et al. (2008) and La Guardia et al., 2007 and La 
Guardia et al., 2010. Egg contents were homogenized, lyophilized and then spiked with 
surrogate standards of PCB 30, 65 and 204 (Ultra Scientific, North Kingstown, RI, USA), 
13C-methoxytriclosan, 13C-PCB-126 and 2,3,4,4′,5,6 – hexabromodiphenyl ether (BDE 
166; Cambridge Isotope Laboratories, Inc., Andover, MA, USA). Blanks, consisting of 
sodium sulfate (baked at 450 °C overnight), were analyzed to evaluate possible 
laboratory contamination. Dried samples underwent accelerated solvent extraction 
(Dionex ASE 200, Sunnyvale, CA, USA) using methylene chloride (DCM) at 100 °C and 
68 atm. 
 Extracts were purified by size exclusion chromatography (SEC, Envirosep-ABC®, 
350 × 21.1 mm column; Phenomenex, Torrance, CA, USA). Each post-SEC extract was 
reduced in volume, added to a solid-phase 2-g silica glass extraction column (Isolute, 
International Sorbent Technology, Ltd., Hengoed Mid Glamorgan, UK) and eluted with 
36 
3.5 mL hexane (to waste), followed by 6.5 mL of 60:40 hexane/DCM and 8 mL DCM. 
The latter two fractions were combined, pooled and then split, with half going for 
coplanar PCB analysis. Coplanar PCBs were separated from nonplanar PCBs by elution 
through a Supleclean ENVI-Carb SPE column (Sigma–Aldrich Co., St. Louis, MO, 
USA). The column was first eluted with 15 mL hexane (to waste). The coplanar 
congeners were obtained by elution with 20 mL hexane/toluene (99:1) and 20 mL 
toluene. The pooled eluent was reduced in volume, spiked with p-terphenyl (Ultra 
Scientific, North Kingstown, RI, USA) as an internal standard, and analyzed by gas 
chromatography/mass spectrometry (GC/MS) on an Agilent 5975C instrument (Agilent 
Technologies, Santa Clara, CA, USA), in electron impact mode and selected ion 
monitoring. A 60 m DB-5 MS GC column (Agilent, 0.32 mm ID × 0.1 μm thickness) was 
used. 
 The second half of the silica SPE fraction retained was spiked with 
decachlorodiphenyl ether (Ultra Scientific) as the internal quantitation standard. 
Identification and quantitation of non-coplanar PCBs was conducted by GC/MS in the 
electron ionization mode on a Varian 2200 GC/MS (Agilent Technologies). 
Organochlorine pesticides and methoxytriclosan were analyzed similarly by GC/MS on a 
Varian 4D MS. Both analyses used 60-m DB-5 columns (0.32 mm ID × 0.25 μm 
thickness). 
 The PBDEs and alt-BFRs were analyzed using this same fraction. PBDEs and alt-
BFRs were separated by ultra-performance liquid chromatography (UPLC, Waters Corp., 
Milford, MA, USA) and analyzed by atmospheric pressure photoionization tandem mass 
37 
spectrometry (APPI/MS/MS, Q-Trap3200 MS, AB Sciex, Framingham, MA, USA). 
Further details of the UPLC-APPI/MS analysis can be found in La Guardia et al. (2013). 
 
Quality control and assurance 
Instrumental analysis and laboratory QA/QC for method development are 
described in Chen et al. (2008). The method detection limit (MDL) for organochlorine 
pesticides and non-coplanar PCB congeners was 0.4 μg/kg ww, coplanar PCBs was 
0.04 μg/kg ww, and PBDEs and alt-BFRs was 0.4 μg/kg ww. Data were corrected based 
on the recovery of surrogate standards in each sample. For quantification of 
organochlorine pesticides and methoxytriclosan, average recovery of the surrogate 
standard PCB-204 (similar physiochemical properties to the organochlorine pesticides) 
and 13C-methoxytriclosan from the eggs were (mean ± SD) 85.9 ± 11.0%. Recoveries of 
the surrogates for the non-coplanar PCBs (PCB-204) averaged 94.9 ± 13.7%, coplanar 
PCBs 97.9 ± 0.13%, and brominated flame retardants (BDE 166) 98.6 ± 24.1%. Overall, 
moisture content (adjusted to fresh weight) in eggs averaged 83.7 ± 1.37%. 
 Due to changes in analytical methodologies and laboratories used between 2000–
2001 (Rattner et al., 2004) and 2011–2012, a subset of 11 egg homogenates (three on the 
Anacostia/middle Potomac Rivers, five from Baltimore Harbor/Patapso Rivers, and three 
on the Elizabeth River) from 2000–2001 were re-analyzed with current samples. Based 
on data from our earlier work (Rattner et al., 2004), we selected those samples containing 
a range (relatively low, intermediate and high) of p,p′-DDE and total PCB residues. A 
comparison of the major stable groups of contaminants (p,p′-DDE, total PCBs, and sum 
of PBDE congeners 47, 99, 100, 153 and 154) was conducted and the percent differences 
38 
between each analysis was calculated (Supplemental Table S1). Individual PCB congener 
patterns were not evaluated on a temporal perspective due to substantial differences in 
analytical methodologies. Percent difference of moisture content of these samples in 
2000–2001 and 2011–2012 changed by only 1.48%, indicating that sample moisture had 
not changed appreciably. However, in 2000–2001, moisture correction was determined 
by mathematically adjusting to the volume of the entire egg (Rattner et al., 2004), while 
in 2011–2012 we physically adjusted by the addition of distilled water (Heinz et al., 
2009a). 
 In 2000–2001, PCBs were analyzed at the Geochemical and Environmental 
Research Group (GERG) at Texas A&M using an Aroclor profile-based analysis. Wet 
samples were mixed with sodium sulfate, spiked with surrogates and extracted with DCM 
in a Teckmar Tissumizer. Extracts were cleaned up on a silica/alumina column and 
purified by HPLC to remove interfering lipids. Identification/quantification was by GC in 
concert with electron capture detection (ECD) using an Aroclor profile-based approach 
(Rattner et al., 2004). In contrast, VIMS conducted a congener-based analysis using 
GC/MS in 2011–2012 (Chen et al., 2008). Thus, some differences in results between the 
two analytical methods were expected. Total PCB concentrations from 11 samples were 
comparable using the two methodologies, with some 2011–2012 estimates being greater 
and others being less (average percent difference of 23.7 ± 20.5%) than values reported 
by GERG in 2000–2001. This seems quite reasonable, as others (Turle et al., 1991) 
reported that total PCB concentrations from Aroclor-based analyses were on average 
46% greater than congener-based analyses. 
39 
 The VIMS conducted PBDE analyses of the 2000–2001 egg samples using 
GC/MS (Rattner et al., 2004) and the re-analysis of a subset of five of these samples 
using ultra-performance liquid chromatography (UPLC)/MS 10 years later. Of the five 
samples reanalyzed in 2011–2012, average percent difference between the original 
samples and their re-analysis was 17.9 ± 13.4% (with some 2011–2012 measurements 
higher and some lower compared to GERG values from 2000–2001). Perhaps most 
surprisingly, the re-analysis of 11 archived samples for p,p′-DDE varied by 44.3 ± 24.5% 
(Supplemental Table S1). A GC/ECD method was used by GERG for the original 2000–
2001 analyses, whereas a GC/MS method was employed for the re-analysis by VIMS. All 
surrogate recoveries within the 2011–2012 samples were acceptable (84.9 ± 11.0%). 
Thus, due to variation between labs, years and sample storage, quantitative comparisons 
between p,p′-DDE values from this current study with published values (Rattner et al., 
2004) are not justified. Notably, values for p,p′-DDE in osprey eggs in 2011–2012 were 
well below the threshold for eggshell thinning (9 of 11 were <1.0 μg/g ww; Wiemeyer 
et al., 1988). 
 
Toxic equivalents for PCB congeners 
 The World Health Organization toxic equivalency factors (TEFs) were used to 
estimate the toxic equivalent (TEQ) for birds (van den Berg et al., 1998). Since 1998, 
TEFs have changed slightly for mammals, but not for avian species (van den Berg et al., 
2006). The TEQ of Ah-receptor active PCB congeners was calculated by multiplying the 
congener concentration (pg/g ww) by the TEF and summing individual values. 
 
40 
DNA damage assay 
Whole blood from osprey nestlings was analyzed for the presence of 8-hydroxy-
2′-deoxyguanosine (8-OH-dG), a biomarker of oxidative DNA damage, using the 
DNA/RNA Oxidative Damage EIA Kit (Cayman Chemical, Ann Arbor, MI, USA). This 
analysis followed the protocol and assay validation described in Rattner et al. (2013). The 
limit of detection for this assay was 1.03 pg/μg DNA. Plates included blanks and 
reference samples to account for inter-assay variability. Samples were analyzed in 
duplicate. 
 Standard curves were fitted and concentrations determined using a 4-parameter 
model (R2 > 0.998; MARS Data Analysis Software 2.10 R3, BMG Labtech). Samples 
collected in 2011 and 2012 were assayed separately. Precision of duplicate 
determinations (intra-assay variation, coefficient of variation, CV ± SD) was 3.8 ± 2.7% 
for samples collected in 2011 and 5.1 ± 3.1% for samples collected in 2012. Duplicate 
analyses with a CV greater than 20% were re-analyzed. Inter-assay variation (CV among 
assay plates) for reference samples was 9.9 ± 11.3%. Due to a number of factors (e.g., 
DNA degradation in samples stored for varying durations, observed variation in assay 
performance among test kit lots), we were unable to justify statistical comparisons of 
results between the 2011 and 2012 collections. 
 
Statistical analysis 
Simple descriptive statistics were generated for measurement endpoints. 
Continuously distributed variables (productivity, eggshell thickness, morphometrics, 
DNA damage and contaminant residues) were tested for normality and homogeneity of 
41 
variance, and log transformed as necessary (SAS 9.3, SAS Institute Inc., Cary, NC, 
USA). Comparisons were made among tributaries using analysis of variance (ANOVA) 
followed by Tukey's HSD multiple mean comparison test to detect site-specific 
differences (α = 0.05). For those variables that did not meet assumptions for parametric 
analysis, a generalized Wilcoxon non-parametric test was used followed by a Bonferroni 
adjustment for comparison of multiple means. A Fisher's Exact test was used to compare 
overall site-specific differences in productivity endpoints. For those sampling sites with 
residues below the MDL in <50% of the samples, the potential range of the mean was 
reported using the Kaplan–Meier method (Helsel, 2005). Any outliers were eliminated 
from the dataset using a Grubb's test. 
Egg, young and nest survival probabilities were calculated using the Mayfield 
method (Mayfield, 1961; Bart and Robson, 1982). For those variables that were not 
normally distributed and/or exhibited heterogeneous variance, a generalized Wilcoxon 
non-parametric test was used. If this was significant, two-sample comparisons were 
conducted and p-values were adjusted using a Bonferroni correction. Results for total 
PCBs and PBDEs were statistically compared between the 2011–2012 and 2000–2001 
from (Rattner et al., 2004) collections using a t-test. 
A logistic analysis of covariance was used to examine site-specific differences in 
the relationship between egg residues (p,p′-DDE, total PCBs, PBDEs and the sum of p,p′-
DDE + total PCBs) and productivity (nest success, egg loss). If no differences were 
detected, logistic regression was used to evaluate the relation between residues and 
productivity across sites. 
 
42 
Results 
Reproduction of ospreys 
In 2011, 36 nests were identified along a 25-km segment on the Anacostia River, 
from the Frederick Douglass Bridge to Dogue Creek, a tributary of the middle Potomac 
River. Thirty nests were found along a 20-km stretch from Baltimore Harbor (Curtis 
Creek) east to the mouth of Patapsco River (Bodkin Point). In 2012, 29 nests were found 
along a 20-km stretch of the Elizabeth River (including the Lafayette River, West and 
South branches). No nests were found on the East branch of the River. At Poplar Island 
(reference site), there were 12 active nests in 2011 and 24 in 2012. Osprey nest density 
based on total water surface area surveyed was 1 nest/1.60 km2 on the Anacostia/middle 
Potomac Rivers, 1 nest/3.9 km2 in Baltimore Harbor/Patapsco River, and 1 nest/3.57 km2 
on Elizabeth River. Poplar Island had the highest density of osprey nests in 2012, at 1 
nest/0.19 km2. 
Productivity, the average number of fledglings per active nest, at each location 
ranged from 1.00 to 1.43 (Table 1). There were no site-specific differences in number of 
eggs laid, egg loss, hatching success, chick loss, and fledging and nest success (Fisher's 
Exact Test and generalized Wilcoxon non-parametric test, p > 0.35). On average, 3.09 
eggs were laid/nest, 39.6% of eggs laid were lost, 91.5% of the eggs retained in nests 
hatched (hatchability), and 94.2% of the hatchlings survived to fledge. While not 
statistically significant, it is noteworthy that the lowest percentages of successful pairs 
and fledglings produced per active nest were on the Anacostia/middle Potomac (55.5% 
successful pairs, 1.11 fledglings/active nest) and the Elizabeth Rivers (66.7%, 1.00). 
43 
Once hatched, most chicks survived to fledge with the exception of the disappearance of 
one nestling on the Potomac River, and one nestling on the Elizabeth River. 
 
Table 1. Reproductive success of osprey nesting in Chesapeake Bay Regions of Concern. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
44 
Eggshell thickness and morphological endpoints 
On Poplar Island, no significant differences (t-test, p > 0.8) were found in osprey 
eggshell thickness between 2011 and 2012; accordingly, samples from both years were 
combined for this reference site. There was an overall difference among sites 
(ANOVA, p = 0.05). Using Tukey's HSD mean separation test, eggshell thickness on the 
Anacostia/middle Potomac Rivers (0.481 ± 0.049 mm) was significantly lower than for 
Poplar Island (0.549 ± 0.057 mm) (p = 0.04). While eggshell thickness was numerically 
lowest on the Anacostia/middle Potomac Rivers compared to the other sites (Baltimore 
Harbor 0.504 ± 0.040 mm and Elizabeth River 0.529 ± 0.044 mm), this trend was not 
significant (p > 0.48). Of the egg that failed to hatch on the Potomac, there was no 
evidence of eggshell thinning (0.503 mm). 
All nestlings examined were in good body condition. One chick on the Elizabeth 
River had a plastic bag around its neck, but otherwise appeared to be fine. Osprey 
nestling body weight (body weight minus estimated weight of food in crop; Schaadt and 
Bird, 1993) was compared among sites. Several of the nestlings on the Elizabeth River 
had mites and there were instances of feather plucking indicative of sibling competition 
for food. There were no differences in body weight between sampling years on Poplar 
Island (t-test, p > 0.6); therefore, measurements were combined. There were no 
differences in osprey nestling body weight among the four sites (p = 0.11, ANOVA; 
Poplar Island 2011–2012, 1603 ± 159 g; Anacostia/middle Potomac Rivers 1480 ± 144 g; 
Baltimore Harbor/Patapsco Rivers 1551 ± 140 g; Elizabeth River 1365 ± 138 g). There 
were no differences in culmen length among sites (p = 0.3, ANOVA) and values ranged b 
 
45 
Organochlorine pesticides and metabolites 
 Twenty-five out of 44 organochlorine pesticides and metabolites were detected in 
eggs at concentrations exceeding the MDL. The Anacostia/middle Potomac Rivers and 
Baltimore Harbor/Patapsco River sites had greatest pesticide and metabolite residues, 
some exceeding Poplar Island values (2.2–11.5 times) by more than an order of 
magnitude (Table 2). The Anacostia/middle Potomac also had the highest frequency of 
detects and measureable concentrations were found in all eggs sampled at this site with 
the exception of trans-chlordane (present in 8 of 9 eggs). Of this dataset, only one egg 
from Baltimore Harbor had p,p’-DDE concentrations (1.8 μg/g ww) that fell within the 
95% confidence interval (1.2–3.0 μg/g ww) associated with 10% shell thinning in osprey 
eggs (2.0 μg/g ww; Wiemeyer et al., 1988). There was no relation between p,p’-DDE 
concentration and eggshell thickness (r = −0.14, p = 0.459, n = 30). 
 
 
 
 
 
 
 
 
 
 
 
46 
 
Table 2.  Concentrations (μg/g ww) of organochlorine pesticides and metabolites from 
Chesapeake Bay Regions of Concern and Poplar Island reference site. a 
 
 
 
 
 
 
 
Polychlorinated biphenyls 
 Total PCB concentrations were up to five times greater in Baltimore Harbor and 
the Anacostia/middle Potomac Rivers (p < 0.0001) than at the reference site (Table 3). 
Some of the greatest values were detected in eggs collected in Baltimore Harbor, ranging 
up to 35.0 μg/g ww. Of the most potent Ah-receptor active PCB congeners (77, 81, 126, 
169), residues of PCB 77 and 126 did not differ from the reference site. However, 
47 
residues of PCB congeners 81 and 169 in eggs were greater on the Elizabeth River 
compared to the Anacostia/middle Potomac Rivers and Baltimore Harbor (p < 0.01), but 
were not detected at Poplar Island. In terms of Ah-receptor active PCB congeners, 
Baltimore Harbor and the Anacostia/middle Potomac Rivers had greater TEQs (16.3 and 
12.8 pg/g respectively; p < 0.04) than the Elizabeth River and the reference site (4.91 and 
2.30 pg/g ww). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48 
Table 3. Concentrations (μg/g ww) of polychlorinated biphenyls and congeners from 
Chesapeake Bay Regions of Concern and Poplar Island reference site. a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Polybrominated diphenyl ether flame retardants 
Congeners 47, 85, 99, 100, 153, 154, 183, 206 and 209 were detected in osprey 
eggs collected from one or more sites (Table 4). Penta-BDE congener 85, hepta-183, 
nona-206 and 209 were detected less frequently than other congeners in samples. 
49 
Congeners 28 and 66 were not detected in any of the samples. Osprey eggs from Poplar 
Island had the lowest PBDE congener residues; concentrations did not differ between 
2011 and 2012 (p > 0.05) and were combined. There were significant site differences for 
both total PBDEs and BDE 47 (p < 0.0001, ANOVA). Specifically, total PBDEs were 
more than three times greater on the Anacostia/middle Potomac and Baltimore 
Harbor/Patapsco Rivers compared to the Poplar Island reference site (p < 0.0001). This 
same site pattern and magnitude of difference was also apparent for BDE congener 47 
(p < 0.0001). The Anacostia/middle Potomac Rivers also had higher concentrations of 
PBDE congeners 99, 100 and 154 compared to Poplar Island and the Elizabeth River 
(p < 0.01). An egg collected near Blue Plains Wastewater Treatment Plant (WWTP) 
contained the greatest total PBDE levels (802 ng/g ww) found in this study. In 2000–
2001, PBDE 206 and 209 were not studied. We re-analyzed 11 archived sample 
homogenates from 2000–2001, and BDE 209 was detected in 3 of 11 samples (range: 
0.66–3.88 ng/g ww; 1/each ROC). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
50 
 
Table 4.  Concentrations (ng/g ww) of polybrominated diphenyl ethers and alternative 
brominated flame retardants from Chesapeake Bay Regions of Concern and Poplar Island 
reference site.a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51 
Other brominated flame retardants and methoxytriclosan 
Five additional BFRs were quantified in osprey eggs (Table 4). Alpha-HBCD was 
most frequently detected (26 of 30 eggs), but tended to be at much lower concentrations 
than the lower brominated BDE congeners. While BTBPE, TBB and TBPH were most 
frequently detected in samples from Baltimore Harbor/Patapsco River, the maximum 
detected concentration for each of these alt-BFRs was found in samples from the 
Anacostia/middle Potomac Rivers. Methoxytriclosan, a moderately bioaccumulative 
degradate of the antibacterial agent triclosan, was detected in all 9 samples from the 
Anacostia/middle Potomac Rivers (1.29–7.40 ng/g ww) and in one sample from 
Baltimore Harbor (5.55 ng/g ww). This compound may serve as a useful marker of 
domestic wastewater. 
The eleven historical egg homogenates were re-analyzed for these five alt-BFRs 
and methoxytriclosan. Alpha-HBCD was detected in 10 of 11 of the archived 2000–2001 
samples with no apparent site-specific differences in concentration (Baltimore 
Harbor/Patapsco River, n = 5 range: 0.53–1.27 ng/g ww; Elizabeth River, n = 3 range: < 
MDL-1.08 ng/g ww; Anacostia/middle Potomac Rivers, n = 3, range: 0.64–1.36 ng/g 
ww). Of the remaining four alt-BFRs, BTBPE, DBDPE, TBB and TBPH were 
consistently detected in two historical samples from Curtis Bay and Shallow Creek off of 
the Patapsco River (0.30–6.74 ng/g ww). However, methoxytriclosan was only detected 
in one sample collected in 2000 on the Potomac River (4.89 ng/g ww). 
 
 
 
52 
Relation between contaminants and osprey productivity 
Analysis of covariance revealed that there were no site-specific associations 
between p,p′-DDE and total PBDEs (p = 0.20), or between total PCBs and PBDEs 
(p = 0.07). However, there was a significant relation between p,p′-DDE and PCB residues 
across sites (p = 0.01), with the greatest correlation on the Anacostia/middle Potomac 
Rivers (r = 0.77, n = 9). When data for p,p’-DDE, total PCBs and total PBDEs were 
combined across sites (n = 30), there was a significant correlation between p,p’-DDE and 
total PCBs (r = 0.84, p < 0.001) and p,p′-DDE and total PBDEs (r = 0.62, p = 0.0003), 
and a marginal relation between total PCBs and total PBDEs (r = 0.32, p = 0.08). There 
were no associations (site-specific or combined sites) between egg loss or nest success 
and the concentrations of p,p′-DDE, PCBs, PBDEs or the sum of p,p′-DDE + PCBs 
(p > 0.09). 
 
Temporal comparisons 
Concentrations of p,p′-DDE, total PCBs and PBDE congeners 47, 99, 100, 153, 
154 in eggs collected in 2011–2012 were compared to results from 2000–2001 (Rattner 
et al., 2004, Figure 4). Total PCB concentrations remained high in Baltimore 
Harbor/Patapsco River between decades (p = 0.45), but decreased on the 
Anacostia/middle Potomac Rivers (9.28 μg/g in 2000–2001 versus 5.57 μg/g ww in 
2011–2012, p < 0.0001). Concentrations of PCBs remained unchanged on the Elizabeth 
River 10 years later (p = 0.96). For equitable comparison of total PBDE concentrations, 
the sum of congeners 47, 99, 100, 153 and 154 were examined between decades. The 
sum of these congeners across all ROCs declined over time (p < 0.04, Figure 2). A 
53 
qualitative comparison of p,p′-DDE levels between decades suggests that concentrations 
were low across all sites, and much lower than historic values (Wiemeyer et al., 1988;  
Rattner et al., 2004). 
 
 
 
 
 
Figure 4.  Temporal comparison of contaminants from Chesapeake Bay Regions of 
Concern in 2000-2001 (adapted from Rattner et al., 2004) and 2011-2012. Geometric 
means and individual data points are presented (●). At each site historical and current 
datasets were compared by a t-test.  
 
54 
DNA Damage 
In 2011, Baltimore Harbor/Patapsco River had significantly greater 
concentrations of 8-OH-dG (36.4 pg/μg DNA) compared to Poplar Island (26.6 pg/μg 
DNA, p = 0.04) (Figure 5), but not the Anacostia/middle Potomac Rivers (35.1 pg/μg 
DNA, p = 0.9). In 2012, the concentration of 8-OH-dG in one sample collected from the 
Elizabeth River (entrance to the Lafayette River) was several times greater (107.5 pg/μg 
DNA) compared to all other values in this study. While Grubb's test identified this 
measurement as an outlier, there is no rationale for excluding this value from our 
statistical analysis. In 2012, there was no significant difference between the Elizabeth 
River (35.5 pg/μg DNA) and Poplar Island (34.2 pg/μg DNA; p > 0.15) with or without 
exclusion of the outlier. 
 
 
 
 
 
 
 
 
 
 
Figure 5. Concentrations of 8-hydroxy-2’-deoxyguanosine (8-OH-dG) in osprey nestling 
whole blood samples. Means, standard deviations and individual data points are 
presented (●). Capital letters indicate a statistically significant difference (p < 0.05) using 
Tukey's HSD test.  
55 
Discussion 
Osprey productivity in Regions of Concern 
 
Use of the insecticide DDT, and specifically its metabolite p,p′-DDE, resulted in 
eggshell thinning and population declines of ospreys (Wiemeyer et al., 1988) and other 
species of piscivorous birds (Blus, 2011). In the 1970s, the Chesapeake Bay osprey 
population was estimated at 1450 pairs, with some population segments producing well-
below normal rates (Henny et al., 1974). Following restrictions in the use of DDT and 
other chlorinated pesticides, the Chesapeake osprey population more than doubled by 
1995–1996 (Watts et al., 2004). 
To determine the recruitment rate needed to maintain a stable osprey population, 
information is needed on nest site competition, nest availability and age of breeding; thus, 
it is challenging to identify which productivity rate is best for Chesapeake Bay ROCs 
(Watts and Paxton, 2007). Historically, it was estimated that 0.8–1.15 fledglings/active 
nest are required to maintain a stable population (Spitzer, 1980; Spitzer et al., 1983; 
Poole, 1989), although these estimates have likely changed. While contemporary 
survivorship data are available for ospreys in Chesapeake Bay (Bryan Watts, unpublished 
data), a model has yet to be developed to predict the reproductive rate necessary to 
maintain a stable population. Historical recruitment rates on the southern Potomac in 
1970 were 0.55 fledglings/active nest (Wiemeyer, 1971) and in 2000 the reproductive 
rate was 0.88 on Anacostia/middle Potomac Rivers (Rattner et al., 2004). By 2011, the 
reproductive rate had further increased to 1.11 fledglings/active nest on the 
Anacostia/middle Potomac Rivers, an increase of 0.56 fledglings/active nest over the 
1970s. Based on historical estimates (Spitzer, 1980; Spitzer et al., 1983; Poole, 1989), 
56 
since 2000, productivity in Baltimore Harbor appears to be adequate to maintain a stable 
population. The Elizabeth River appeared to have lower productivity (1.00 
fledgling/active nest) compared to the other sites, but still adequate to maintain a stable 
population. However, this lower productivity rate may be related to sample size (only 
n = 6 nests). By inclusion of data from seven additional Elizabeth River nests, the 
productivity estimate increased to 1.28 fledglings/active nest, which is more than 
adequate to maintain a stable population. Osprey productivity rates are primarily driven 
by food availability and brood reduction, but could also be adversely affected by p,p’-
DDE and other contaminant burdens in osprey eggs. 
Although there were no differences in productivity (i.e., number of eggs laid, 
hatching and fledging success) among sites, there were some notable changes in osprey 
nesting in ROCs. In 2000, only one osprey nesting pair was observed on Bear Creek off 
of Baltimore Harbor/Patapsco River, which is considered to contain the most highly toxic 
sediments in Baltimore Harbor (McGee et al., 1999). In 2011, at least four pairs were 
observed nesting on channel markers and power transmission towers. On the Anacostia 
River, only one pair was observed in 2000, but by 2011 there were at least four nesting 
attempts, two of which were successful. While it is tempting to conclude that decreasing 
contaminants in eggs could account for differences in productivity, temporal and site-
specific differences in other factors (e.g., food availability, nest site quality/availability, 
weather, predation, inexperienced breeders) could also contribute to fluctuations in 
fecundity (Levenson and Koplin, 1984; Stinson et al., 1987; Machmer and Ydenberg, 
1989). Compared to 2000–2001, nest density slightly increased at all study sites 
(Baltimore Harbor/Patapsco River 1 nest/2.30 km2 in 2000 vs. 1 nest/1.60 km2 current 
57 
study; Anacostia/middle Potomac 1 nest/4.7 km2 vs. 1 nest/3.9 km2; Elizabeth River 1 
nest/4.7 km2 in 2001 vs. 1 nest/3.57 km2 in 2012). 
It has been suggested that the collection of a sample egg may bias productivity 
rates. For some datasets, a difference has been noted in productivity between sampled 
and unsampled nests, and some investigators (Henny and Kaiser, 1996) adjust 
productivity to account for sample egg collection. However, in other studies no 
differences in productivity between sampled and unsampled nests have been noted 
(Rattner et al., 2004; Toschik et al., 2005). Nests at the Poplar Island reference site from 
which a sample egg was collected fledged 1.38 young, while unsampled nests from this 
site fledged 1.31 young. Unlike Henny and Kaiser, 1996 and Henny et al., 2004, the 
present study limited sampling to nests containing three or more eggs, which may 
account for this difference. 
 
 
Eggshell thickness 
 
Eggshell thickness in Chesapeake Bay ROCs was on average 0.514 ± 0.054 mm, 
which is close to the average pre-DDT-era value (0.505 mm, n = 365, Anderson and 
Hickey, 1970). In the 1970s, eggshell thickness on the Potomac River was about 19% 
lower (0.402–0.416 mm) than the pre-DDT-era value (Wiemeyer et al., 1988). In the 
1970s, Wiemeyer et al., 1988 found an association (r = −0.70) between p,p’-DDE 
residues in osprey eggs and eggshell thickness. Notably, concentrations of organochlorine 
pesticides and metabolites (Table 2) are now well below the threshold associated with 
10% eggshell thinning (Wiemeyer et al., 1988). In fact, no significant relationship was 
observed between concentrations of p,p’-DDE and eggshell thickness in the 2011–2012 
58 
dataset. 
 
 
Contaminant exposure in regions of concern 
 
Residues of p,p′-DDE on the middle Potomac River declined about 75% between 
1971–1977 and 2011 (3.1 μg/g ww in the 1970s, Wiemeyer et al., 1988; 1.16 μg/g ww in 
2000, Rattner et al., 2004, 0.73 μg/g ww, present study). Concentrations of p,p′-DDE in 
eggs collected in other ROCs were low and did not seem to change in the past decade 
(Baltimore Harbor 0.443 μg/g ww in 2000 versus 0.553 μg/g ww in 2011; Elizabeth 
River 0.660 μg/g ww in 2001 versus 0.488 μg/g ww in 2012). Mean p,p′-DDE 
concentrations in Chesapeake Bay (<0.725 μg/g ww; Table 2) remain lower than those 
reported in the most industrialized and urbanized segment of Delaware River between 
C&D canal and Easton, PA (means < 1.77 μg/g ww; Toschik et al., 2005). 
From a temporal perspective, PCB residues in osprey eggs from Baltimore 
Harbor/Patapsco and Elizabeth Rivers remained unchanged over the decade. Notably, 
eggs collected from Curtis Creek (Baltimore Harbor) in 2011 contained the greatest PCB 
residues (12.9 and 35.0 μg/g ww), and were comparable or exceeded values reported 
across the Northeast U.S. in the 1970s (up to 23 μg/g ww; Wiemeyer et al., 1988). 
However, a 40% decline in average total PCB concentrations was observed in eggs from 
the Anacostia/middle Potomac ROC (Figure 2). This suggests that PCB levels or their 
bioavailability in the osprey food web on the Anacostia/middle Potomac Rivers have 
decreased over the past decade. In support of this hypothesis, a reduction in total PCB 
concentrations in sediments has been reported for the Anacostia, where residues have 
declined by an order of magnitude over the past 25 years (Velinsky et al., 2011). Due to 
59 
moderate biomagnification factors from fish to osprey eggs (e.g., 11 times; Henny et al., 
2003), these changes in chlorinated biphenyl concentrations in sediments may be 
translated up the food web. 
Coplanar PCBs have been associated with embryonic deformities and adverse 
reproductive effects along with other toxic responses in fish-eating birds (Rice et al., 
2003;  Su et al., 2014). There was no evidence of embryonic deformities or reproductive 
impairment in the present study. Despite differences in analytical methods and detection 
limits between this current study and Rattner et al. (2004), all TEQs were well below the 
threshold associated with cytochrome P450 induction in osprey nestlings (Elliott et al., 
2001). 
Over the past decade, congener patterns of PBDEs did not change dramatically, 
with BDE 47, 99, 100, 153 and 154 being the most prominent congeners in osprey eggs 
across all sites. However, total PBDEs declined by about 40%. This is similar to a 55% 
decline reported by Henny and coworkers (2011) for osprey eggs collected on the 
Willamette River, Oregon. Since the last ecotoxicological study in the Chesapeake, the 
penta- and octa-BDE formulations have been phased out of use in the U.S. (end of 2004) 
(US EPA, 2014a).  More recently, use of the deca-formulation has been curtailed (US 
EPA, 2014b). While the manufacturer and use of PBDEs in new consumer products has 
been banned, many long-lived goods containing these compounds remain in use and end 
up in landfills. Thus, it is unknown how quickly environmental levels may change in the 
food web in response to regulatory action (Chen and Hale, 2010; Zhang et al., 2014). 
Notably, PBDE residues in osprey eggs collected along the Anacostia/middle 
Potomac Rivers in 2000 were over four times greater than those found at the reference 
60 
site (Rattner et al., 2004), with the maximum values detected near the Blue Plains 
WWTP. Other studies (Henny et al., 2011) have shown that volume of discharge and 
distance from a WWTP may be reflected in contaminant concentrations in osprey eggs 
and are presumably responsible for the high residues of PBDEs found in eggs near Blue 
Plains. 
The detection of PBDEs in osprey eggs in ROCs in 2000–2001 stimulated a series 
of studies examining the effects of flame retardants on raptorial birds (e.g., Fernie et al., 
2005; Fernie et al., 2006; Fernie et al., 2008; McKernan et al., 2009; Rattner et al., 2013). 
These studies provided a context for placing field residue data into perspective. In 
controlled exposure studies, the presence of PBDE flame retardants in developing 
American kestrel (Falco sparverius) embryos have been associated with oxidative stress, 
DNA damage, delay in hatch time, shorter humerus length, reduced total thyroid weight 
and reproductive and courtship behavior changes in adults. In field studies, Chen et al. 
(2010) have estimated a biomagnification factor of 25.1 for total PBDEs in the osprey 
food web. Reduced osprey productivity was associated with PBDEs at concentrations 
exceeding 1000 ng/g ww in eggs from Oregon and Washington (Henny et al., 2009). 
These findings were suggested to be equivocal in a subsequent study (Henny et al., 
2011). In the Chesapeake, the greatest residues of total PBDEs in osprey eggs 
(0.928 μg/g in 2000 and 0.802 μg/g ww in 2011) approached the lowest-observed-
adverse-effect-level (LOAEL) of 1.8 μg/g ww for pipping and hatching success                            
(McKernan et al., 2009), but there was no evidence of impaired hatching success or 
overall productivity in Chesapeake Bay ospreys. Many uncertainty factors (differences in 
species sensitivity and metabolism, exposure to mixtures of toxicants) and knowledge 
61 
gaps for toxicity thresholds make extrapolations from the lab to the field difficult. 
In place of PBDE formulations, alt-BFRs are now being manufactured and there 
are concerns regarding their potential persistence and bioaccumulation (de Wit et al., 
2011 and Stieger et al., 2014). Like PBDEs, these BFRs can leach into the environment 
and have been detected in piscivorous birds (Glaucous gulls Larus hyperboreus, 
Verreault et al., 2007; herring gulls Larus argentatus, Gauthier et al., 2007; white-tailed 
sea eagles Haliaeetus albicilla, Eulaers et al., 2014). Among the alt-BFRs, HBCD has 
been shown to cause endocrine and behavioral effects in captive American kestrels 
(Marteinson et al., 2011; 2012). Concentrations of HBCD in Chesapeake Bay osprey 
eggs were much lower than the HBCD toxicity thresholds determined to date. In this 
current study, α-HBCD was detected most frequently in osprey egg samples, but at much 
lower concentrations than other locations, including the Norwegian Arctic (7.23–
63.9 ng/g ww, Verreault et al., 2007 versus < MDL-3.72 ng/g ww in the present study). 
Our results reveal that mixtures of flame retardants are present in egg samples from the 
Chesapeake Bay, but the biological significance of these levels is unclear. 
 
 
DNA damage 
 
This study documents evidence of mild DNA damage in Baltimore 
Harbor/Patapsco and Anacostia/middle Potomac Rivers compared to the reference site, 
and also variations of 8-OH-dG concentrations between individuals. Xenobiotics 
including PCBs, PBDEs and even PAHs are known to increase reactive oxygen species in 
cells and can damage lipids, proteins and DNA. Mild oxidative stress has been detected 
in American kestrels (and to a lesser degree in common terns) in response to embryonic 
62 
exposure to the penta-BDE commercial mixture DE-71 (Fernie et al., 2005, McKernan 
et al., 2009 and Rattner et al., 2013). Osprey nestlings are exposed to a multitude of 
stressors, and may have adapted antioxidant defense mechanisms. Life history traits 
including trophic level and life span may influence their capability to tolerate oxidative 
stress leading to large variations in responses even between individuals (Costantini, 
2008). Current evidence suggests that oxidative stress may impair immune function, 
longevity and reproduction (Costantini, 2008). Ultimately, damage to DNA may make 
the individual susceptible to disease progression and epigenetic effects that may be 
reflected in later generations (Cooke et al., 2003). 
 
 
Conclusions 
 
The present findings indicate that organochlorine pesticides and PBDE flame 
retardants continue to decline in osprey eggs in Chesapeake Bay, consistent with 
discontinuation of their use. Low levels of organochlorine pesticides were detected at all 
sites, well below the threshold associated with 10% eggshell thinning. Interestingly, 
PBDE residues in eggs were greatest near the Blue Plains WWTP on the Potomac River, 
which suggests that wastewater discharge and sewage sludge are a source of continued 
input of PBDEs to the environment (LaGuardia et al., 2010). While total PCB 
concentrations declined at some sites, they remain high in Baltimore Harbor/Patapsco 
River and continue to drive fish consumption advisories in Chesapeake Bay (MDE, 2014; 
VDH, 2009). Overall, there were no apparent large-scale effects of the contaminants 
examined here on osprey productivity. The Bay osprey population continues to increase 
and productivity rates are at or have exceeded the threshold to maintain a stable 
63 
population. Since reaching their nadir in the 1970s the Chesapeake Bay osprey 
population has demonstrated its resilience in the face of anthropogenic threats. Although 
osprey productivity appears stable at the population level, DNA damage assays suggest 
more subtle biochemical and cellular effects that could have consequences on the fitness 
of some individuals. 
 
 
 
 
 
 
 
 
 
 
64 
Supplemental Table S1.  Concentrations of p,p’-DDE, total PCBs and PBDEs in archived eggs analyzed in 2000-2001 and re-
analyzed in 2011-2012. 
Sample ID Site 
Analytical 
Laboratory 
p,p'-DDE  
(µg/g ww) 
p,p'-DDE  
Percent 
Difference 
Total PCBs  
(µg/g ww) 
Total PCBs  
Percent 
Difference 
Total PBDEs 
(ng/g ww) 
Total PBDEs 
Percent Difference 
BHRC3-1 
Baltimore 
Harbor/Patapsco/River 
Hale 2011 0.926 
56.2 
7.75 
38.3 
243 
11.5 
GERG 2000 0.519 11.4 272 
BHCC1-1 
Baltimore 
Harbor/Patapsco/River 
Hale 2011 0.453 
35.6 
5.07 
4.14 
394 
15.5 
GERG 2000 0.316 4.86 337 
BHORB1-1 
Baltimore 
Harbor/Patapsco/River 
Hale 2011 1.955 
110.2 
16.6 
68.7  
 GERG 2000 0.566 8.10 
 
BHBP7-1 
Baltimore 
Harbor/Patapsco/River 
Hale 2011 0.758 
49.9 
7.94 
12.0  
 GERG 2000 0.455 7.04 
 
BHSC5-1 
Baltimore 
Harbor/Patapsco/River 
Hale 2011 0.790 
49.3 
10.3 
2.16  
 GERG 0.478 10.0 
 
PRSNRL-1 
Anacostia/middle Potomac 
Rivers 
Hale 2011 1.100 
21.2 
7.42 
10.0  
 GERG 2000 0.889 8.20 
 
PRDC-2 
Anacostia/middle Potomac 
Rivers 
Hale 2011 1.720 
26.8 
7.01 
30.6  
 GERG 2000 1.314 9.55 
 
PR88-1 
Anacostia/middle Potomac 
Rivers 
Hale 2011 1.513 
26.2 
9.44 
4.58  
 GERG 2000 1.163 9.02 
 
ERWBP-3 Elizabeth River 
Hale 2012 0.650 
36.8 
4.89 
30.8 
149 
26.6 
GERG 2001 0.448 3.58 195 
ER8-1 Elizabeth River 
Hale 2012 1.370 
32.2 
11.6 
41.1 
174 
35.3 
GERG 2001 0.990 7.65 248 
CINW-1 Elizabeth River 
Hale 2012 0.911 
42.8 
1.12 
18.5 
153 
0.73 
GERG 2001 0.590 1.35 154 
 
aIn 2000-2001 osprey eggs were analyzed by the Geochemical and Environmental Research Group (GERG) at Texas A&M.  These eleven eggs were re-analyzed in 2011-2012 respectively 
by the lab of Dr. Robert C.  Hale at the Virginia Institute of Marine Science.  In 2000-2001, only a subset of the eggs were analyzed for PBDEs, which limited the number of samples (n = 5) 
that could be compared.  Percent difference = (absolute difference/mean) x 100%.  
65 
 
 
 
 
 
 
 
 
 
CHAPTER 3 
CHESAPEAKE BAY FISH-OSPREY (Pandion haliaetus) FOOD WEB: 
EVALUATION OF CONTAMINANT EXPOSURE AND GENETIC DAMAGE 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66 
Introduction 
Ospreys (Pandion haliaetus) are a well-known ecotoxicological sentinel species. 
Many studies have utilized this charismatic fish-hawk as a bioindicator to increase 
knowledge of spatial and temporal trends in contaminants (Golden and Rattner, 2003; 
Grove et al., 2009).  For example, ospreys feed at a high trophic level and biomagnify 
lipophilic contaminants.  The Chesapeake Bay provides breeding habitat to the largest 
osprey population in the world due its shallow waters and high productivity (Poole, 
1989).  
 The decline and later recovery of the Chesapeake Bay osprey population was 
associated with the patterns of use of the organochlorine pesticide DDT, in particular, the 
p,p’-DDE metabolite that induces eggshell thinning (e.g., Wiemeyer et al., 1975; Spitzer, 
1980; Poole, 1989; Audet et al., 1992; Watts et al., 2004).  While several studies have 
examined trends in osprey productivity in relation to contaminants, few have addressed 
this issue from an integrative food web perspective (Henny et al., 2003, 2009; Chen et al., 
2010).  Studies of osprey diet and foraging ecology in the Chesapeake are limited (e.g., 
McLean and Byrd, 1991; Glass and Watts, 2009).  These investigators documented that 
energy rich menhaden (Brevoortia tyrannus) was a dominant osprey prey item in high 
salinity sites such as the lower James River.  However, ospreys nesting in low salinity 
sites on the upper James and York Rivers primarily consumed catfish (Ictaluridae) 
species and gizzard shad (Dorosoma cepedianum).  
 The last large-scale ecotoxicological study of ospreys in the Chesapeake Bay was 
conducted in 2001-2002 (Rattner et al., 2004), and focused upon the United States 
Environmental Protection Agency (US EPA) designated Regions of Concern (ROCs; viz., 
Anacostia/middle Potomac, Baltimore Harbor/Patapsco and Elizabeth Rivers).  In order 
67 
to meet the requirements described in the Chesapeake Bay Executive Order 13508 
(2009), federal agencies expanded their research and monitoring efforts to generate 
strategies to reduce toxics in the major tributaries of the Chesapeake.  This entailed 
assessments of contaminant exposures and effects on fish and wildlife.  In 2011-2012, we 
re-evaluated contaminants in ospreys nesting in Chesapeake Bay ROCs and found that 
concentrations of halogenated pollutants in eggs had declined, but there remained 
evidence of genetic damage in nestlings found in the most industrialized regions (Lazarus 
et al., 2015a).   
There are limited ecotoxicological data for wildlife for the northern regions of the 
Bay.  The Susquehanna is the largest freshwater inflow to the Chesapeake Bay (Rattner 
and McGowan, 2009).  Between 2011 and 2013, we studied contaminant exposure, food 
web transfer and potential effects on ospreys nesting in the Susquehanna, James and 
Potomac Rivers.   Due to the distribution of osprey nests on the Susquehanna River, 
sampling efforts focused on the lower Susquehanna River and flats.  In addition, 
contaminant exposure was monitored in Back River, Maryland (northeast of Baltimore 
Harbor and the Patapsco River).  Relative to other sites in the Chesapeake Bay, sediments 
from Back River contain very high concentrations of PCBs (McGee et al., 1999; 
Klosterhous, 2007), and preliminary observations (Lazarus unpublished data) indicated 
poor osprey productivity in the vicinity of the Back River wastewater treatment plant 
(WWTP).  Herein, we describe findings of a study measuring contaminant concentrations 
and their transfer (biomagnification) between whole fish and osprey eggs.  Reproductive 
success, eggshell thickness and oxidative genetic damage in nestlings were employed to 
68 
investigate contaminant-related effects at various biological levels (population, individual 
and molecular) among several Chesapeake Bay tributaries.   
 
Materials and Methods 
Study sites  
From 2011 to 2013, three major Chesapeake Bay tributaries (lower Susquehanna 
River and flats, middle Potomac and James Rivers), Back River and Poplar Island 
reference site were studied (Figure 6).  In 2011, whole fish and osprey eggs and nestling 
blood samples were collected along a 45 km stretch of the (i) Anacostia/middle Potomac 
River (Frederick Douglass Bridge, Washington, District of Columbia to Mattawoman 
Creek, Maryland).  In 2012, similar sampling was conducted along a 60 km stretch of the 
(ii) James River (Richmond to Milton, Virginia).  Finally, in 2013, sampling was 
undertaken along a 20 km stretch along the (iii) Susquehanna River and flats (from 
Aberdeen, Maryland to the I-95 Millard E Tydings Memorial Bridge). In 2013, we also 
studied an 11 km stretch on (iv) Back River (Back River Wastewater Treatment Plant 
(WWTP) to Hart Miller Island).  Based on results from previous studies, the (v) Paul S. 
Sarbanes Ecosystem Restoration Project at Poplar Island, Maryland (Poplar Island), a 
mid-Bay location, was used as a reference site for all three years (Rattner et al., 2013; 
Lazarus et al., 2015a; 2015b).   
 
Osprey reproduction and foraging activity  
All procedures involving fish and ospreys were conducted under approval of the 
Institutional Animal Care and Use Committees of the U.S. Geological Survey’s Patuxent 
69 
Wildlife Research Center (USGS/PWRC) and the University of Maryland, and with 
appropriate federal and state scientific collection permits.  Starting in late-March, osprey 
nests were visited every 7-10 days to determine the number of eggs laid, eggs hatched 
and young present at ≥ 40 days.  These data were used to calculate productivity 
(Mayfield, 1961; Postupalsky, 1977; Steenhof and Newtown, 2007).  Additional nests 
were monitored at each site as a potential source of nestling blood samples in the event 
that the selected study nest failed.  
During the osprey reproductive period, dietary preferences of nesting adults were 
monitored using a variety of techniques (Johnson et al., 2008; Glass and Watts, 2009).  
Game camera (Bushnell 8MP Trophy Cam, Overland Park, KS, USA) images of prey 
items captured by osprey, direct identification of fish scraps found in osprey nests, and 
direct observations of prey deliveries (photo documented Nikon D3100 DLSR camera, 
AF VR-Nikkor 80-400mm lens, Nikon, Tokyo, Japan) were used to characterize osprey 
diet.  Using these observations, we determined the 2-3 dominant prey items (based on the 
highest percent catch) to sample for the food web component of this study. 
 
 
 
 
 
 
 
 
70 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Osprey egg sample collection 
Using egg sampling technique described by Blus (1984), upon completion of a 
clutch (3 or more eggs) one fresh egg was randomly collected from each study nest 
(Susquehanna, n=10; Anacostia/middle Potomac, n=13, James River, n=12; Back River 
n=5; Poplar Island, n=4/year).  Eggs were transported to the U.S. Geological Survey 
Patuxent Wildlife Research Center (USGS/PWRC), cleaned, weighed and the length and 
Figure 6.  Locations of osprey nests sampled from in Chesapeake Bay and Poplar 
Island reference site, (•) indicates a sampled nest.  
 
71 
width were measured to the nearest 0.01-mm (Lazarus et al., 2015a).  A 2.4-mm hole was 
drilled into the blunt end of the egg (Dremel® MultiPro® 7.2V, Model 770, Mt. Prospect, 
IL, USA) and distilled water was injected into the air cell to return contaminant 
concentrations to that of a freshly laid egg (Heinz et al., 2009).  Each egg was opened, 
contents (excluding shell membrane) were transferred to a chemically clean jar (I-CHEM, 
VWR Scientific, Radnor, PA, USA), examined and weighted, and then stored -80ºC.  
Shells were dried for 3-4 months at room temperature and thickness measurements were 
taken at three points along the equator using a micrometer (Model 1010M; L.S. Starrett 
Co., Athol, MA, USA). 
 
Nestling blood samples and morphological endpoints 
 Osprey nestling blood samples were collected from 40-45-day-old young at each 
sampled nest or a nearby nest in the event of reproductive failure (n=46 accessible nests 
from the Susquehanna, Potomac and James Rivers and Poplar Island; n=3 nestlings for 
Back River because there were no nearby replacement nests).  Briefly, one nestling/nest 
was removed for approximately 10 min.  After physical examination, body weight, crop 
contents (size of food contents present in crop determined via palpation) and culmen 
length were measured.  A 5-7 mL brachial blood sample was collected using a 23-gauge 
25.4 mm needle into a heparinized syringe (Sarstedt International, Newton, NC, USA).  
About 100 µL of blood was immediately transferred to a microcentrifuge tube, frozen on 
dry ice and stored at -80ºC for subsequent DNA damage assays.  The remainder of the 
blood sample was centrifuged, and plasma was harvested for pharmaceutical (Lazarus et 
al., 2015b) and stable isotope analyses. 
72 
Fish sampling 
 In each year of the study, fish sample collection was undertaken in early July, 
with target fish ranging from 25-35 cm in length (preferential prey size for ospreys; Poole 
et al., 1989).  On the Anacostia/middle Potomac and the James Rivers, fish were caught 
by electroshocking.  On the Susquehanna, a combination of electroshocking and hook 
and line were used to capture live fish.  At Poplar Island, fish were caught using a 
midwater trawl and a commercial pound net.  Fish (n=201) were collected and stored at -
20ºC, and then composited (n=3 fish/composite) by location, species and size.  Due to the 
limited number of osprey nests on Back River, only a small sampling of menhaden 
(Brevoortia tyrannus), gizzard shad (Dorosoma cepedianum) and white perch (Morone 
Americana) were collected.  
 
Chemical analysis of osprey eggs, whole fish and quality assurance  
Whole fish and osprey egg contents were analyzed for 129 PCB congeners, 44 
organochlorine pesticides, methoxytriclosan, 11 PBDE congeners, and 5 alternative 
brominated flame retardants [(alt-BFRs: α, β, γ hexabromocyclododecane (HBCD), 1,2-
bis (2,4,6-tribromophenoxy)ethane (BTBPE), di(2-ethylhexyl)-2,3,4,5-
tetrabromophthalate (TBPH), 2-ethylhexl 2,3,4,5-tetrabromobenzoate (TBB), and 
decabromodiphenyl ether (DBDPE)].  During analysis for PBDEs, one osprey egg 
sample from the Back River was lost during sample processing (n reduced from 5 to 4).  
Chemical analyses were conducted at the Virginia Institute of Marine Science 
(VIMS) following the methods described in Lazarus et al. (2015b), Chen et al. (2008) and 
La Guardia et al. (2007, 2010, 2013).  Briefly, egg contents were homogenized, 
73 
lyophilized and spiked with surrogate standards PCB 30, 65 and 204 (Ultra Scientific, 
North Kingstown, RI, USA), 13C-methoxytriclosan, 13C-PCB-126 and 2,3,4,4’,5,6-
hexabromodiphenyl ether (BDE-166; Cambridge Isotope Laboratories, Inc., Andover, 
MA, USA).  Sodium sulfate blanks were analyzed.  Dried samples underwent solvent 
extraction using methylene chloride (DCM) at 100ºC and 68 atm.  Extracts were purified 
by size exclusion chromatography (SEC, Envirosep-ABC®, 350 x 21.1 mm column; 
Phenomenex, Torrance, CA, USA).  Each post-SEC extract was reduced in volume, 
added to the top of a 2-g silica gel glass solid phase extraction column  (Isolute, 
International Sorbent Technology, Ltd., Hengoed Mid Glamorgan, UK) and eluted with 
3.5-mL hexane (to waste), followed by 6.5 mL of 60:40 hexane/DCM  and 8 mL DCM.   
The latter two fractions were combined and then split, with half going for coplanar PCB 
analysis.  Coplanar PCBs were separated from nonplanar PCBs by elution through a 
Supleclean ENVI-Carb SPE column (Sigma-Aldrich Co., St. Louis, MO, USA).   The 
column was first eluted with 15 mL hexane (to waste).  The coplanars were obtained by 
elution with 20 mL hexane/toluene (99:1) and 20 mL toluene.  The pooled eluent was 
reduced in volume, spiked with p-terphenyl (Ultra Scientific, North Kingstown, RI, USA) 
as an internal standard, and analyzed by gas chromatography/mass spectrometry 
(GC/MS) on an Agilent 5975C (Agilent Technologies, Santa Clara, CA, USA), in 
electron impact mode and selected ion monitoring.  A 60 m DB-5 MS GC column 
(Agilent, 0.32 mm ID x 0.1 µm thickness) was used.  
The second half of the silica SPE fraction retained was spiked with 
decachlorodiphenyl ether (Ultra Scientific) as the internal quantitation standard.  
Identification and quantitation of non coplanar PCBs was conducted by GC/MS in the 
74 
electron ionization mode on a Varian 2200 GC/MS (Varian now owned by Agilent 
Technologies).  The organochlorine pesticides and methoxytriclosan were analyzed 
similarly by GC/MS on a Varian 4D MS.  Both analyses used 60 m DB-5 columns (0.32 
mm ID x 0.25 µm thickness).  PBDEs and alt-BFRs were separated by ultra-performance 
liquid chromatography (UPLC, Waters Corp. Milford, MA, USA) and analyzed by 
atmospheric pressure photoionization tandem mass spectrometry (APPI/MS/MS, Q-
Trap3200 MS, AB Sciex, Framingham, MA. USA).  Further details of the UPLC-
APPI/MS analysis can be found in La Guardia et al. (2013).  
Method detection limits on a wet weight basis were converted to a lipid weight 
basis using the following formulas: MDL dry = MDL wet/(1-% water) and MDL lipid = 
(MDL dry/% lipid).  Percent lipid and percent moisture used in this calculation were from 
the lowest samples to allow for the most conservative estimates.  The method detection 
limit (MDL) for organochlorine pesticides and non-coplanar PCB congeners for osprey 
eggs was 0.4 µg/kg wet weight (ww) and 11.9 µg/kg on a lipid weight (lw) basis and for 
fish was 0.2 µg/kg and 7.2 µg/kg lw. The MDL for coplanar PCB congeners was 0.04 
µg/kg ww and 1.19 µg/kg lw in osprey eggs and 0.02 µg/kg and 0.72 µg/kg lw in fish.  
The MDL for PBDEs and alt-BFRs was 0.4 µg/kg ww in eggs or 11.9 µg/kg lw and 0.2 
µg/kg ww and 7.2 µg/kg lw in fish.   
All data were corrected based on the recovery of surrogate standards in each 
sample and corrected for moisture loss back a to fresh weight basis.  The average 
recovery of the surrogate standard PCB 204 and 13C-methoxytriclosan from eggs was 
(mean ± SD) 81.8 ± 14.0% and 84.7 ± 10.7% in fish.  Mean recovery for non-coplanar 
PCBs (surrogate standard PCB 204) was 87.9 ± 18.0% in eggs and 84.0 ± 21.2% in fish.  
75 
For coplanar PCBs (surrogate standard PCB 126) average recovery was 87.8 ± 20.7% 
and 96.1 ± 24.9% in fish.  Average recovery of PBDEs (surrogate standard BDE 166) 
was 101.1 ± 25.0% and 104.4 ± 30.7% in fish.  Overall, mean moisture content in eggs 
averaged 83.9 ± 0.52% and in fish 74.7 ± 3.54%.   
 
Biomagnification factors      
 Biomagnification factors (BMFs) were calculated by relating the concentrations 
of detected chemicals in prey (whole fish) to those in the predator (osprey).  Beyer and 
Biziuk (2009) present a simple formula for calculation of BMFs as the concentration of 
the chemical in the organism (CB) to the concentration of the chemical in its prey (CA): 
BMF= CB/CA.  In order to calculate BMFs in this study, we applied the model presented 
in Elliott et al. (2005) and used in Chen et al. (2010).  This equation adjusts residues 
based on diet composition: 
Y=BMF [(F1(X1) + F2(X2) +…+ Fn (Xn)]    (1) 
Y represents the geometric mean contaminant concentration in the predator, Fn represents 
the percentage of each prey species in the diet and Xn reflects the geometric mean 
contaminant residue per species consumed.  Biomagnification factors were calculated 
using both ww and lw basis for the major groups of contaminants analyzed.   The lipid 
weight of each sample was calculated by dividing the dry weight recovery corrected 
values by the % lipid.  For those contaminant residues presented as a Kaplan Meier range 
of means, the minimum and maximum BMFs were calculated.   
 
 
76 
Stable isotopes  
Stable isotope analyses were performed at Colorado Plateau Stable Isotopes 
Laboratory (CPSIL) at Northern Arizona University to determine 13C and 15N content.  
Nitrogen stable isotopes indicate relative trophic level for the local food web. 
Approximately 1 mL of osprey nestling plasma was freeze dried and analyzed using a 
Thermo-Electron Delta V Advantage Isotope Ratio Mass Spectrometer (IRMS) (Thermo 
Fisher Scientific Inc., Waltham, MA, USA). The IRMS was configured through the 
Finnigan CONFLO III (Thermo Finnigan, Barkhausenstr, Germany) using a Carlo Erba 
NC2100 elemental analyzer CE Elantech, Inc., Lakewood, NJ, USA).  Carbon and 
nitrogen stable isotope compositions were obtained in a single run.  CPSIL uses 
biological standards for calibration and raw data normalization from the National 
Institute of Standards and Technology and the International Atomic Energy Agency 
(Mauldin et al., 2012).  Stable isotope values for δ13C were reported permille (parts per 
thousand) (‰) according to the Vienna Pee Dee belemnite standard and δ15N were 
reported relative to atmospheric air (AIR).  CPSIL has uncertainty factor of ≤ 0.10 ‰ for 
δ13C and ≤ 0.20 ‰ for δ 15N.   
 
DNA damage assays 
Whole blood samples (n=45, one excluded from the James River due to 
consistently high % coefficient of variation after multiple re-runs) were analyzed for 8-
hydroxy-2’-deoxyguanosine (8-OH-dG), as an indicator of DNA damage (assay methods 
and validation also described in Rattner et al., 2013; Lazarus et al., 2015a).  Briefly, 
samples were analyzed using the DNA/RNA oxidative damage EIA kit (Cayman 
77 
Chemical, Ann Arbor, MI, USA).  Plates included blanks and all samples were analyzed 
in duplicate.  Standard curves were fit using a 4-parameter model (R2>0.998; MARS 
Data Analysis Software 2.10; BMG Labtech, Cary, NC, USA).  Intra-assay variation 
(precision of duplicate determinations, coefficient of variation, CV ± SD) was 3.5 ± 3.0% 
in 2011 and 7.4± 4.2% for samples collected in 2012 and 2013.  Any samples with a 
CV>20% were re-analyzed.  Inter-assay variation among plates for reference samples 
was 4.1 ± 11.5% in 2012 and 9.93 ± 11.3% in 2014 assays.  Those samples collected in 
2011 on Poplar Island and the Anacostia/middle Potomac Rivers and those samples from 
the Susquehanna and James Rivers were analyzed separately in 2014.  However, due to 
variations in performance between manufacturing lots we were unable to quantitatively 
compare all study sites.  The limit of detection for this assay was 1.03 pg/µg DNA 
determined by evaluating the mean minus 3 standard deviations from the standard curve 
(Rattner et al., 2013).  
 
Statistical analyses 
Descriptive statistics were generated for continuously distributed variables 
(eggshell thickness, morphological endpoints, DNA damage and contaminant residues).  
Variables were tested for normality, homogeneity of variance and log transformed as 
necessary (Quinn and Keough, 2002).  Analysis of variance was used to detect overall 
differences among sites, and specific comparisons were conducted using Tukey’s HSD 
test (α=0.05).  If the assumptions for an ANVOA were not met, Wilcoxon non-parametric 
statistics were used followed by a Bonferroni correction to adjust for multiple 
comparisons.  For contaminants with residues < MDL in <50% of samples, the Kaplan-
78 
Meier method was used to estimate the extremes of the mean (Helsel, 2005).  Fisher’s 
Exact Test was used to compare site-specific differences in productivity endpoints.  In 
osprey eggs, concentrations of Ah-receptor active PCB congeners were multiplied by 
toxic equivalency factors (TEFs) to estimate toxic equivalence (TEQs, van den Berg et 
al., 1998).  A correlation analysis was conducted to examine relationships among all 
variables.  A logistic analysis of convariance was first used to examine site specific 
relationships between egg residues (p,p’-DDE, PCBs and PBDEs) and osprey nest 
success.  If there were no site specific differences data were combined to evaluate overall 
differences in productivity and contaminant residues. 
 Redundancy analysis (RDA) was conducted using data from Lazarus et al. 
(2015a) combined with this current study to increase statistical power.  All predictor 
variables (egg contaminant concentrations, DNA damage and stable isotope data) were 
log transformed to correct for normality and homogeneity of variances (Quinn and 
Keough, 2002). The δ13C isotope data were –log(x) transformed because values were 
negative.  Redundancy analysis (RDA; Legendre and Anderson, 1999) was used to assess 
if concentrations of PCB, PBDE, p,p’-DDE, DNA damage and carbon and nitrogen stable 
isotope measurements differed by study site: 
DNA + PCB + PBDE + DDE + C + N ~ Site  (2) 
We used RDA and not a more generalized distance based-RDA because Euclidian 
distance was appropriate for our data.  We specifically examined if axes explained more 
variability than would be expected by chance alone using a permutation-based ANOVA.  
Year was used as a blocking factor since DNA damage assays from 2011 versus 2012 
and 2013 were not comparable (Lazarus et al., 2015a).  This analysis was conducted 
79 
using the RDA function from the Vegan package in R (R. Core Team, 2013; Oksanen et 
al., 2011).  Interaction terms between contaminant concentrations were also evaluated 
and then a mixed effects linear regression was used to examine if any contaminant data 
affected DNA damage (sample site random effect).    
 
Results 
Productivity 
Productivity for the five study sites ranged from 1.17 to 1.33 fledglings/active 
nest (Table 5).  There were no site-specific differences in productivity (eggs laid, eggs 
lost, hatching, fledging and nest success) (Fisher’s Exact Test p>0.60, Table 1) among 
the Susquehanna River and flats, Potomac, James Rivers and Poplar Island.  On average, 
of the 47 nests sampled, 3.10 eggs were laid per nest, 68.5% hatched, 96.9% of the 
nestlings that hatched fledged, 77.7% of the active pairs fledged young. On Back River, 
there were only 5 active nests.  Hatchability was adequate (83.3%) and fell within range 
of the other study sites (78.9-90.0%).  However, of these intensively studied nests, only 1 
fledgling was produced/active nest and 60.0% of the successful pairs fledged young.   
 
80 
 
 
 
 
 
 
 
 
 
 
Table 5. Reproductive success of ospreys nesting in Chesapeake Bay regional waterways of 
concern. 
81 
Osprey eggshell thickness 
Eggshells were thinner on the Anacostia/middle Potomac Rivers (p=0.003) 
compared to the reference site (Anacostia/middle Potomac Rivers, mean ± S.D.:  0.49 ± 
0.05 mm vs. Poplar Island reference site: 0.55 ± 0.05).  However, there were no 
differences in eggshell thickness among the other sites (Susquehanna: 0.52 ± 0.03; James: 
0.52 ± 0.04 mm) compared to the Anacostia/middle Potomac and Poplar Island reference 
site.  Back River had significantly thicker eggshells compared to the Potomac River  
(0.55 ± 0.03 mm; p=0.048). 
 
Nestling body weight and culmen length  
 There were no significant differences in body weight and culmen length among 
sampling years (2011, 2012, and 2013) on Poplar Island (p > 0.49), and thus reference 
site measurements were combined.  Overall, there were no differences in osprey nestling 
body weight across all study sites (p=0.18) and body weight averaged 1595.7 ± 156.7g.  
Culmen length did not vary among sites (p=0.34) and averaged 31.0 ± 1.1 mm. Although 
not included in statistics, four additional chicks on Back River were studied and their 
body weight averaged 1481.7 ± 28.8 g and culmen length averaged 28.6 ± 1.3 mm. 
82 
 
 
 
DNA damage 
 DNA damage assays were conducted in 2012 and 2014; however, due to 
variations in performance between assay kit lots, results cannot be quantitatively 
compared among years.  Results were generated for 48 nestling blood samples 
(CV<20%), with one sample from the James River excluded that consistently exhibited a 
poor precision in three separate assays.    
In 2011, nestlings from the Anacostia/middle Potomac River (33.9 ± 6.1 pg/µg 
DNA) exhibited greater DNA damage than those from the Poplar Island reference site 
Figure  7.  Concentrations of 8-hydroxy-2’-doxyguanosine (8-OH-dG) in osprey 
nesting whole blood. Capital letters indicate a statistically significant difference 
(p<0.05) using Tukey’s HSD test.  Means, standard deviations and individual values 
are presented (•). Data from 2011 and 2012/2013 were analyzed separately. 
 
83 
(26.6 ± 2.9 pg/µg DNA)  (p=0.04, Figure 7).  There was no difference (p=0.1532) in 
oxidative DNA damage in nestling blood samples from Poplar Island in 2012 and 2013, 
and thus data for these two years were combined (27.6 ± 12.6 pg/µg DNA).  There were 
no significant differences (p=0.09) in DNA damage in nestling blood samples from the 
Susquehanna, James and Back Rivers and Poplar Island reference sites in 2012 and 2013.   
Of the three nestling blood samples collected from the Back River, 8-OH-dG averaged 
46.1 ± 28.1 pg/µg DNA; the nestling sampled closest to the WWTP had the greatest 8-
OH-dG value (78.1 pg/µg DNA) of all samples analyzed in this study.  
 
Osprey diet characterization  
Over the three-year study, a total of 1,662 osprey prey items (15 fish species) 
were documented (Supplemental Table S2).  On Poplar Island osprey predominately fed 
on striped bass (Morone saxatilis) (47.8% of diet) and Atlantic menhaden (Brevoortia 
tyrannus) (44.3% of diet).  On the Susquehanna, Anacostia/middle Potomac and James 
Rivers osprey fed predominately on catfish (Ictaluridae sp.) and gizzard shad (Dorosoma 
cepedianum).  Mean 15N content in nestling plasma, which is a proxy for trophic level, 
was similar on the Anacostia River (19.4 ‰) and Poplar Island (18.2 ‰) (p=0.02, α = 
0.008 for Bonferroni Correction), but slightly lower on the Susquehanna (17.5 ‰) and 
James Rivers (15.3 ‰, p<0.006).  This indicates that ospreys are feeding at fairly similar 
trophic levels across study sites.  A complete reconstruction of osprey diet was not the 
specific goal of this study; however, these basic dietary observations were used to 
identify the fish species to sample for the food web component of this study.  
 
84 
Contaminants in fish 
 Organochlorine pesticides, PCBs, PBDEs, and alt-BFRs were measured in the 
dominant fish species in osprey diet (Supplemental Table S3).  Of the contaminants 
quantified, 19 of 44 organochlorine pesticides, 111 of 129 PCB congeners, 6 of 11 PBDE 
congeners and 5 of 7 alt-BFRs were detected in fish.  Fish from the Anacostia/middle 
Potomac Rivers had mean concentrations of p,p’- DDE (39.1 ng/g ww) that were about 4 
times greater than Poplar Island (9.24 ng/g ww) (p< 0.0001), with values on the 
Susquehanna and James Rivers in the intermediate range.  Similarly, fish on the 
Anacostia/middle Potomac Rivers contained the greatest total PCB residues (481.2 ng/g 
ww, ranging up to 1,145.2 ng/g ww compared to Poplar Island (49.1 ng/g ww, ranging up 
to 102.3 ng/g ww) (p< 0.0001).  Like p,p’-DDE, total PCBs on the Susquehanna and 
James exhibited intermediate values.  The only coplanar PCB congener detected in fish 
samples was PCB 77, and there were no differences among sites (p=0.72).  There were no 
differences in total PBDE residues in fish among study sites.  Alternative (alt)-BFRs were 
detected at low quantities across all study sites (all < 8.3 ng/g ww).  Qualitatively, p,p-
DDE residues were not unlike those on the Anacostia/middle Potomac Rivers, while total 
PCB and PBDE concentrations in gizzard shad were the greatest compared to the other 
study sites.  
 
Contaminants in osprey eggs 
Overall, there were significant relationships between contaminants and osprey 
productivity as indicated by logistic regression (p>0.22).  In ospreys, 24 of 44 
organochlorine pesticides, 110 of 129 non-coplanar PCB congeners, 4 of 4 coplanar PCB 
85 
congeners, 8 of 11 PBDE congeners and 5 of 7 alt-BFRs were detected in osprey eggs.  
In total, the same 19 organochlorine pesticides, 6 PBDE congeners, 4 alt-BFRs, 1 
coplanar PCBs and 77 non-coplanar PCBs were detected in both fish and osprey samples. 
Residues of organochlorine pesticides were greater on the Anacostia/middle Potomac 
Rivers for p,p’-DDE, cis-chlordane, cis-nonachlor, trans-nonachlor (p<0.02) compared to 
other study sites (Table 6).  Residues of p,p’-DDE in osprey eggs on the 
Anacostia/middle Potomac ranged up to 1.00 µg/g, which is 2.5 times greater than the 
maximum value on Poplar Island (0.414 µg/g ww).  The metabolite of the antimicrobial 
agent triclosan, methoxytriclosan, was detected at most study sites (Poplar Island 1/12, 
range <MDL-1.49 ng/g ww; Susquehanna River all <MDL; Anacostia/middle Potomac 
12/12, range 0.40-6.29 ng/g ww; James River 1/12, range <MDL-1.49 ng/g ww; and 
Back River 1/5, range <MDL-3.77 ng/g ww).  
Total PCBs (sum of non-coplanar plus coplanar congeners) were elevated at all 
study sites compared to Poplar Island (p<0.0001) (Table 7).  There were no significant 
differences in PCB concentrations in eggs among the other 3 sites (p>0.47). Upon 
inspection of the data, geometric mean PCB residues for Back River were similar to the 
other site means.  Congener 169 was detected most frequently on the James.  The TEQ on 
the Anacostia/middle Potomac Rivers (0.19 ng/g) was greater compared to the other 
study sites (p<0.025). 
Total PBDE concentrations were greatest on the Susquehanna and 
Anacostia/middle Potomac Rivers compared to other sites (p<0.002) (Table 8). The 
maximum PBDE residue (801.8 ng/g ww) was detected in vicinity of the Blue Plains 
WWTP on the middle Potomac River.  Congeners 47, 100, 153 and 154 followed a 
86 
similar pattern with values being higher on the Susquehanna and middle Potomac Rivers 
compared to the other study sites.  Upon inspection of the data, α-HBCD, BTBPE, 
DBDPE, TBB, and TBPH were most frequently detected on the Anacostia/middle 
Potomac Rivers with 5 of the 7-alt BRFs detected in osprey eggs.  Generally, alt-BFR 
values were two orders of magnitude lower than PBDE concentrations.   Beta (β) and 
gamma (γ)-HBCD were not detected in any osprey egg samples.   
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
87 
Table 6. Osprey egg concentrations (µg/g ww) of organochlorine pesticides and 
metabolites from Chesapeake Bay regional waterways and Poplar Island reference sitea. 
 
 
88 
Table 7. Osprey egg concentrations (µg/g ww) of polychlorinated biphenyls and 
congeners from Chesapeake Bay regional waterways and Poplar Island reference sitea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
89 
Table 8. Osprey egg concentrations (ng/g ww) of polybrominated diphenyl ethers and 
alternative brominated flame retardants from Chesapeake Bay regional waterways and 
Poplar Island reference sitea.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
90 
Biomagnification factors  
Biomagnification factors were used to relate contaminant residues in fish to those 
in osprey eggs (Table 9).  Using dietary observations, the ratio of predominant fish prey 
species on Poplar Island (menhaden and striped bass) and the Susquehanna (catfish and 
gizzard shad) was approximately 1:1.  On the Anacostia/middle Potomac, birds were 
feeding predominately on catfish and gizzard shad (ratio of 3:2), and prey species of 
ospreys on the James River were similar but at a slightly different ratio (3:1).  Using 
these proportions of prey species consumed by ospreys, the BMFs for osprey eggs (ww 
basis) averaged 16.5 for p,p’-DDE, 25.4 for PCBs, and about 17.9 for total PBDEs and 
15.4-15.5 for BDE congeners 47, 99, 100 and 154.  Values were approximately in the 
same ratios when calculated on a wet weight and lipid weight basis. Many of the 
compounds had a BMF <1 (trans-nonachlor, cis-nonachlor, cis-chlordane, trans-
chlordane, and methoxytriclosan) indicating no biomagnification in the upper trophic 
levels in Chesapeake Bay.   
 
91 
Table 9. Biomagnification factors (BMFs) and a wet weight (ww) and lipid weight (lw) from whole fish to osprey egg contents by 
study site. 
 
 
 
92 
Redundancy analysis 
The RDA incorporated data from samples from our previously published study 
(n=13; Lazarus et al., 2015a) and the present study (n=47; because there were only 
complete data for three nests on Back River, this site was excluded from the RDA, Figure 
8). The RDA model explained more variability than would be expected by chance alone 
(Fdf =5 =11.53, p<0.001).  The first two axes (Table S3, and Figure S3) explained more 
variability than would be expected by chance alone, Fdf =1 =46.89, p<0.001 and Fdf=1= 
6.99, p<0.001), and axis 3 was near the threshold for significance (Fdf=1=2.54, p=0.0667).  
A distinct grouping emerged among sites (Figure 8, Supplemental Figure S4).  Axis one 
had the largest contribution from PBDEs, PCBs and p,p’-DDE, while axis 2 had the 
largest contributions from PBDEs.  Thus, sites on the left of Figure 3 are most polluted.  
Those sites on the top left (Patapsco River/Baltimore Harbor, Elizabeth River) had more 
PCB and p,p’-DDE contaminations in osprey eggs, and those on the bottom left 
(Anacostia/middle Potomac Rivers and the Susquehanna) had more PBDEs (Axis 2).  
Overall, PBDEs appear most closely associated with DNA damage compared to the other 
chemicals.  Poplar Island (furthest to the right) had the least contamination, which is 
consistent with it being the reference site (Figure 8, Table S4 and Figure S1).  There were 
no univariate predictors for DNA damage (PCB tdf =48=0.445, p=0.658; PBDE tdf =48=-
0.087, p=0.931; p,p’-DDE tdf =48=0.915, p=0.365).   
 
 
 
93 
 
 
 
 
 
Discussion 
Osprey productivity 
 Due to the presence of the organochlorine pesticide DDT and its metabolites 
(p,p’-DDE) in the osprey food web, by the 1970’s the osprey population had contracted 
to the main stem of the Bay with few nesting pairs present  north of the Bay Bridge 
(Henny et al., 1974; Wiemeyer et al., 1975, 1988).  During the DDT use era, productivity 
Figure 8. Redundancy analysis (RDA) biplot of DNA Damage, PCBs, PBDEs, 
DDE and carbon and nitrogen stable isotopes and their relationships across study 
sites. Data are projected onto the ordination axes.  The large dots are the centroids 
and the shaded areas represent the 95% confidence ellipses.  The vectors (black 
lines) represent the contribution of different variables to each axis. 
 
 
94 
rates were low (e.g., 0.55 fledglings/active nest on the middle Potomac in 1970; 
Wiemeyer et al., 1971).  Spitzer (1980) and Poole (1989) stated that an osprey population 
producing 0.8-1.15 fledglings/active nest is required to maintain a stable population in the 
Chesapeake Bay watershed.  By the mid-90’s population estimates had more than 
doubled (Watts et al., 2004).  Current productivity estimates (>1.17 fledglings/active nest, 
Table 1) indicate that osprey reproduction is adequate to maintain a stable osprey 
population in many parts of the Chesapeake (Rattner et al. 2004, Watts and Paxton, 2007; 
Lazarus et al. 2015a and present study).   
 
Contaminants in ospreys 
 The present study further demonstrates that residues of p,p’-DDE in osprey eggs 
have declined since the 1970’s (e.g., averaging 3.1 µg/g ww on the middle Potomac in 
1971-1977, Weimeyer et al. 1988 to 0.63 µg/g ww in the present study). Current p,p’-
DDE concentrations are well-below the threshold associated with 10% eggshell (i.e., 2.0 
µg/g ww, Wiemeyer et al., 1988), and at these low levels there was no relationship 
between p,p’-DDE and eggshell thickness.   
 PCB concentrations were similar among the Susquehanna River and flats, 
Anacostia/middle Potomac and James Rivers.  Compared to historical values reported by 
Wiemeyer et al., (1988) on the Potomac River in 1973 (9.8 µg/g ww), PCB 
concentrations in this study declined by approximately 50%.  Mean total PCB 
concentrations at all study sites were lower than levels described in eggs from the 
Baltimore Harbor/Patapsco River in 2011 (geometric mean 7.77 µg/g ww, maximum 
35.0 µg/g ww, Lazarus et al., 2015a).   Although PCBs have been associated with many 
95 
adverse effects in fish-eating birds (Harris and Elliott, 2011), none were documented in 
this study as indicated by logistic regression.   The TEQs found in some study sites 
exceeded the no-observed-adverse-effect-level for osprey nestling growth (0.136 ng 
TEQ/g ww egg) (Woodford et al., 1998).  Although growth rate was not examined in the 
present study, body weight was used as a surrogate measure.  There was no evidence of a 
relation between toxic TEQ in Chesapeake Bay osprey eggs and body weight of 40-45 
day old nestlings (R2=-0.12, p=0.42).  
 Average total PBDEs were greatest on the Anacostia and Susquehanna (368.8 
ng/g and 343.5 ng/g ww respectively).  These values are lower than those reported in 
Chesapeake Bay osprey eggs from 2000-2001 (Rattner et al., 2004). Manufacture of the 
penta BDE commercial formulation from which these congeners derive ceased in 2004 
(La Guardia et al., 2007). The expanded subset of nests sampled on the Anacostia/middle 
Potomac River revealed that PBDE egg residues decreased with downstream distance 
from the Blue Plains WWTP. Such treatment facilities are documented sources of PBDEs 
(LaGuardia et al., 2007; Henny et al., 2011). Notably, the greatest residues of PBDEs on 
both the middle Potomac River (801.8 ng/g ww) and Susquehanna (648.6 ng/g ww) were 
found in nests in close proximity to WWTPs.  Values in the present study were below the 
lowest-observed-adverse-effect-level associated with reduced pipping and hatching 
success in American kestrels (Falco sparverius) (1.8 µg/g ww; McKernan et al., 2009), 
but exceeded the recommended toxicity reference value for aquatic birds (180 ng/g ww; 
Zhang et al., 2014). The paucity of ecotoxicity data for aquatic birds from Susquehanna 
and northern Bay makes it difficult to place the present findings into a historical 
perspective, but these data can serve as valuable documentation for future monitoring 
96 
studies.  Alternative-BRFs including, α-HBCD, BTBPE, DBDPE, TBB and TBPH were 
detected at some of the study sites at low concentrations compared to the PBDE flame 
retardants (Table 4), but it is difficult to determine their significance as toxicity reference 
values have yet to be derived for these compounds in birds.  
 
Osprey diet and biomagnification of contaminants 
 Dietary observations documented in this study were similar to those reported in 
other characterizations of osprey foraging behavior in Chesapeake Bay (McLean and 
Bird 1991; Viverette et al., 2007; Glass and Watts, 2009).  These studies similarly 
document that Atlantic menhaden are an important component in the diet of ospreys in 
much of the Chesapeake Bay.  In tidal freshwater tributaries, osprey diet shifts from 
estuarine species to catfish and gizzard shad (Glass and Watts, 2009; Supplemental Table 
1).  Although osprey diet differed among study sites, only slight differences in 15N 
signatures were observed (e.g., both striped bass and catfish species are opportunistic 
predators, while gizzard shad and menhaden are both planktivores).   
Biomagnification of lipophilic compounds has been well studied in the field of 
ecotoxicology (e.g., Kelly et al., 2007).  In the present study, biomagnification factor 
estimates are generally comparable to other reports examining the osprey food web 
(Henny et al., 2003, 2009; Chen et al., 2010).  Overall, BMFs were similar when 
expressed on a ww or lw basis.  The BMF for p,p’-DDE  averaged 19.8 ng/g lw, which 
was comparable to Chen et al. (2010; i.e., 18 ng/g lw) for the Chesapeake Bay, but is one-
fourth than that reported by Henny et al. (2003; estimated to be 85 ng/g lw in an osprey 
egg) on the Willamette River in Oregon.   This difference may be attributable to greater 
97 
p,p’-DDE concentrations in fish that ospreys are consuming on their wintering grounds 
(Henny et al., 2003). Again, the BMF for total PCBs in the present study was comparable 
to that reported previously for Chesapeake (27.9 ng/g lw, present study vs. 25.1 ng/g lw, 
Chen et al., 2010), but three times greater than that on the Willamette River, OR 
(estimated to be 11 ng/g lw for osprey eggs, Henny et al., 2003). In general, average 
residues of PCBs in Chesapeake Bay osprey eggs are 6 times greater than residues along 
the Willamette River, while for fish total PCB concentrations were only about 2 times 
greater than those collected along the Willamette. 
Several of the organochlorine pesticides and methoxyriclosan had a BMF < 5 
indicating only modest biomagnification (p,p’-DDD, cis -chlordane, trans-chlordane, cis-
nonachlor, trans-nonachlor, and methoxytriclosan) (Thomann et al., 1992; Barron, 2003).  
No biomagnification was observed for the alt-BFRs even though their log kow’s are just 
as great as the hexa-, octa- and deca- BDE formulations (US EPA, 2010; Wu et al., 
2011). Even though deca-BDE has been used worldwide, it was not found in great 
concentrations in this study.  This is consistent with findings reported in Chen and 
coworkers (2009), where they indicate that BDE-209 was detected in peregrine falcon 
(Falco peregrinus) eggs, but not in the aquatic birds studied.  Other factors may play a 
role including biotransformation, and the unstudied metabolites may actually be more 
bioaccumulative than the parent compounds (Kelly et al., 2007).   
 
Relation of concentrations of PCBs, DDE, and PBDEs with DNA damage 
The redundancy analysis indicated that DNA damage was most closely related to 
PBDE concentrations in osprey eggs, and presumably in nestling ospreys; however, 
98 
univariate analysis was not significant between PBDE concentrations and DNA.  
Although this analysis separates PBDEs as a large contributor to DNA damage, other 
chemicals including PAHs, PCBs, perfluorinated compounds, radiation and some metals 
have been shown to cause DNA damage in songbirds and Japanese quail (Cortunix 
cortunix japonica) (Custer et al, 2000; Devaux et al., 1998; Misra et al., 1998; Kim et al., 
2010; Luloff et al., 2011).  Damage may have been caused by some other co-occurring 
but unanalyzed chemical or chemicals. DNA damage has been observed in laboratory 
studies where American kestrels and common terns (Sterna hirundo) were exposed to the 
commercial PBDE formulation DE-71 (Rattner et al., 2013) and was elevated in the 
higher dose groups. The increased production of oxyradicals (above endogenous 
production) can lead to a variety of consequences in the body including mutations, 
lesions and disease progression as the body’s natural repair mechanisms become 
overwhelmed (Valavanidis et al., 2009).  
 
Conclusions 
Ospreys in the Chesapeake Bay are now thriving and our estimates in several 
tributaries including historic ROCs, indicate that productivity is adequate to maintain a 
stable population.  Both legacy and current use flame-retardants had limited effects on 
osprey productivity across study sites and are well-below established toxicity thresholds.  
Biomagnification factor estimates for principal contaminants are similar to those reported 
in other studies. Similar to findings by Lazarus and coworkers (2015a), there is increased 
evidence of genetic damage in ospreys nesting in the most polluted areas.   
 
99 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Supplemental Figure S1. Linear contribution (eigenvalues) of predictor variables 
to each axis. 
100 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Supplemental Table S2. Osprey diet composition from each study site. 
 
101 
 
 
 
 
 
 
 
 
 
 
 
Supplemental Table S3. Contaminant residues in whole fish collected across all 
study sitesa,b. 
 
102 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Supplemental Table S3 cont’d. Contaminant residues in whole fish collected across all 
study sitesa,b. 
 
103 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Axis Variance 
Proportion 
of 
variance 
Cumulative 
proportion 
explained 
Degrees of 
freedom 
F P 
1 0.1396 0.814 0.814 1 46.9 <0.001 
2 0.0207 0.121 0.935 1 6.99 <0.001 
3 0.0076 0.044 0.979 1 2.54 0.0667 
4 0.0032 0.019 0.997 1 1.08 0.337 
5 0.0004 0.002 1 1 0.147 0.938 
Supplemental Table S4. Amount of variance explained by each axis as well as
 the proportion and cumulate proposition of variance explained. 
 
104 
 
 
 
 
 
 
 
 
CHAPTER 4 
 
EXPOSURE AND FOOD WEB TRANSFER OF PHARMACEUTICALS IN OSPREYS 
(Pandion haliaetus): PREDICTIVE MODEL AND EMPIRICAL DATA 
 
 
 
 
 
 
 
 
 
 
 
 
 
105 
Introduction 
In parallel with human population growth and a myriad of veterinary and human 
health uses, pharmaceuticals and their metabolites primarily enter the environment 
through waste- water from bulk drug production, sewage plants and septic systems, and 
in biosolids applied to agricultural lands (Kolpin et al., 2002; Ramirez et al., 2009). The 
development of advanced analytical techniques and widespread monitoring has revealed 
the presence of pharmaceuticals in a variety of environmental matrices (sediments, 
sewage sludge, water, and fish). Pharmaceuticals may not be completely removed by 
traditional wastewater treatment systems, and with constant wastewater inputs, even 
labile compounds may exhibit pseudo-persistence in surface waters (Daughton and 
Ternes, 1999; Celiz et al., 2009). Their detection in the environment has raised concerns 
about bioaccumulation, transfer through the food web, and potential effects that 
pharmaceutical “cocktails” may elicit on ecosystems.    
 Understanding ecological risks of pharmaceuticals to free-ranging wildlife 
(amphibians, reptiles, birds, and mammals) remains a major research need (Boxall et al., 
2012), the one exception being the nonsteroidal anti-inflammatory drug (NSAID) 
diclofenac used to treat livestock. Diclofenac use resulted in nontarget poisoning and 
endangerment of several species of Asian vultures feeding on carcasses of cattle that had 
been treated with this drug (Oaks and Watson, 2011). The catastrophic effects of 
diclofenac on old world vultures resulted in detailed investigations of several NSAIDs in 
birds. A recent workshop evaluated the risk of pharmaceuticals to wildlife and identified 
major information gaps including the need to conduct food web exposure modeling and 
environmentally realistic risk assessments (Arnold et al., 2013). Hernout et al. (2011) also 
106 
suggested that prioritizing chemicals (e.g., metals, pesticides, pharmaceuticals) in a food 
web framework before intensive and costly investigation would be beneficial to natural 
resource managers and policymakers.      
 Ospreys (Pandion haliaetus) are a high trophic level species that have served as 
sentinels of ecosystem health and environmental change (Grove et al., 2009). Their eggs 
and blood are excellent matrices to document spatial and temporal trends and to elucidate 
exposure, bioaccumulation, and biomagnification of contaminants. Ospreys are strictly 
piscivo- rous and this aspect makes their diet easy to monitor and link to sources of 
localized contaminant exposure. Their diet can vary with salinity (Glass and Watts, 
2009), prey availability and trophic position and can range from anadromous fish in 
polyhaline regions to nonmigratory fish in oligohaline waters. Ospreys are adaptable to 
human landscapes and can be found nesting in highly industrialized and urbanized areas 
and even in proximity to wastewater treatment plants (WWTPs).    
 To date, no studies have examined the bioaccumulation of pharmaceuticals and 
their fate in the water-fish-osprey food web. This study describes a framework and the 
findings of a screening-level exposure assessment to estimate the daily and cumulative 40 
day intake of pharmaceuticals that are being analyzed by some environmental research 
laboratories (Du et al., 2012; Furlong et al., 2014). This was complemented by empirical 
analyses of 23 compounds and an artificial sweetener analyzed in water and blood plasma 
of fish and osprey nestlings from sites located along potentially impaired waterways in 
Chesapeake Bay.  
 
 
107 
Materials and Methods 
Screening-level exposure model: daily intake 
 The daily intake (DI) of 113 pharmaceuticals, metabolites, and an artificial 
sweetener (Table A2) by an adult female osprey was calculated to determine which 
compounds reached or exceeded the human therapeutic dose (HTD) (assumes 
comparable intestinal absorption for both ospreys and humans) (Figure 9). These 
compounds are quantified at the US Geological Survey National Water Quality 
Laboratory (Furlong et al., 2014) and some of these compounds are analyzed in fish 
plasma (Du et al. 2012). Three hypothetical exposure regimes (10, 100, and 1000 ng/L) 
were chosen (ranging from “dilution dominated” high flow to “wastewater effluent 
dominated” low flow) (Brooks et al., 2006) and modeled across 3 pH values (pH 6, pH 7, 
pH 8) that are representative of surface water gradients at field study sites. The pH 
consideration is important, because the drugs examined are all potentially ionizable and 
their bioconcentration factors (BCFs) are pH-specific and dependent on log D (a measure 
taking into account ionized and un-ionized forms of a molecule) (Meylan et al. 1999; Fu 
et al. 2009). These factors can influence bioaccumulation and toxicity in fish (Valenti et 
al. 2009, 2011, 2012), and ultimately, their absorption (bioaccessibility) in the 
gastrointestinal tract of birds. Predicted BCFs for each substance (ACS, 2014) were used 
to calculate the quantity of a pharmaceutical accumulated in a generic fish in 24 hours 
(Berninger et al., 2011). 
 
 
108 
Figure 9. Theoretical screening-level exposure assessment framework used to model 
the daily intake (DI) for an osprey. 
 
 
 
 
Calculated pharmaceutical residues in fish from each scenario were used to 
estimate the DI for a 1568 g adult female osprey (US EPA, 1993). Due to the 
complexities in modeling cumulative exposure of a growing osprey nestling (e.g., logistic 
109 
growth plateauing at 40 days, changing food intake, and metabolic demands, etc.), a 40 
day exposure assessment for an adult osprey was conducted. Food intake rate (FIR) was 
estimated 2 ways. The first estimate used osprey bite size and body weight (BWt) to 
calculate a FIR of 329 g of fish wet weight (ww) per day (Poole, 1985; US EPA, 1993). 
The second estimate used dry weight (dw) consumption rates based on the relationship 
between BWt and metabolic energy for birds (FIR g dw/day= 0.648BWt0.651) (Nagy, 
1987; US EPA, 1993). The FIR for an adult female osprey (77.94 g fish dw/day) 
converted to ww (assuming 75% water content for a generic fish) was 312 g/day. These 2 
estimates yielded similar results and the metabolic-based estimate was selected for use. 
The DI (µg pharmaceutical/kg Bwt) was calculated using Equation 1 
  DI = (residue in fish) (FIR)/(kg BWt).    (1) 
The DIs for varying degrees of absorption were compared to the oral HTD for an 
adult. Human therapeutic doses were obtained as the minimum daily dose to exert a 
therapeutic effect (RxList, 2008; FDA 2012; Drugsite Trust, 2014). 
 
Screening-level exposure model: 40 day cumulative intake 
 To estimate cumulative body burden of ospreys, assumptions included that diet 
was the principal exposure route, BWt and FIR were constant, and intestinal absorption 
was comparable between ospreys and humans. Clearance was incorporated assuming a 
first-order kinetic elimination equation to calculate total exposure (i.e., µg/kg BWt) 
because the majority of ionizable pharmaceuticals follow this type of elimination (Bardal 
et al. 2011). Using DI (t = 1 d), exposure (E) oscillated following a saw-tooth pattern 
between peak (Epeak) and trough (Etrough) 
110 
  Epeak = (DIremaining)e
-kt + DI (just after meal)    (2) 
  Etrough = (DIremaining)e
-kt + (DI)e-kt (just before a meal)  (3) 
  t1/2 = ln (2)/ k (half-life elimination constant    (4) 
There are limited data on the half-lives (t1/2) of pharmaceuticals in birds to apply 
in these equations. To place the 40 day exposure into perspective, the drug t1/2 in ospreys 
needed to reach or exceed the HTD within 40 days at most extreme scenario (1000 ng/L 
concentration, pH 8, complete absorption) was back calculated. Equation (3) (daily 
exposure at nadir) was used to conservatively estimate cumulative daily body burden. 
The back calculated t1/2 for ospreys was compared to the t1/2 in humans (Ebadi, 2008; 
Wishart et al., 2008; FDA, 2012). 
 
Empirical pharmaceutical exposure data 
Study sites were selected in urbanized areas in proximity to WWTPs, combined 
sewer outflows, and effluent dominated low flow sites. These sites include the 
Susquehanna River (MD, PA), Back River (MD), James River (VA), and the US 
Environmental Protection Agency (USEPA) designated Regions of Concern (Baltimore 
Harbor [MD], Anacostia River/middle Potomac [DC, MD, VA], and the Elizabeth River 
[VA]), all of which appear on the 303d list for impaired waterways (Figure 10) (US EPA 
2013). Sampling was undertaken during osprey nesting seasons of 2011, 2012, and 2013. 
The Paul S. Sarbanes Ecosystem Restoration Project at Poplar Island (MD), a remote 
mid-Bay location, was used as a reference site. 
111 
Figure 10. Map detailing the locations of Chesapeake Bay study sites. 
 
 
 
Duplicate water samples were collected from 12 select sampling sites (2–3 
locations along a stretch of the Susquehanna River, Back River, Anacostia/middle 
Potomac Rivers, James River, and at Poplar Island). Surface water samples were 
collected in clean 4 L amber glass jugs. Field blanks were taken by opening an empty jar 
to account for other sources of contamination. Water quality parameters (pH, dissolved 
112 
O2, temperature, conductivity) were measured concurrently (YSI Multimeter Yellow 
Springs OH). Water samples were stored on wet ice and shipped overnight to Baylor 
University. 
All procedures involving fish and ospreys were conducted under approval of the 
Institutional Animal Care and Use Committees of the US Geological Survey (USGS) and 
the University of Maryland, and appropriate scientific collection permits. Game camera 
(Bushnell 8MP Trophy Cam, Overland Park, KS) images of prey items delivered to 
osprey nests, direct observations, and identification of scraps were used to reconstruct 
osprey diet and identify target species for sampling. Based on osprey diet reconstruction, 
a combination of gizzard shad (Dorosoma cepedianum), catfish (blue catfish Ictalurus 
furcatus, brown bullhead Ameiurus nebulosus, and channel catfish Ictalurus punctatus), 
and carp (Cyprinus carpio) were sampled on the Susquehanna, Anacostia/middle 
Potomac, and James Rivers. Atlantic menhaden (Brevoortia tyrannus), striped bass 
(Morone saxatilis), and white perch (Morone americana) were sampled at the more 
saline Poplar Island site, and a combination of carp, catfish, gizzard shad, and white 
perch were sampled on Back River. These fish species reflect different trophic levels 
ranging from primary consumers (herbivorous) such as Atlantic menhaden and gizzard 
shad, to secondary consumers (carnivorous) including white perch and striped bass, to 
catfish and carp (omnivorous) representing a combination of both primary and secondary 
consumers. Fish were captured by electroshocking in upriver sites. At Poplar Island, fish 
were captured using a midwater trawl and a commercial pound net. Plasma was sampled 
from the 2 to 3 dominant prey fish species found at each site that fell within the osprey 
foraging size range (25–35 cm) (Poole 1989). All fish (n = 233) were anesthetized 
113 
(MS222, tricane methanesulfonate), weighed, measured, and 1 to 2 mLs of blood were 
sampled using a heparinized syringe. Fish blood was stored on wet ice and transported to 
the USGS-Leetown Science Center in WV. The blood was centrifuged at 2000 g at 4°C 
for 10 min on the same day of collection. Plasma was harvested for pharmaceutical 
analyses and stored at −80°C. Fish tissue was saved for analysis of organic contaminants 
as part of a concurrent study. 
Osprey nests were identified in mid-March along a 25 to 35 km stretch of river. A 
sample egg was collected for analysis of legacy contaminants and nests were visited 
weekly to determine reproductive success as part of a concurrent study. Once nestlings 
reached 40 to 45 days of age, a single chick was briefly removed from the nest (<10 min). 
Body weight and culmen length were measured, and a 5 to 7 mL brachial blood sample 
was drawn into a heparinized syringe. Samples were stored on wet ice and centrifuged at 
1500 g at 4°C for 10 min on the same day of collection. Plasma was harvested and 
samples (n = 69) were stored at the USGS-Patuxent Wildlife Research Center at −80°C. 
All plasma samples were shipped frozen to Baylor University for the quantification of 
pharmaceuticals. 
 
Analysis of pharmaceuticals 
 
A suite of 23 pharmaceuticals and metabolites and an artificial sweetener (Tables 
S5 and S6) were quantified in water and plasma samples from fish and osprey nestlings 
via isotopic dilution liquid chromatography-tandem mass spectrometry (LC-MS/MS). 
These compounds included analgesics (acetaminophen, codeine), antibiotics 
(erythromycin, sulfamethoxazole, trimethoprim), an anticoagulant (warfarin), 
114 
antidepressants (paroxetine, fluoxetine, sertraline, and primary metabolites norfluoxetine 
and desmethylsertraline), an antihistamine (diphenhydramine), antihypertensives 
(atenolol, diltiazem, propranolol), anti-inflammatories (celecoxib, diclofenac), an 
antilipemic (gemfibrozil), an antiseizure (carbamazepine), a parasiticide (ivermectin), 
psychostimulants (diazepam, methylphenidate), a stimulant (caffeine), and an artificial 
sweetener (sucralose; conservative tracer of effluent discharges) (Soh et al., 2011). 
For water, sample filtration and extraction generally followed previously 
described protocols (Du et al. 2014). A mixture of 24 internal standards (deuterated 
analogues of target compounds, except for ivermectin for which abamectin was the 
internal standard) and 5 mL of methanol was added to 500 mL of each water sample 
before extraction and acidification (pH adjusted with 100 µL of 85% [v/v] phosphoric 
acid) (Lajeunesse et al. 2008). Resulting concentrations of internal standards were 
approximately 100 ng/g. Samples were subsequently loaded onto strong cation-exchange 
cartridges (Strata-SCX, 500 mg; Phenomenex, Torrance, CA) preconditioned with 4 mL 
of methanol and 8 mL of nano-pure water. Each cartridge was washed with 4 mL of HCl 
(0.1 N) and 4 mL of methanol, followed by elution of the 5 antidepressant serotonin 
reuptake inhibitors with 6 mL of 5% (v/v) NH4OH in methanol. Extraction of 19 other 
analytes generally followed a previously reported protocol (Vanderford and Snyder 
2006). Each sample (500 mL subsample) was spiked with a mixture of internal standards 
and loaded onto a preconditioned HLB cartridge (200 mg, Waters, Milford, MA). These 
loaded cartridges were air-dried and subsequently eluted with 5 mL methanol followed by 
5 mL 10:90 (v/v) methanol-methyl tertiary butyl ether. The eluate from 2 separate 
extractions was evaporated to dryness under a stream of N and reconstituted in 1 mL of 
115 
chromatographic mobile phase (i.e., methanol-0.1% [v/v] aqueous formic acid). Before 
LC-MS/MS analysis, samples were sonicated for 1 min and filtered using Pall Acrodisc 
hydrophobic Teflon Supor membrane syringe filters (13 mm diameter; 0.2 µm pore size; 
VWR Scientific, Suwanee, GA). 
For plasma samples, a slightly modified extraction method was used (Fick et al., 
2010a). An aliquot of fish and osprey plasma (typically 1 mL), combined with the same 
mixture of internal standards that was used for water, was diluted to 5 mL using 0.1% 
(v/v) aqueous formic acid and mixed thoroughly by sonication. The mixture was loaded 
on preconditioned (5 mL of methanol and 5 mL of nano-pure water) HLB SPE cartridges 
(200 mg, Waters). Each cartridge was air-dried and subsequently eluted with 5 mL of 
methanol. The eluate was reconstituted, and analytes were quantified by LC-MS/MS as 
previously described (Du et al. 2012). 
For water and fish plasma, method detection limits (MDLs) were less than 
11 ng/L with the exception of ivermectin and sucralose (Table S6). Osprey plasma from 
the reference site, spiked with the mixture of internal standards, was used to determine 
MDLs, which were similar to that of water and fish (Table S2). For quality control 
purposes, 1 pair of matrix spike samples and 1 method blank sample was added for each 
batch analysis. Spike recoveries ranged from 81% to 111% in water, 81% to 113% in 
fish, and 81% to 89% in ospreys. 
 
Statistical analyses 
For the daily and 40 day screening-level exposure models, DI and half-life 
elimination constants were estimated using Microsoft Excel. For empirical exposure data, 
116 
concentrations of pharmaceuticals and an artificial sweetener were first recovery 
corrected and only values above the MDL were reported. If the analyte was present in all 
samples, the arithmetic mean and standard deviation were obtained using SAS (SAS 
Institute, NC). If an analyte was detected in only 1 of the 2 duplicate water samples, one 
randomly selected value was included in the statistical analysis. If analytes were detected 
in over half (but not all) of the samples at a study location, the Kaplan-Meier method was 
used to estimate an interval that contains the theoretical mean (Helsel 2005, 2009). 
 Parametric statistics were conducted for those analytes detected in all samples. 
Continuously distributed analyte concentrations in water, and fish and osprey plasma 
were tested for homogeneity of variance (Levene's test) and normality (Shapiro-Wilk 
test). In 2 instances, variables were log or square root transformed to correct for 
normality or heterogeneous variances. A 1-way analysis of variance followed by Tukey's 
honestly significant difference method for multiple comparisons was performed 
(α = 0.05). For those sites with nondetects, a generalized Wilcoxon nonparametric test 
was used (Helsel, 2005). 
For all detectable compounds, a hazard quotient (HQ) was calculated by dividing 
the maximum concentration found in fish or osprey plasma by the human therapeutic 
plasma concentration (Cmax). The larger the HQ value, the greater the potential for a 
compound to exert a pharmacological effect in fish or ospreys. 
 
 
 
 
117 
Results 
Screening-level exposure model 
 Of 114 compounds, 31 had a BCF less than or equal to 1.00 (Table S6). These 31 
compounds plus the bronchodilator tiotropium (no BCF available) were excluded from 
the model. Of the 83 remaining compounds, 15 had both a BCF greater than 100 and an 
estimated DI greater than 20 µg/kg-day and are predicted to have the greatest potential to 
bioaccumulate. The calculated DIs for these 15 compounds at concentrations of 10, 100, 
and 1000 ng/L in an adult female osprey are presented in Table 10. At concentrations of 
1000 ng/L water at pH 8, the DI of orlistat, fenofibrate, tamoxifen, and loperamide would 
be 1.1 to 4.4 times greater than the oral HTD. Based on the information in Table 10, 
orlistat is the only compound that still exceeds the HTD even if the intestinal absorption 
in ospreys was only half that of humans (12 198 µg/kg BWt-day, 2.4 times greater than 
the HTD). Pharmaceutical concentrations in water and pH values selected in the model 
were environmentally realistic (analyte concentrations in Chesapeake Bay range from 
0.029 to 10,249 ng/L and site pH ranged from 6.15 to 8.34) (Tables 3 and A4).  
 
Cumulative 40 day exposure model  
For the top 15 compounds (BCF > 100, DI > 20 µg/kg BWt-day), the theoretical 
half-life (calculated from the half-life elimination constant, k) for an osprey to reach the 
HTD within a 40 day period ranged from 1 to 231 days. For this subset of 
pharmaceuticals, the half-lives in humans ranged from 0.04 to 7 days. Notably, the half-
lives in ospreys for fenofibrate, tamoxifen and ezetimibe were less than that in humans, 
with the HTD being exceeded in just 3 to 7 days (Table 10).
118 
 
 
 
 
 
 
 
 
 
 
Table 1. Daily intake (DI) at three water concentrations and t
1/2
 to accumulate a human therapeutic dose (HTD) within 40 days for 16 
pharmaceuticals compared to human values.
Rank Compound
BCF      
pH 8
HTD                
(µg/kg BW)
t
1/2
 ospreys to exceed HTD 
100% absorption
a 
t
1/2 
humans (Days)
10 ng/L pH 8 100 ng/L pH 8 1000 ng/L pH 8 1000 ng/L pH 8
1 Orlistat 123000 244 2440 24396 5143 1.79 0.04-0.08
2 Fenofibrate 15100 29.9 299 2995 1714 0.42 (HTD in 3 days) 0.83
3 Piperonyl butoxide 2400 4.76 47.6 476 Topical NA
b
NA
b
4 Tamoxifen 797 1.58 15.8 158 143 1.00 (HTD in 4 days) 5-7
5 Ketoconazole 646 1.28 12.8 128 2857 22.8 0.14
6 Ezetimibe 589 1.16 11.6 117 143 1.19 (HTD in 7 days) 0.79-1.25
7
Iminostilbene
c
556 1.10 11.0 110 5714 34.3 0.08-2.71
8 Loratadine 538 1.06 10.7 107 143 1.27 0.35
9 Loperamide 519 1.03 10.3 103 57.1 0.67 0.45
10 Promethazine 358 0.71 7.10 71.1 357 4.07 0.67-0.79
11 Diltiazem 343 0.68 6.80 68.0 2571 231 0.12-0.18
12 Raloxifene 327 0.65 6.50 65.0 857 11.3 1.15
13 Dextromethorphan 228 0.45 4.52 45.2 571 11.3 0.05-0.16
14 Desmethylsertraline 213 0.42 4.20 42.3 NA
b
NA
b
2.58-4.33
15 Sertraline 142 0.28 2.82 28.2 357 11.5 1.04-1.08
a
Shaded boxes indicate the estimated DI or t
1/2
 for ospreys exceeds HTD or human t
1/2
.
b
NA=not applicable; not injested for therapeutic uses or a metabolite of a parent compound.
c
Iminostilbenes are a group of antiseizure drugs which includes carbamazepine and oxcarbamazepine.  HTD and half-lives are given based on the lowest dose in 
this group that is needed for a therapeutic effect.
DI Osprey (100% absorption)                                                                  
(µg/kg BW-day)
a
Table 10. Daily intake (DI) at three water concentrations and t1/2 to accumulate a human therapeutic dose (HTD) within 40 
days for 16 pharmaceuticals compared to human values. 
 
119 
Empirical assessment of pharmaceuticals in water, fish, and osprey 
 Of the 24 analytes measured, 17 were detected in water and 8 of these were also 
detected in fish (Tables 2, A4 and A5). Diltiazem was the only compound detected in all 
3 matrices (Table 11, all values presented on a ng/L basis). When compared to the human 
therapeutic plasma concentration (Cmax), all detected compounds had a HQ less than or 
equal to 0.08. 
 In water, concentrations were averaged from the 2 to 3 sampling sites per 
tributary. There were 73 out of 77 instances where the analyte was detected in the 
duplicate water samples and the median relative percent difference between samples was 
12.02%. Samples were collected from the reference site each year of the study, but 
diltiazem was only detected in water samples in 2011 and 2012. At Back River, 18 
analytes were detected in water with concentrations being 2 to 154 times greater than 
other sites. Carbamazepine, diltiazem, sulfamethoxazole, diphenhydramine, and caffeine 
were detected in water at all intensively sampled sites. Statistical analysis revealed that 
the Anacostia/middle Potomac Rivers had the greatest carbamazepine (log transformed) 
and caffeine concentrations in water (Back River excluded from analysis as there was 
only a single sample), followed by the James, Poplar Island, and Susquehanna (p < 0.04). 
There were no differences in diltiazem concentrations in water among all 5 sites.
 Although 7 pharmaceuticals and sucralose were found in fish, detection frequency 
was low, rarely exceeding half of the samples per site. Thus, parametric statistical 
analyses could not be conducted among analytes, fish species, and sites (Table A5). By 
inspection of these data, diphenhydramine and diltiazem were present in fish from all 
study sites. Both the Anacostia/middle Potomac and Back Rivers had the largest suite of 
120 
pharmaceuticals detected. Diltiazem was found in 13% of fish samples and present in all 
species, with the greatest plasma concentration (2.4 ng/mL) found in a catfish collected 
on the Susquehanna (Table 11). 
Diltiazem was detected in all 69 osprey nestling plasma samples (1.6–24.6 times 
greater than the MDL). Osprey nestlings on the Anacostia/middle Potomac, Baltimore 
Harbor, and Back River had higher (p < 0.04) diltiazem plasma concentrations (square 
root transformed) compared to the James, Elizabeth, Susquehanna Rivers, and the Poplar 
Island reference site. Plasma diltiazem concentrations at Poplar were higher than those on 
the Elizabeth and James Rivers (p < 0.007). For the Potomac River, which had osprey 
nests evenly spaced downriver from Blue Plains WWTP, diltiazem in nestling plasma did 
not exhibit a spatial concentration gradient (p > 0.35).
 
 
 
121 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table 2. Summary of compounds detected in water and fish across study sites
a
Anticoagulant Antihistamine
Acetaminophen Codeine Sulfamethoxazole Trimethoprim Erythromycin Warfarin Diphenhydramine Methylphenidate Diazepam
Poplar Island
+
Water +
Atlantic Menhaden
Striped Bass +
White Perch +
Anacosita/middle Potomac 
Rivers
Water + + + + + +
Catfish sp. +
Gizzard Shad +
Carp +
Back River
Water + + + + + + + + +
Catfish sp. +
Gizzard Shad + +
Carp +
White Perch +
James River
Water + + +
Catfish sp. +
Gizzard Shad +
Susquehanna River
Water + +
Catfish sp.
Gizzard Shad 
Antilipemic Antiseizure
Artificial 
Sweetener
Stimulant
Atenolol Diltiazem Propanolol Celecoxib Diclofenac Gemfibrozil Carbamazepine Sucralose Caffeine
Poplar Island
Water + + + +
Atlantic Menhaden +
Striped Bass + +
White Perch +
Anacosita/middle Potomac 
Rivers
Water + + + + + + +
Catfish sp. +
Gizzard Shad + + +
Carp +
Back River
Water + + + + + + + + +
Catfish sp. +
Gizzard Shad + +
Carp +
White Perch
James River
Water + + + + + +
Catfish sp. + +
Gizzard Shad +
Susquehanna River
Water + + +
Catfish sp. +
Gizzard Shad 
a
+ compound detected
Psychostimulant
Antihypertensive Anti-inflammatories
Site/Class
Site/Class
Analgesic Antibiotic
Table 11. et weight concentrations of polychlroinated biphenyls from Chesapeake Bay 
Regions of Concern and Poplar Island reference sitea. 
 
 
 
 
122 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Discussion 
Screening-level exposure assessment 
Whereas several studies have examined uptake of pharmaceuticals from water by 
fish (Brown et al. 2007; Ramirez et al. 2009), their transfer to high trophic level wildlife 
has not been evaluated. The likelihood for a broad suite of potentially ionizable 
Table 3. Diltiazem concentrations in water, fish and osprey on ng/L basis
a
Water (ng/L) Osprey (ng/L plasma)
Detects/n Detects/n
Mean “ std. dev. Mean “ std. dev.
Extremes Extremes
HQ
b
Catfish Gizzard Shad Carp Rockfish Menhaden Perch
2/3 NS NS NS 3/17 1/10 1/10 13/13
1.06-1.14
c
- - - 2199 “ 1524
C
<MDL-2.05 <MDL-1800 <MDL-410 <MDL-410 605-4458
0.06 0.01 0.01 0.15
2/3 0/30 3/33 1/18 NS NS NS 13/13
1.08-1.62
c
- - - 4517 “ 1384
A
<MDL-2.47 - <MDL-410 <MDL-330 NS NS NS 3503-8630
0.01 0.01 0.29
1/1 1/2 1/9 0/9 NS NS 0/5 7/7
- - - - - 2353 “ 1207
B,C
173 <MDL-350 <MDL-420 - - 1049-4288
0.01 0.01 0.29
3/3 8/27 3/27 NS NS NS NS 12/12
5.85 +2.51 - - 912 “ 225
D
2.96-7.49 <MDL-770 <MDL-570 537-1355
0.03 0.02 0.05
2/2 9/18 0/18 NS NS NS NS 10/10
1.67 + 0.45 - - 1434 “ 372
C,D
1.35-1.99 <MDL-2400 - 4049-2099
0.08 0.07
Additional sites where only osprey nestlings sampled
6/6
966 “ 352
D
564-1320
0.04
8/8
3786 “ 714
A,B
2885-5110
0.17
a
NS=Not sampled because fish species was not a large component of osprey diet at a particular site; - indicates no mean calculated, contaminant was detected in fewer than 
half of the samples; MDL = method detection limit. Means with different captial letter superscripts are significantly different (p < 0.05). 
b
Hazard Quotient (HQ) is the upper extreme concentration found in fish or osprey plasma divided by the human therapeutic plasma concentration (C
max
)
 
for diltiazem (30,000 ng/L).
c
If non-detects were present in less than half of the samples, the Kaplan-Meier method was used to estimate the extremes of the mean followed by a generalized Wilcoxin 
non-parametric test.
Detects/n
Fish (ng/L plasma)
HQ
b
Baltimore Harbor/ 
Patapsco River
Elizabeth River
Poplar Island
Anacosita/                  
middle Potomac 
Rivers
Back River
James River
Susquehanna 
River
Mean “ std. dev.
Extremes
Site
Table 12. Concentrations (ng/g ww) of polybrominated diphenyl ethers, additional 
brominated flame retardants and methoxytriclosan from Chesapeake Bay Regions of 
Concern and Poplar Island reference sitea. 
 
 
 
123 
pharmaceuticals to bioaccumulate in a water–fish–osprey food web was modeled using 
their concentration in water, pH-specific BCF and FIR of an adult female osprey. Those 
compounds with high pH-specific BCFs and long half-lives near low flow point sources 
(i.e., low dilution scenario) were predicted to exceed the HTD.  
This screening-level assessment identified a subset of 15 of 114 compounds that 
warrant further investigation based on their potential to exceed the HTD. Over a narrow 
range of pH (6–8), there was little effect on the BCF of 6 of these compounds (orlistat, 
fenofibrate, piperonyl butoxide, ezetimibe, iminostilbene, and loratadine), with BCFs 
fluctuating by less than 20% (Table S1). Although ionizable at pH extremes (low or high 
pKa), these 6 compounds were in their neutral state from pH 6–8 and predicted to be the 
most bioaccumulative. It has been suggested that compounds with such characteristics 
could evoke pharmacological responses and possibly toxicity in invertebrates and fish at 
their isoelectric point (Ebadi 2008; Rendal et al. 2011). The remaining 10 compounds 
(mean pKa 8.37) are not ionizable until pH exceeds environmentally relevant conditions. 
The use of pH-specific BCFs appears to be a valuable tool to identify and prioritize 
pharmaceuticals and metabolites that have the greatest potential to bioaccumulate at 
environmentally relevant conditions. 
An estimate of the half-life is required to model first-order kinetic elimination of 
drugs over a specific period of time and provides a measure of the persistence of a 
xenobiotic. Based on our screening-level exposure model, HTDs for fenofibrate, 
tamoxifen, and ezetimibe were exceeded in adult ospreys at theoretical half-lives (0.24–
1.19 days) that were less than their half-lives in humans (0.83–1.25 days) (Ebadi 2008; 
Wishart et al. 2008). Such theoretical half-lives are not unreasonable. For the 
 
 
 
124 
aforementioned compounds, it might be possible for an osprey to accumulate a HTD 
within 3 to 7 days of exposure in a low-flow scenario. 
Uncertainty factors and model assumptions 
 
Eighteen pharmaceuticals and an artificial sweetener were detected in water 
samples from Chesapeake Bay. Frequency of detection and concentrations were greatest 
in water samples collected on the Back River, which receives appreciable WWTP input 
(180 million gallons/day from 1.3 million residents from Baltimore) (Baltimore County 
Watershed Management Program 2012). Despite greater input from Blue Plains WWTP 
and population size (330 million gallons/day from 2.1 million residents of the 
Washington District of Columbia metropolitan area) (District of Columbia Water and 
Sewage Authority 2014), concentrations were seemingly lower on the Anacostia/middle 
Potomac Rivers. 
Sulfamethoxazole, diphenhydramine, diltiazem, carbamazepine, sucralose, and 
caffeine were frequently detected in water samples, and 3 of these (diphenhydramine, 
diltiazem, and carbamazepine) were often detected in fish plasma (Tables S3 and S4). In 
fish, detection frequency, concentrations, and HQs were low and far less than critical 
environmental concentrations hypothesized to cause pharmacological effects in fish 
(Schwab et al. 2005; Fick et al., 2010b; Du et al. 2014). This is not unexpected as other 
reports indicate that both sucralose and caffeine do not bioaccumulate in fish. Sucralose 
was detected at lower concentrations in fish than in water samples from the 
Anacostia/middle Potomac Rivers and did not bioconcentrate (maximum detected 
concentration in fish/maximum detected concentration in water = 0.50; Tables S3 and 
S4). Notably laboratory studies have reported a BCF less than 1 for sucralose in zebrafish 
 
 
 
125 
(Danio rerio) (Lillicrap et al. 2011) and is not that different from literature estimates 
(BCF = 1, Table S1) (ACS 2014). Of the compounds most frequently detected in fish, our 
estimated BCFs were within an order of magnitude compared to literature values 
presented in Table S1 (diphenhydramine: estimated 79.1 compared with literature value 
of 16.7; carbamazepine: 44.1 versus 16.2; diltiazem: 319 versus 343). Interestingly, 
antidepressants were not found in the present study despite being detected in many urban 
rivers in North America (Brooks et al. 2005; Ramirez et al. 2009; Lajeunesse et al. 2011; 
Schultz et al. 2010).  
Diltiazem was the only analyte detected in water, fish, and biota. For diltiazem, 
concentrations in water were low (mean 2.44 ng/L), and with the exception of Back River 
(173 ng/L), was generally an order of magnitude below those found in urban inland 
waters of the United States and Sweden (36–1800 ng/L) (Kolpin et al. 2002, 2004; Fick, 
et al., 2010a; Du et al. 2014). In Chesapeake Bay, diltiazem fish plasma concentrations 
were 2 times greater than those observed at 3 WWTPs in Sweden (MDL-1000 ng/L) 
(Fick et al., 2010a). Out of 10 commonly used pharmaceuticals tested in Daphnia magna 
and Japanese medaka (Oryzias latipes), diltiazem exhibited the greatest acute toxicity 
(96 h LC50 = 8.2 and 15.0 mg/L, respectively) (Kim et al. 2007). The predicted no-effect 
concentration for diltiazem based on the lowest acute EC50 values was estimated to be 
8.2 µg/L (Kim et al. 2007), which is over an order of magnitude greater than the 
maximum value observed in the Back River. Evaluating these data in a more complete 
assessment should also include chronic responses linked to therapeutic hazard (Brausch 
et al. 2012; Valenti et al. 2012). The aquatic hazards and risk of diltiazem and many other 
pharmaceuticals remain poorly characterized (Brooks, 2014). 
 
 
 
126 
Several interspecific differences in pharmaceutical bioaccumulation were found 
among fish species (Table S8). For example, diltiazem was detected in channel catfish, 
but not gizzard shad from the Susquehanna River, whereas carbamazepine was observed 
in blue and channel catfish, but again not gizzard shad from the Anacostia/middle 
Potomac Rivers. Spatial variations in fish migration patterns may explain such 
differences in pharmaceutical bioaccumulation. For example, anadromous gizzard shad 
migrate downstream to deeper waters in the winter, whereas catfish remain in upper 
estuarine sites where they may be continuously exposed to wastewater discharge. The 
influence of trophic position (e.g., herbivorous gizzard shad and omnivorous catfish) on 
pharmaceutical bioaccumulation in fish is not well understood. 
 
Empirical findings in osprey nestlings 
This screening-level exposure assessment suggests that only 3 of 24 analytes 
quantified in osprey plasma (diltiazem, sertraline, and desmethylsertraline) are likely to 
exceed the HTD. Of these 3, diltiazem has the highest pH-specific BCF (343 at pH 8) and 
was detected in all osprey nestling plasma at low concentrations (0.56–8.63 ng/mL 
plasma), with the maximum value being 28% of the HTD. Although present in all osprey 
samples, there were no overt signs (therapeutic or toxicological) observed in our 
companion study examining reproductive success. Ospreys are thriving in Chesapeake 
Bay, including the most contaminated sites, and reproduction is generally adequate to 
sustain stable populations (>1.15 fledglings per active nest) (Lazarus et al. 2015a). 
Of the 15 compounds identified in the screening-level model as having the 
greatest potential to bioaccumulate, 3 were measured in osprey plasma (Table 1; 
 
 
 
127 
diltiazem 11, desmethylsertraline 14, and sertraline 15), and only diltiazem was detected. 
The accumulation of diltiazem and other antidepressants is theoretically pH-dependent. 
The bioaccumulation characteristics (partition coefficients log p and log D at pH 8) of 
sertraline (log p = 5.08 and log D = 3.60) and its metabolite desmethylsertraline (log 
p = 4.89 and log D = 3.73), are not unlike diltiazem (i.e., log p = 4.73 and log D = 3.90). 
Thus, diltiazem may be bioaccumulating not only because of its high pH-specific BCF, 
but also because of other biological characteristics including specific binding 
mechanisms. It is clear that diltiazem concentrations were greatest in osprey nestlings 
followed by fish and water concentrations. Across sites, the maximum diltiazem 
concentrations in water, fish plasma, and osprey plasma were averaged for each matrix to 
approximate a biomagnification factor. Diltiazem concentrations in fish plasma were 21.6 
times greater than those in water, and osprey plasma concentrations were 4.71 times 
greater than fish. It should be noted that the biomagnification factor from fish to osprey 
most certainly varies with osprey diet composition. 
 
Conclusions 
This screening-level exposure assessment identified 15 out of 113 
pharmaceuticals and an artificial sweetener that warrant further investigation in fish-
eating birds due to their high BCF and DI. Some of these compounds might even exceed 
the HTD. The antihypertensive drug diltiazem was detected in all osprey nestling plasma 
samples in several tributaries of Chesapeake Bay. Twelve additional analytes that were 
predicted to bioaccumulate, but not measured in environmental samples should receive 
priority for further investigation. Although diltiazem in ospreys did not exceed the HTD 
 
 
 
128 
and was well below the Cmax, our findings indicate that it can bioaccumulate to levels that 
are over 4 times greater than values in fish plasma. Even though empirical concentrations 
of drugs in the present study are well-below therapeutic levels for humans, the paucity of 
effect threshold data for birds and lower vertebrates makes interpretation of these 
observations challenging. Our knowledge of mammalian pharmacology can assist in 
extrapolation of effects to wildlife (Huggett et al. 2003), but in some (and hopefully rare) 
instances, birds and other perhaps other classes of vertebrates may be sensitive to low-
level environmental exposures (Oaks and Watson 2011). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
129 
 
Supplemental Table S5. List of 114 pharmaceutical compounds and sucralose for the 
screening-level exposure assessment, their CAS number, and bioconcentration factor 
(BCF). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table S1. List of 114 compounds for the screening level exposure assessment, their CAS number and bioconcentration factor (BCF).
Number
Compound
a
Primary Use Class CAS Number
BCF                                    
pH 8
1 1,7-Dimethylxanthine Stimulant 611-59-6 1.00
2 10-Hydroxy-amitriptyline Metabolite of amitriptyline 1159-82-6 2.51
3 Abacavir Antiviral, reverse transcriptase inhibitor 136470-78-5 4.46
4
Acetaminophen
*
Analgesic, antipyretic 103-90-2 1.34
5 Acyclovir Antiviral 59277-89-3 1.00
6 Albuterol β-2 adrenergic receptor agonist for asthma treatment 18559-94-9 1.00
7 Alprazolam Benzodiazepine to treat anxiety 28981-97-7 16.8
8 Amitriptyline Tricyclic antidepressant 50-48-6 83.2
9 Amphetamine Psychostimulant 300-62-9 1.00
10 Antipyrine Analgesic, antipyretic 60-80-0 1.27
11
Atenolol
*
Antihypertensive 29122-68-7 1.00
12 Benztropine Anticholinergic used in treatment of Parkinson's 86-13-5 2.47
13 Betamethasone Synthetic glucocorticoid steroid 378-44-9 20.7
14 Bupropion Antidepressant, smoking cessation aid 34911-55-2 30.0
15
Caffeine
*
Psychoactive stimulant 58-08-2 1.00
16
Carbamazepine
*
Anticonvulsant and mood stabilizer 298-46-4 16.2
17 Carisoprodol Muscle Relaxant 78-44-4 23.2
18
Celecoxib
**
Anti-inflammatory 169590-42-5 53.9
19 Chlorpheniramine Antihistamine 132-22-9 4.54
20 Cimetidine Antacid, Histamine H-2 receptor antagonist 51481-61-9 1.00
21 Citalopram Antidepressant/antianxiety (SSRI) 59729-33-8 6.99
22 Clonidine Antihypertensive 4205-90-7 16.5
23
Codeine
*
Analgesic, antitussive, antidiarrheal 76-57-3 2.32
24 Cotinine Metabolite of nicotine, tobacco constituent 486-56-6 1.00
25 Dehydronifedipine Metabolite of the antihypertensive nifedipine 67035-22-7 6.76
26 Desmethyldiltizaem Metabolite of diltiazem 86408-45-9 29.8
27
Desmethylsertraline
**
Metabolite of sertraline 87857-41-8 213
28 Desvenlafaxine Antidepressant, and metabolite of venlafaxine 93413-62-8 1.00
29 Dextromethorphan Cough suppressant 125-71-3 228
30
Diazepam
*
Antiseizure, antianxiety and insomnia 439-14-5 79.2
31
Diclofenac
**
Anti-inflammatory 15307-86-5 1.00
32
Diltiazem
*
Antihypertensive 42399-41-7 343
33
Diphenhydramine
*
Antihistamine 58-73-1 16.7
34 Duloxetine Antidepressant 116539-59-4 27.3
35
Erythromycin
*
Antibiotic 114-07-8 7.73
36 Ezetimibe Antilipemic to reduce cholesterol 163222-33-1 589
37 Fadrozole Aromatase inhibitor 102676-47-1 6.91
38 Famotidine Antacid, Histamine H-2 receptor antagonist 76824-35-6 1.00
39 Fenofibrate Antilipemic to reduce cholesterol 49562-28-9 15100
40 Fexofenadine Antihistamine 83799-24-0 1.19
41 Fluconazole Antifungal 86386-73-4 1.29
42
Fluoxetine
*
Antidepressant/antianxiety (SSRI) 54910-89-3 5.46
43 Fluticasone propionate Corticosteroid 90566-53-3 22.1
44 Fluvoxamine Antidepressant/antianxiety (SSRI) 54739-18-3 15.7
45
Gemfibrozil
**
Antilipemic 25812-30-0 1.00
46 Glipizide Antidiabetic 29094-61-9 1.00
47 Glyburide Antidiabetic 10238-21-8 1.33
48 Hydrocodone Analgesic, antitussive 125-29-1 12.30
49 Hydrocortisone Anti-inflammatory 50-23-7 12.90
50 Hydroxyzine Antihistamine, sedative 68-88-2 32.7
51 Iminostilbene Intermediate for manufacture of carbamazepine 256-96-2 556
52
Ivermectin
**
Parasiticide 70288-86-7
60.0
c
53 Ketoconazole Antifungal 65277-42-1 646
54 Lamivudine Antiviral 134678-17-4 1.00
55 Lidocaine Topical anesthetic 137-58-6 17.3
56 Loperamide Anti-diarrheal 53179-11-6 519
57 Loratadine Antihistamine 79794-75-5 538
58 Lorazepam Anticonvulsant and muscle relaxant 846-49-1 38.0
59 Meprobamate Carbamate derivative, anxiolytic 57-53-4 2.00
60 Metaxal ne Muscle Relaxant 1665-48-1 4.65
61 Metformin Antidiabetic 657-24-9 1.00
62 Methadone Analgesic 76-99-3 47.4
63 Methocarbamol Muscle Relaxant 532-03-6 1.16
64 Methotrexate Antimetabolite and antifoliate drug used to treat cancer 59-05-2 1.00
65
Methylphenidate
**
Psychostimulant 298-59-9 1.02
66 Metoprolol Antihypertensive 51384-51-1 1.00
67 Morphine Analgesic 57-27-2 1.03
 
 
 
130 
68 Nadolol Antihypertensive, used to prevent migraines 42200-33-9 1.00
69 Nevirapine Antiviral 129618-40-2 59.7
70 Nicotine Stimulant 54-11-5 1.00
71 Nizatidine Antacid 76963-41-2 1.00
72 Nordiazepam Anti-anxiety metabolite of diazepam 1088-11-5 75.8
73 Norethindrone Contraceptive component 68-22-4 87.6
74
Norfluoxetine
*
Antidepressant 83891-03-6 35.2
75 Norverapamil Metabolite of verapamil 67018-85-3 5.99
76
Omeprazole + esomeprazole
b
Antacid 73590-58-6 34.3
77 Orlistat Anti-obesity 96829-58-2 123000
78 Oseltamivir Antiviral 196618-13-0 4.92
79 Oxazepam Anti-anxiety 604-75-1 28.4
80 Oxycodone Analgesic and antidiarrheal 76-42-6 6.42
81
Paroxetine
*
Antidepressant/antianxiety (SSRI) 61869-08-7 8.10
82 Penciclovir Antiviral 39809-25-1 1.00
83 Pentoxifylline Improves blood flow, cardiovascular drug 6493-05-6 1.00
84 Phenazopyridine Analgesic 94-78-0 28.4
85 Phendimetrazine Appetite suppressant drug 634-03-7 7.88
86 Phenytoin Antiepileptic 57-41-0 5.23
87 Piperonyl butoxide Pediculicides 51-03-6 2400
88 Prednisolone Corticosteroid 50-24-8 10.3
89 Prednisone Corticosteroid 53-03-2 9.12
90 Promethazine Antihistamine, antiemetic, sedative 60-87-7 358
91 Propoxyphene Analgesic 469-62-5 46.5
92
Propranolol
*
Antihypertensive 525-66-6 2.60
93
Pseudoephedrine + ephedrine
b
Decongestant, appetite suppressant, stimulant 90-82-4, 299-42-3 1.00, 1.00
94 Quinine Antimalarial, analgesic 130-95-0 4.34
95 Raloxifene Selective Estrogen Receptor Modulator 84449-90-1 327
96 Ranitidine Antacid 66357-35-5 1.00
97
Sertraline
*
Antidepressant/antianxiety (SSRI) 79617-96-2 142
98 Sitagliptin Antihyperglcemic 486460-32-6 19.9
99
Sucralose
**
Artifical Sweetener 56038-13-2 1.00
100 Sulfadimethoxine Antibiotic 122-11-2 1.00
101 Sulfamethizole Antibiotic 144-82-1 1.00
102
Sulfamethoxazole
*
Antibiotic 723-46-6 1.00
103 Tamoxifen Estrogen receopotor agonist used to treat breast cancer 10540-29-1 797
104 Temazepam Hypnontic 846-50-4 27.1
105 Theophylline Antiasthmatic, diuretic 58-55-9 1.00
106 Thiabendazole Parasiticide, fungicide 148-79-8 44.4
107 Tiotropium Bronchodilator 186691-13-4
.
108 Tramadol Analgesic 27203-92-5 1.16
109 Triamterene Diuretic 396-01-0 4.35
110
Trimethoprim
*
Antibiotic 738-70-5 1.50
111 Valacyclovir Antiviral 124832-26-4 1.00
112 Venlafaxine Antidepressant 93413-69-5 3.13
113 Verapamil Antihypertensive 52-53-9 65.7
114
Warfarin
*
Anticoagulant 81-81-2 1.00
a 
*=Subset of analytes quantified at Baylor University; **=analytes quantified at Baylor but not at USGS National Water Quality Laboratry 
b
Mixture of stereoisomers.
c
BCF predicted from SAR relationship not pH specific (Sanderson et al. 2007).
Supplemental Table S5 cont’d. List of 114 pharmaceutical compounds and sucralose 
for the screening-level exposure assessment, their CAS number, and bioconcentration 
factor (BCF). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
131 
Table S2. Analytes quantified in water and plasma from fish and ospreys, 
and their method detection limits (MDL).
Class Analyte
Water 
(ng/L)
Fish 
(ng/mL)
Osprey 
(ng/mL) 
Analgesics Acetaminophen 2.90 4.70 4.60
Codeine 0.830 6.90 8.10
Antibiotics Erythromycin 8.60 8.60 14.0
Sulfamethoxazole 1.30 1.90 4.00
Trimethoprim 1.30 2.80 0.40
Anticoagulant Warfarin 0.78 0.93 0.62
Antidepressant Fluoxetine 8.40 5.50 5.00
Norfluoxetine 8.50 3.90 7.40
Paroxetine 11.0 1.80 4.00
Sertraline 6.10 3.70 5.00
Desmethylsertraline 5.40 8.30 9.70
Antihistamine Diphenhydramine 0.22 0.12 0.18
Antihypertensive Atenolol 4.30 2.90 11.0
Diltiazem 0.24 0.12 0.35
Propranolol 1.80 0.58 1.90
Anti-inflammatories Celecoxib 11.0 7.90 8.40
Diclofenac 2.80 0.84 1.40
Antilipemic Gemfibrozil 2.10 6.90 8.00
Antiseizure Carbamazepine 0.53 0.53 0.74
Artificial Sweetener Sucralose 36.0 37.00 22.0
Parasiticide Ivermectin 93.0 61.0 140.0
Psychostimulant Diazepam 4.60 2.80 3.50
Methylphenidate 0.30 0.27 0.58
Stimulant Caffeine 4.50 4.70 7.80
Supplemental Table S6. Analytes quantified in water and plasma from fish and ospreys 
and their method detection limits (MDL). 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
132 
Table S3.  Analytes detected in water samples
a 
Poplar Island                   
(2011-13)                  
Detects/n                            
Mean “ std. dev.                                
Extremes           
Susquehanna River 2013               
(2 sites)                                    
Detects/n                                                               
Mean “ std. dev.                                                                              
Extremes
James River 2012              (3 
sties)                        Detects/n                    
Mean “ std. dev            
Extremes  
Anacostia/middle Potomac Rivers 
2011                                                      
(3 sites)                                                                                                
Detects/n                                                                         
Mean “ std.dev.                                  
Extremes  
Back River 2011                                                                          
(1 site)                              
Detects/n                             
Mean                                     
Extremes  
2/3 1/1
63.7-64.7
b
271
<MDL-174 -
1/1
219
-
1/3 1/1
- 461
<MDL-11.3 -
3/3 2/2 3/3 3/3 1/1
6.78 “ 3.51 3.26 “ 0.57 33.5 “ 10.6 35.6 “ 21.1 572
3.67-10.6 2.85-3.66 21.7-42.3 17.6-58.8 -
2/3 3/3 1/1
2.41-2.89
b
5.00 “ 1.68 83.0
<MDL-4.23 3.60-6.88 -
1/1
163
-
3/3 2/2 3/3 3/3 1/1
1.56 “ 0.82 5.20 “ 6.31 1.09 “ 0.60 5.40 “ 6.53 10.9
0.32-2.59 0.74-9.67 0.37-1.72 1.20-12.9 -
2/3 1/3 1/1
12.1-13.6
b
- 114
<MDL-21.1 <MDL-44.8 -
2/3 2/2 3/3 2/3
1/1
1.06-1.14
b
1.68 “ 0.45 5.86 “ 2.52 1.08-1.62
b
-
<MDL-2.05 1.36-2.00 2.97-7.51 <MDL-2.47 173
1/1
245
-
1/1
440
-
1/1 1/1
- 20.3
<MDL-12.2 -
3/3 3/3 1/1
4.77 “ 1.62 5.57 “ 3.18 39.5
2.97-6.11 3.55-9.23 -
3/3 2/2 3/3 3/3 1/1
3.69 “ 1.31
B
3.44 “ 0.88
B
14.4 “ 1.85
A,B
27.5 “ 21.7
A
220
2.79-5.22 2.81-4.06 12.3-15.88 7.20-50.5 -
1/3 3/3 3/3 1/1
- 2257 “ 548 2644 “ 1946 10249
ND-145 1717-2813 699-4590 -
1/1
177
-
1/3 1/1
- 196
<MDL-1.05 -
3/3 2/2 3/3 3/3 1/1
22.7 “ 10.8
B
13.5 “ 1.61
B
55.67 “ 14.5
AB
79.4 “ 21.5
A
255
11.7-33.3 12.3-14.6 40.2-69.1 57.1-100 -
a
 - indicates no mean calculated, contaminant was detected in fewer than half of the samples; MDL = method detection limit.  Means with different captial letter superscripts 
are significantly different (p<0.05). 
b
If non-detects were present in less than half of the samples, the Kaplan-Meier method was used to estimate the extremes of the mean followed by a generalized Wilcoxin 
non-parametric test.
Water Concentrations (ng/L)
Acetaminophen
Codeine
Propanolol
Atenolol
Diltiazem
Trimethoprim
Warfarin
Erythromycin
Caffeine
Sucralose
Analyte 
Sulfamethoxazole 
Diphenhydramine
Celecoxib
Diclofenac
Gemfibrozil
Carbamazepine
Diazepam
Methylphenidate
Supplemental Table S7. Analytes detected in water samples. 
 
 
 
 
 
 
 
 
 
133 
Supplemental Table S8. Concentrations of pharmaceuticals in fish plasma at each study 
site.  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Table S4. Concentrations of pharmaceuticals in fish plasma at each study site 
a
C
max 
(ng/mL)
30 50 100 30 710 2,000 NA 2,000
Catfish 
2/30 0/30 2/30
- - -
<MDL-0.42 - <MDL-2.23
0.01 0.001
2/2 1/2
0.89+0.46 -
0.57-1.22 <MDL-0.35
0.02 0.01
1/27 8/27 2/27
- - -
<MDL-0.51 <MDL-0.77 <MDL-10.9
0.01 0.03 0.01
9/18
-
<MDL-2.4
0.08
Gizzard Shad
10/33 1/33 3/33 0/33 4/33
- - - - -
<MDL-1.02 <MDL-6.82 <MDL-0.42 - <MDL-2301
0.02 0.07 0.01 -
4/9 8/9 1/9 1/9
- 0.69-0.70
c
- -
<MDL-67 <MDL-1.10 <MDL-0.42 <MDL-1.29
2.23 0.02 0.01 0.001
4/27 3/27
- -
<MDL-0.39 <MDL-0.57
0.01 0.02
0/18
-
-
Carp
6/18 1/18 0/18
- - -
<MDL-0.38 <MDL-0.33 -
0.01 0.01 -
1/9 0/9 1/9
- - -
<MDL-0.35 - <MDL-1.58
0.01 0.001
White Perch
1/5 0/5
- -
<MDL-0.8 -
0.02
3/10 1/10
- -
<MDL-0.31 <MDL-0.41
0.01 0.01
Rockfish
5/17 3/17 2/17
- - -
<MDL-0.98 <MDL-1.82 <MDL-26.3
0.02 0.06 0.04
Atlantic Menhaden
1/10
-
<MDL-0.41
0.02
a
 - indicates no mean calculated, contaminant was detected in fewer than half of the samples; MDL = method detection limit. Means with captial letter superscripts indicate a 
a significant difference (p < 0.05).  
b
Hazard Quotient (HQ) is the upper extreme concentration found in fish or osprey plasma divided by the human therapeutic plasma concentration (C
max
).
c
If non-detects were present in less than half of the samples, the Kaplan-Meier method was used to estimate the extremes of the mean followed by a generalized Wilcoxin 
non-parametric test.
Back River (n=5)
Poplar Island (n=10)
Poplar Island (n=17)
Poplar Island (n=10)
                                            C oncentrations in Fish Plasma (ng/mL)
Diphenydramine 
Detects/n             
Mean + std.dev.            
Extremes                   
HQ
b
Atenolol 
Detects/n               
Mean + std. dev.      
Extremes               
HQ
b
Diltiazem 
Detects/n             
Mean + std. dev.      
Extremes                
HQ
b
Celecoxib  Detects/n             
Mean + std.dev.          
Extremes              
HQ
b
Carbamazepine 
Detects/n                
Mean + std.dev.               
Extremes               
HQ
b
Sucralose 
Detects/n              
Mean + std.dev.  
Extremes                  
HQ
b
Caffeine    
Detects/n                   
Mean + std.dev.              
Extremes           
HQ
b
Codeine Detects/n               
Mean + std.dev.     
Extremes               
HQ
b
Analyte
Anacostia/middle Potomac 
Rivers (n=30)
Back River (n=2)
James River (n=27)
Susquehanna River (n=18)
Anacostia/middle Potomac 
Rivers (n=33)
Back River (n=9)
James River (n=27)
Susquehanna River (n=18)
Anacostia/middle Potomac 
Rivers (n=18)
Back River (n=9)
 
 
 
134 
 
 
 
 
 
 
 
 
 
CHAPTER 5 
OVERALL DISCUSSION 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
135 
Summary of Findings  
These three studies investigated spatial and temporal trends of contaminants in 
osprey eggs, their transfer in the food web, patterns of emerging pollutants and potential 
effects on ospreys.  Since their population nadir in the 1970’s, Chesapeake Bay ospreys 
have demonstrated resilience in the face of anthropogenic threats. These studies indicate 
that osprey productivity remains stable across three major Chesapeake Bay tributaries 
and Regions of Concern (greater than 0.8-1.15 fledglings/active nest). Contaminant 
residues in osprey eggs seem to be below adverse effect thresholds, although there was 
modest evidence of genetic damage in the most industrialized and urbanized areas of the 
Bay.   
All organochlorine pesticides, PCBs, and PBDE flame retardants were below 
thresholds that have been documented to cause an adverse effect in laboratory and 
controlled exposure studies (Elliott et al., 2001; Reviewed in Beyer and Meador, 2011; 
Zhang et al., 2014).  Eggshell thinning, which has been associated with the historical use 
of the pesticide DDT (specifically its metabolite p,p’-DDE)  was not documented in this 
study.  Current eggshell thickness measurements (0.514 ± 0.054 mm) have increased 
from the DDT use era (from first use in 1942 to its ban in 1972; 0.402-0.416 mm; 
Wiemeyer et al., 1988).  Thickness measurements from 2011-2013 are similar to those 
reported during the pre-DDT era (0.505, n=365; Anderson and Hickey, 1970).  This 
pattern is consistent with the discontinuation of the use of the organochlorine pesticide 
DDT in 1972, declining residues of p,p’-DDE in eggs, and the recovery of the osprey 
population in the Bay.   
 
 
 
136 
Total PCB concentrations in eggs declined slightly on the Anacostia/middle 
Potomac Rivers compared to 2000-2001, but remain unchanged in industrialized hotspots 
in Baltimore Harbor/Patapsco River.  Eggs from the Anacostia/middle Potomac Rivers 
had total PCB concentrations in 2011 that were ~40% lower than in 2000 (9.28 µg/g ww 
in 2000 vs. 5.56 µg/g ww in 2011).  These are similar to findings reported by Velinsky 
and coworkers (2011), who documented that residues in sediments declined by over an 
order of magnitude in the past 25 years.  As Velinsky and coworkers point out, reductions 
can be attributed to use bans (manufacture banned in the U.S. in 1979; USEPA, 2013) 
and better control at point source facilities (e.g., Haywood and Buchanan, 2007).  
However, concentrations of PCBs in osprey eggs continued to remain elevated in 
industrialized hotspots in Baltimore Harbor/Patapsco River (especially in Curtis and Bear 
Creeks).  Notably, the Coast Guard Yard in Curtis Creek was listed as a Superfund site in 
2002 due to contamination with PCBs, pesticides and dioxins (US EPA, 2013).  Although 
remediation efforts have been completed (as of 2013), PCBs continue to be localized in 
sediments from the areas around their discharge (Ashley et al., 2009).  Physical processes 
in Baltimore Harbor may help explain why contaminants are retained in this location and 
much of the sediment material in the Harbor doesn’t move (Ashley et al., 2009). Various 
solutions include capping or dredging the hotspots to remove contaminated sediments. 
However, there is still evidence in this study to indicate PCBs are biomagnifying 
(BMF=25.4), and it may take more time for this change to be completely reflected in the 
food web. 
In 2000-2001, PBDE residues were reported in bird eggs on the Chesapeake and 
other major estuaries around the U.S.  At that time, there were limited data to interpret 
 
 
 
137 
these values measured in the field.  A laboratory study conducted by McKernan and 
coworkers (2009) and other work conducted by Fernie and coworkers (2003, 2005, 2006, 
2008), provided information to place field data into perspective (Zhang et al., 2014).  
Since the last ecotoxicology studies in ROCs, PBDEs have been phased out (penta- and 
octa- formulations at end of 200 and deca- formulation at the end of 2009).  Production of 
the fully brominated deca-BDE 209 ceased at the end of 2012 (US EPA, 2015).  
Paralleling the phase out of BDE use, there has been an overall decline of flame-
retardants in ROCs.  However, they have only declined slightly near WWTPs and may 
continue to be associated with their processing capacity, upgrade status and river dilution.  
The highest relative concentration of PBDEs across study sites was detected on 
the middle Potomac (up to 800 ng/g ww) in the vicinity of the largest advanced WWTP 
in the world (Blue Plains WWTP, processing up to 370 million gallons per day, mgd).  
Residues on the middle Potomac followed a predictable spatial gradient.  Concentrations 
increased starting on the Anacostia, peaked at Blue Plains and then declined downriver.  
Concentrations around Blue Plains were similar to maximum total PBDEs in the northern 
segment of Delaware Bay (up to 820 ng/g ww; Toschik et al., 2005).  Similar patterns are 
reflected across other study sites.  Residues on Back River were the highest (666.1 ng/g 
ww and 616.1 ng/g ww) in two of the three eggs collected in vicinity of the City of 
Baltimore Back River WWTP (capacity of 180 mgd).  Next, followed eggs in Havre de 
Grace, MD on the Susquehanna River containing maximum total PBDE values of 648 
ng/g ww in the vicinity of the Havre de Grace WWTP (2 mgd).  The WWTP in Havre de 
Grace has recently completed its upgrade, so perhaps changes have not been reflected in 
the food web to date. Unfortunately, there are no avian monitoring data for PBDEs in this 
 
 
 
138 
tributary and future monitoring efforts on the Susquehanna would be valuable to track 
progress.  PBDE residues were low in Baltimore Harbor, but there was a slight peak in 2 
of the 3 nests sampled above the Key Bridge, potentially related to the Patapsco WWTP 
(processing 63 mgd).  Overall, the James and Elizabeth Rivers had low PBDE residues 
compared to these other sites.   This is surprising due to the abundance of WWTPs on the 
James River. 
Methoxytriclosan was detected in every egg sample on the Anacostia/middle 
Potomac Rivers.  Although values are low (consistent with a low BMF), they peaked 
around the Frederick Douglass Bridge on the Anacostia River (5.08 ng/g ww) and again 
downriver from the Blue Plains WWTP (7.40 ng/g ww).  Even though methoxytriclosan 
was detected at low concentrations and rarely detected at other study sites, this “signal” 
may indicate contributions from additional non-point sources on the Anacostia River (i.e., 
combined sewer overflows). Even the antihypertensive drug diltiazem discussed in 
Chapter 4 peaked on the Anacostia/middle Potomac Rivers thus consistent with the 
“signal” of a sewage dominated system.  Diltiazem was found in all 69 osprey plasma 
samples (540–8630 ng/L), with 41% of these samples exceeding maximum 
concentrations found in fish.  Diltiazem levels in fish and osprey plasma were below the 
human therapeutic plasma concentration (30,000 ng/L).  However, blood diltiazem 
concentrations between 2,500-8,000 µg/L have been reported in fatal case reports in 
humans (Roper et al., 1993).  These values are much higher than detected in osprey 
nestling plasma samples, and at this time there is no knowledge on effect thresholds of 
diltiazem in wildlife.  
 
 
 
139 
Studies at a regional scale that can document the occurrence and concentrations of 
these sewage-affiliated compounds in wildlife on a spatial scale can help reveal patterns 
and narrow down specific contributions of these emerging contaminants.  Effect 
thresholds for diltiazem, other personal care products, pharmaceuticals, are unknown in 
ospreys at this time, and there is no evidence to suggest adverse effects.   
 One study site that warranted special attention was Back River.  The first year 
nests were monitored in 2011, two eggs were collected on Back River, but no nestlings 
were produced.  Four nests were observed being built but never became active.  In 2012, 
there was a similar situation.  In 2013 a final effort was made on Back River.  Only a 
small number of nests (n=5) were found and productivity was low (1 fledgling/active 
nest). This limited us from conducting a robust analysis that would be adequate to 
compare to other study sites.  Notably, Back River sediment concentrations of PBDEs 
were elevated at Cox Point (Klosterhous, 2007), but we did not detect remarkable 
residues of contaminants in osprey eggs; however, fish (e.g., gizzard shad and carp) 
contained some of the greatest residues of organic contaminants and pharmaceuticals 
compared to other sites.  Interestingly, Back River also contained an osprey nestling with 
high levels of DNA damage relative to other study sites.  Compared to other Chesapeake 
Bay tributaries, Back River was an osprey ghost town.  Perhaps additional contaminants 
that were not examined or physical factors (water clarity, food availability and quality) 
limited birds from nesting at this site.  
 
 
 
 
 
 
140 
In Silico Tools for Assessment 
Other tools besides field studies can assist in identifying those chemicals that 
warrant further investigation.  Empirical analyses for pharmaceuticals can be costly and 
timely depending on the number of analytes, types and quantities of samples.  Over 1,453 
drugs have been introduced going back to 1840’s (Kinch et al., 2014); thus, it is critical to 
narrow this list down to the key pharmaceuticals that warrant closer examination.  An 
environmentally relevant in silico screening-level exposure assessment was used to 
evaluate the bioaccumulation potential of 113 pharmaceuticals and metabolites (a subset 
of drugs analyzed at the U.S Geological Survey National Water Quality Laboratory), and 
an artificial sweetener (sucralose). Hypothetical concentrations in water reflecting 
“wastewater effluent dominated” or “dilution dominated” scenarios were combined with 
pH-specific bioconcentration factors (BCFs) to predict uptake in fish.  Residues in fish 
and osprey food intake rate were used to calculate the daily intake (DI) of compounds by 
an adult female osprey.  Fourteen pharmaceuticals and a drug metabolite with a BCF 
greater than 100 and a DI greater than 20 µg/kg were identified as being most likely to 
exceed the adult human therapeutic dose (HTD). These 15 compounds were also 
evaluated in a 40 day cumulative dose exposure scenario using first-order kinetics to 
account for uptake and elimination. Assuming comparable absorption to humans, the 
half-lives (t1/2) for an adult osprey to reach the HTD within 40 days were calculated. For 
3 of these pharmaceuticals, the estimated t1/2 in ospreys was less than that for humans, 
and thus an osprey might theoretically reach or exceed the HTD in 3 to 7 days. 
Continuing to adapt these models based on new studies (field and lab) can assist in 
making robust predictions.   
 
 
 
141 
Examination of an Effects Hierarchy 
 In this study, effects of ospreys nesting in ROCs and other Bay tributaries were 
examined at genetic (oxidative damage to DNA), individual (body weight and culmen 
length) and population levels (osprey productivity number of fledglings/active nest).   
Historically, productivity rates have been a critical measure of monitoring effects of 
contaminants and declines and recovery of populations over temporal and spatial scales.  
Case studies of large-scale wildlife die-offs at the population level are certainly the most 
striking to natural resource managers and regulators.  However, by the time a population 
reaches a tipping point (local level vs. widespread issue), it may already be too late and 
intense recovery efforts must be undertaken, or they may not be possible.  
 Based on findings in this study the osprey population in Chesapeake Bay has 
adequate productivity rates to maintain a stable population, but there is subtle evidence of 
DNA damage in the most industrialized and urbanized areas. Unfortunately, there are no 
data to indicate the point at which genetic damage measured in this study translates from 
a short term repairable response (i.e., DNA repair mechanisms kick in, or presence of 
adaptations to counter this damage) to an irreversible effect.  Major questions emerge 
including: 1) how do you study effects on wildlife when the population is thriving? and 
2) what do effects at the macromolecular level mean? Thus, perhaps a re-examination of 
the hierarchy effects is warranted.  The hierarchy of effects (Figure 11) indicates how a 
perturbation at one level is translated to changes in higher levels. 
Often it is challenging from both a research and economic standpoint to justify 
and show clear and meaningful effects on the left side of the hierarchy.  Indicating the 
presence of increased DNA damage in osprey nestlings in industrialized areas is one step 
 
 
 
142 
towards completing the picture.  Although subtle, oxidative DNA damage may have an 
effect in the long term health of an individual or even in subsequent generations.  The 
application of the adverse outcome pathway (Ankely et al., 2010) may help place our 
findings into perspective.  Research gaps remain to find a different type of canary in the 
coal mine, but this at sublethal levels, which is much different than the previous train of 
thought.  
 
 
 
 
 
 
 
 
Figure 11. Gradient illustrating the hierarchy of effects and when 
macromolecular effects may be meaningful. 
 
Utility of these Data  
Our data can serve to assist the Chesapeake Bay Program (composed of federal, 
academic, state and non-profit partners) in the assessing of the progress and effectiveness 
of existing reduction strategies (i.e., TMDLs) and the development of new ones.  These 
studies address many of the issues discussed in the Bay Program’s 2015 Toxic 
Contaminants Research Outcome Management Strategy (CBP, 2015). Once these data 
are gathered, the Bay Program agencies can then begin to focus on reduction, remediation 
and prevention. The Chesapeake Bay Program not only addressed legacy contaminants, 
 
 
 
143 
but also new classes of contaminants including PBDEs, alt-BFRs and pharmaceuticals 
(CBP, 2015). 
 
Final Conclusions 
Overall, contaminants in Chesapeake Bay have declined in osprey eggs; however, 
issues at specific sites remain the same.  Although the osprey population is stable, we 
have detected evidence of modest genetic damage.  Unfortunately, at this time there is 
limited knowledge on the implications of increased oxidative DNA damage in wildlife.  
Additional lab and field studies may assist us in how to interpret these effects and at what 
point they may translate to something that is ecologically meaningful.  Future directions 
may include the examination of complete blood counts (i.e., white blood cells and 
lymphocytes, Vitamin A) from those birds exhibiting the highest levels of DNA damage 
to indicate potential adverse effects in higher levels of the hierarchy (Figure 11). 
In particular, for pharmaceuticals, limited effect threshold data for birds and other 
vertebrates makes this interpretation challenging.  Additional in situ studies of 
pharmaceuticals in the vicinity of sewage outfalls may be warranted to identify other 
compounds that may be bioaccumulating either in nestling blood or osprey eggs.  More 
robust knowledge of species sensitivities (i.e., birds compared to humans) will assist in 
interpretation of these data in future ecological risk assessments and development of 
management strategies.    
  
 
 
 
 
 
144 
Compiled References 
[ACS] American Chemical Society, 2014. SciFinder substance identifier. Available. 
http://www.scifinder-cas-org (accessed 02.19.2014).  
 
Albina, M.L., Alonso, V., Linares, V., Belles, M., Sirvent, J.J., Domingo, J.L., Sanchez, 
D.J., 2010. Effects of exposure to BDE-99 on oxidative status of liver and kidney in adult 
rats. Toxicology. 271, 51-56. 
 
Anderson, D.W., Hickey, J.J., 1972. Eggshell changes in certain North American birds. 
Proc. Int. Ornithol. Cong. 15, 514-540.  
Ankley, G.T., Bennett, R.S., Erickson, R.J., Hoff, D.J., Hornung, M.W., Johnson, R.D., 
Mount, D.R., Nichols, J.W., Russom, C.L., Schmieder, P.K., Serrrano, J.A., Tietge, J.E., 
Villeneuve, D.L., 2010. Adverse outcome pathways: a conceptual framework to support 
ecotoxicological research and risk assessment. Environ. Toxicol. Chem.29, 730-741.  
Arnold, K.E., Boxall, A.B.A., Brown, A.R., Cuthbert, R.J., Gaw, S., Hutchinson, T.H., 
Jobling, S., Madden, J.C., Metcalfe, C.D., Naidoo, V., Shore, R.F., Smits, J.E., Taggart, 
M.A., Thompson, H.M.,2013. Assessing the exposure risk and impacts of 
pharmaceuticals in the environment on individuals and ecosystems. Biol. Lett. 9, 1–4. 
Alvarez, D.A., Cranor, W.L., Perkins, S.D., Schroeder, V.L., Iwanowicz, L.R., Clark, 
R.C., Guy, C.P., Pinkney, A.E., Blazer, V.S., Mullican, J.E., 2009. Reproductive health 
of bass in the Potomac, USA, drainage: Part 2. Seasonal occurrence of persistent and 
emerging organic contaminants. Environ. Toxicol. Chem. 28, 1084-1095. 
 
Ashley, J.T.F., Baker, J.E., 2009. Hydrophobic organic contaminants in surficial 
sediments of Baltimore harbor: Inventories and sources. Environ. Toxicol. Chem. 18, 
838-849. 
 
Audet, D.J., Scott, D.S., Wiemeyer, S.N., 1992. Organochlorines and mercury in osprey 
eggs from the eastern United States. J. Raptor Res. 26, 219-224. 
 
Baltimore County Watershed Management Program, 2012. Back River Watershed. 
Available. http://www.baltimorecou ntymd.gov/Agencies/environment/waters 
heds/backmain.html (accessed 02.19.14).  
 
Bardal, S., Waechter, J., Martin, D., 2011. Applied Pharmacology. Elsevier Saunders. St. 
Louis, MO, USA. 
Bart, J., Robson, D.S., 1982. Estimating survivorship when the subjects are visited 
periodically. Ecology 63, 1078-1090.  
 
Barron, M.G., 2003. Bioaccumulation and bioconcentration in aquatic organisms. In: 
Hoffman, D.J., Rattner, B.A., Burton, G.A., Cairns, J. Handbook of Ecotoxicology. 
Lewis Publishers. New York, NY pp. 877-892. 
 
 
 
145 
Bean, T.G., Boxall, A.B.A., Lane, J., Herborn, K.A., Pietravalle, S., Arnold, K.E., 2014. 
Behaviorual and physiological responses of birds to environmentally relevant 
concentrations of an antidepressant. Phil. Trans. R. Soc. B. 369, http://dx.doi.org 
/10.1098/rstb.2013.0575 
 
Bellanca, M.A., Bailey, D.S., 1977. Effects of chlorinated effluents on aquatic 
ecosystems in the lower James River. J. Water Pollut. Control Fed. 49, 639-645. 
 
Berninger, J.P., Du, B., Connors, K.A., Eytcheson, S.A., Kolkmeier, M.A., Prosser, K.N., 
Valenti, T.W., Chambliss, C.K., Brooks, B.W., 2011. Effects of the antihistamine 
diphenhydramine on selected aquatic organisms. Environ. Toxicol. Chem. 30, 2065-
2072. 
 
Beyer, W.N., Day, D., Morton, A., Pachepsky, Y., 1998. Relation of lead exposure to 
sediment ingestion in mute swans on Chesapeake Bay, USA. Environ. Toxicol. Chem. 
17, 2298-2301. 
 
Beyer, A., Biziuk, M., 2009. Environmental fate and global distribution of 
polychlorinated biphenyls. Rev. Environ. Contam. Toxicol. 201, 137-158.  
 
Beyer, W.N., Meador, J.P., 2011. Environmental Contaminants in Biota. 2nd Ed. CRC 
Press, New York, NY, USA.  
 
Blazer, V.S., Iwanowicz, L.R, Iwanowicz, D.D., Smith, D.R., Young, J.A., Hedrick, J.D., 
Foster, S.W., Reeser, S.J., 2007. Intersex (testicular oocytes) in smallmouth bass 
Micropterus dolomieu from the Potomac River and selected nearby drainages. J. Aquat. 
Anim. Health. 19, 242-253. 
 
Blus, L.J., 1984. DDE in birds' eggs: comparison of two methods for estimating critical 
levels. Wilson Bull. 96, 268-276.  
Blus, L.J., 2011. DDT, DDD, and DDE in birds. In: Beyer, W.N., Meador, J.P. (Eds.), 
Environmental Contaminants in Biota: Interpreting Tissue Concentrations, 2nd ed. CRC 
Press, Taylor & Francis Group, Boca Raton, FL, USA, pp. 425-444.  
Boxall, A.B.A., Rudd, M.A., Brooks, B.W., Caldwell, D.J., Choi, K., Hickmann, S., 
Innes, E., Ostapyk, K., Staveley, J.P., Verslycke, T., et al., 2012. Pharmaceuticals and 
personal care products in the environment: What are the big questions? Environ. Health 
Perspect.120. 1221-1229.  
 
Brausch, J.M., Connors, K.A., Brooks, B.W., Rand, G.M., 2012. Human pharmaceuticals 
in the aquatic environment: A recent review of recent toxicological studies and 
considerations for toxicity testing. Rev. Environ. Contam. Toxicol. 218, 1-99. 
 
 
 
 
146 
Brodin, T., Piovano, S., Fick, J., Klaminder J., Heynen, M., Jonsson, M., 2014. 
Ecological effects of pharmaceuticals in aquatic systems-impacts through behavioral 
alterations. Phil. Trans. R. Soc. B. 396, 20130580.  
 
Brooks, B.W., Chambliss, K.C., Stanley, J.K., Ramirez, A., Banks, K.E., Johnson, R.D., 
Lewis, R.J. 2005. Determination of select antidepressants in fish from an effluent-
dominated stream. Environ. Toxicol. Chem. 24, 464-469. 
 
Brooks, B.W., Riley, T.M., Taylor, R.D., 2006. Water quality of effluent-dominated 
stream ecosystems: Ecotoxicological, hydrological, and management considerations. 
Hydrobiologia 556, 365-379. 
 
Brooks, B.W., 2014. Fish on Prozac (and Zoloft): Ten years later. Aquat. Toxicol. 151, 
61–67. 
 
Brown, J.N., Paxéus, N., Förlin, L., Larsson, D.G.J., 2007. Variations in bioconcentration 
of human pharmaceuticals from sewage effluents into fish blood plasma. Environ. 
Toxicol. Pharmacol.24, 267–274. 
 
Carro, T., Dean, K., Ottinger, M.A., 2013a. Effects of an environmentally relevant 
polychlorinated biphenyl (PCB) mixture on embryonic survival and cardiac development 
in the domestic chicken. Environ. Toxicol. Chem. 32, 1325-1331. 
Carro, T., Taneyhill, L.A., Ottinger, M.A., 2013b. The effects of an environmentally 
relevant 58-congener polychlorinated biphenyl (PCB) mixture on cardiac development in 
the chick embryo. Environ. Toxicol. Chem. 32, 1317-1324.  
Chen, D., LaGuardia, M.J., Harvey, E., Amaral, M., Wohlfort, K., Hale, R.C., 2008. 
Polybrominated diphenyl ethers in peregrine falcon (Falco peregrinus) eggs from the 
Northeastern U.S. Environ. Sci. Technol. 42, 7594-7600.  
Chen, D., Hale, R.C., 2010. A global review of polybrominated diphenyl ether flame 
retardant contamination in birds. Environ. Int. 36, 800-811.  
Chen, D., Hale, R.C., Watts, B.D., LaGuardia, M.J., Harvey, E., Mojica, E.K., 2010. 
Species-specific accumulation of polybrominated diphenyl ether flame re- tardants in 
birds of prey from the Chesapeake Bay region, USA. Environ. Pollut. 158, 1883-1889.  
Chesapeake Bay Executive Order, 2009. Chesapeake Bay Protection and Restoration. 
Office of the Press Secretary. Available. https://www.whitehouse.gov/the-press-
office/executive-order-chesapeake-bay-protection-and-restoration. (accessed 03.27.15).  
 
 
 
 
 
 
147 
(CBP) Chesapeake Bay Program, 1994. Chesapeake Bay basinwide toxics reduction 
strategy reevaluation report. U.S. EPA Chesapeake Bay Program: Annapolis, MD. 
CBP/TRS 117/94. 
 
2000. Chesapeake 2000 Agreement. 28 June 2000. Annapolis, MD. 13pp. 
 
2012. Map: Chemical Contaminants. Available. http://www.chesa 
peakebay.net/maps/map/chemical_contaminants_2012.(accessed 09.02.15.).  
 
2015. Toxic Contaminants Research Outcome. Available. 
http://www.chesapeakebay.net/documents/22047/3d_toxics_research_5-22-
15_ff_formatted.pdf. (accessed 09.09.2015).  
 
Cooke, M.S., Evans, M.D., Dizdaroglu, M., Lunec, J., 2003. Oxidative DNA damage: 
mechanisms, mutation and disease.  FASEB J. 17, 1195-1214.  
Costantini, D., 2008. Oxidative stress in ecology and evolution: lessons from avian 
studies. Ecol. Lett. 11, 1238-1251.  
Covaci, A., Harrad, S., Mohamed, A.E., Nadeem, A., Law, R.J., Herzke D., Wit, C.A., 
2010. Novel brominated flame retardants: a review of their analysis, environmental fate 
and behavior. Environ. Int. 37, 532-556. 
 
Custer, T.W., Custer, C.M., Hines, R.H., Sparks, D.W., Melancon, M.J., Hoffman, D.J., 
Bickham, J.W., Wickliffe, J.K., 2000. Mixed-function oxygenases, oxidative stress, and 
chromosomal damage measured in lesser scaup wintering on Indiana Harbor Canal. 
Arch.Environ. Toxicol Chem. 38, 522-529.  
 
Daughton, C.G., Ternes, T.A. 1999. Pharmaceuticals and personal care products in the 
environment: agents of subtle change? Environ. Health Persp. 107, 907-938.  
 
de Wit, C.A., Kierkegaard, A., Ricklund, N., Sellström, U., 2011. Emerging brominated 
flame retardants in the environment. In: Eljarrat, E., Barceló, D. (Eds.), Brominated 
Flame Retardants, The Handbook of Environmental Chemistry, vol. 16. Heidelberg, 
Germany, pp. 241-286.  
Devaux, A., Flammarion, P., Bernardon, V., Garric, J., Monod, G., 1998. Monitoring of 
the chemical pollution of the River Rhône through measurement of DNA damage and 
cytochrome P4501A induction in chub (Leuciscus cephalus). Mar. Environ. Res. 46, 257- 
262.  
 
District of Columbia Water and Sewage Authority, 2014. Blue Plains advanced 
wastewater treatment plant. Available. http://www.dcwater.com /wastewater 
/blueplains.cfm (accessed 02.19.2014.). 
Drugsite Trust, 2014. Drug information database. Available. http://www.drugs.com 
(accessed 02.19.2014.). 
 
 
 
148 
 
Du, B., Perez-Hurtado, P., Brooks, B.W., Chambliss, C.K., 2012. Evaluation of an 
isotope dilution liquid chromatography tandem mass spectrometry method for 
pharmaceuticals in fish. J. Chrom. A. 1253, 177-183.  
 
Du, B., Price, A.E., Scott, W.C., Kristofco, L.A., Ramirez, A.J., Chambliss, C.K., 
Yelderman, J.C., Brooks, B.W., 2014. Comparison of contaminants of emerging concern 
removal, discharge, and water quality hazards among centralized and on-site wastewater 
treatment system effluents receiving common wastewater influent. Sci.Tot. Environ. 
466–467, 976–984. 
Ebadi, M., 2008. Desk reference of clinical pharmacology. 2nd ed. CRC Press, Inc. Boca 
Raton, FL, USA.  
Elliott, J.E., Machmer, M.M., Henny, C.J., Wilson, L.K., Norstrom, R.J., 1998. 
Contaminants in ospreys from the Pacific Northwest: I. Trends and patterns in 
polychlorinated dibenzo-p-dioxins and –dibenzofurans in eggs and plasma. Arch. 
Environ. Contam. Toxicol. 35, 620-631.  
 
Elliott, J.E., Wilson, L.K., Henny, C.J., Machmer, M.M., 2000. Contaminants in ospreys 
from the Pacific Northwest: II. Patterns and trends in organochlorine pesticides, 
polychlorinated biphenyls and mercury in eggs and plasma. Arch. Environ. Contam. 
Toxicol. 38, 93-106. 
 
Elliott, J.E., Wilson, L.K., Henny, C.J., Trudeau, S.F., Leighton, F.A., Kennedy, S.W., 
Cheng, K.M., 2001. Assessment of biological effects of chlorinated hydrocarbons in 
osprey chicks. Environ. Toxicol. Chem. 20, 866-879.  
 
Elliott, J.E., Miller, M.J., Wilson, L.K., 2005. Assessing breeding potential of peregrine 
falcons based on chlorinated hydrocarbon concentrations in prey. Environ. Pollut. 134, 
353-361.  
 
Elliott, J.E., Morrissey, C.A., Henny, C.J., Inzunza, E.R., Shaw, P., 2007. Satellite 
telemetry and prey sampling reveal contaminant sources to Pacific Northwest ospreys. 
Ecol. Appl. 17, 1223-1233. 
 
Elizabeth River Project, 2008. State of the Elizabeth River. 22p. http://elizabethriver. 
org/Publi cations/ERP% 20Pubs/State_of_River_2008.pdf.  
 
Ernst, H.R., 2003. Chesapeake Bay Blues: Science, Politics, and the Struggle to Save the 
Bay. Rowman and Littlefield Publishers. New York, NY, USA. 
 
Eulaers, I., Jaspers, V.L.B., Halley, D.J., Lepoint, G., Nygård, T., Pinxten, R., Covaci, A., 
Eens, M., 2014. Brominated and phosphorous flame retardants in white-tailed eagle 
Haliaeetus albicilla nestlings: bioaccumulation and associations with dietary proxies 
(d13C, d15N and d34S). Sci. Tot. Environ. 478, 48-57.  
 
 
 
149 
 
Ewins, P.J., Postupalsky, S., Hughes, K.D, Weseloh, D.V., 1999. Organochlorine 
contaminant residues and shell thickness of eggs from known-age female ospreys 
(Pandion haliaetus) in Michigan during the 1980s. Environ. Pollut. 104: 295-304. 
 
Fernie, K., Bortolotti, G., Drouillard, K., Smits, J., Marchant, T., 2003. Developmental 
toxicity of in ovo exposure to polychlorinated biphenyls. II. Effects of maternal or 
paternal exposure on second-generation nestling American kestrel. Environ. Toxicol. 
Chem. 88, 375-383.  
 
Fernie, K.J., Shutt, J.L., Mayne, G., Hoffman, D., Letcher, R.J., Drouillard, K.G., Ritchie, 
I.J., 2005. Exposure to polybrominated diphenyl ethers (PBDEs): changes in thyroid, 
vitamin A, glutathione homeostasis, and oxidative stress in American kestrels (Falco 
sparverius). Toxicol. Sci. 88, 375-383.  
Fernie, K.J., Shutt, J.L., Ritchie, I.J., Letcher, R.J., Drouillard, K., Bird, D.M., 2006. 
Change in the growth, but not the survival, of American kestrels (Falco sparverius) 
exposed to environmentally relevant polybrominated diphenyl ethers. J. Toxicol. 
Environ. Health A. 69, 1541-1554. 
Fernie, K.J., Shutt, J.L., Letcher, R.J., Ritchie, I.J., Sullivan, K., Bird, D.M., 2008. 
Changes in reproductive courtship behaviors of adult American kestrels (Falco sparver- 
ius) exposed to environmentally relevant levels of the polybrominated diphenyl ether 
mixture, DE-71. Toxicol. Sci. 102, 171-178.  
[FDA] Food and Drug Administration.,  2012. FDA approved drug products. Available.  
http://www.accessdata.fda.gov/scripts/cder/drugsatfda/index.cfm (accessed 09.01.2014.).  
 
Fick, J., Lindberg, R.H., Parkkonen, J., Arvidsson, B., Tysklind, M., Larsson, D.G., 
2010a. Therapeutic levels of levonorgesterol detected in blood plasma of fish: results 
from screening rainbow trout exposed to treated sewage effluents.  Environ. Sci. Techol.  
44, 2661-2666. 
 
Fick, J., Lindberg, R.H., Tysklind, M., Larsson, D.G.J., 2010b. Predicted critical 
environmental concentrations for 500 pharmaceuticals. Regul. Toxicol. Pharmacol. 58, 
516–523. 
 
Furlong, E.T., Noriega, M.C., Kanagy, C.J., Kanagy, L.K., Coffey, L.J., Burkhardt, M.R., 
2014. Determination of human-use pharmaceuticals in filtered water by direct aqueous 
injection—High-performance liquid chromatography/tandem mass spectrometry: US 
Geological Survey techniques and methods. Book 5, chapter B10. Available. 
http://dx.doi.org/10.3133/tm5b10 (accessed 09.29.2014.).  
 
Furness, R.W., Greenwood, J.J.D, Jarvis, P.J., 1993. Can birds be used to monitor the 
environment? In: Furness, R.W., Greenwood, J.J.D. (Ed.), Birds as Monitors of 
Environmental Change. Chapman & Hall, New York, NY, USA. pp. 1-143.  
 
 
 
 
150 
Fu, W., Franco, A, Trapp, S., 2009. Methods for estimating the bioconcentration factor of 
ionizable chemicals. Environ. Toxicol. Chem. 28, 1372-1379.  
 
Gauthier, L.T., Hebert, C.E., Weseloh, D.V.C., Letcher, R.J., 2007. Current-use flame 
retardants in the eggs of herring gulls (Larus argentatus) from the Laurentian Great Glass 
K.A., and B.D. Watts. 2009. Osprey diet composition and quality in high- and low-
salinity areas of lower Chesapeake Bay. J. Rapt. Res. 43, 27-36. 
 
Glass, K.A., Watts, B.D., 2009. Osprey diet composition and quality in high and low 
salinity areas of lower Chesapeake Bay. J. Raptor Res. 43, 27–36. 
 
Golden N.H., Rattner B.A., 2003. Ranking terrestrial vertebrate species for utility in 
biomonitoring and vulnerability to environmental contaminants. Environ. Contam. 
Toxicol. 176, 67-136. 
 
Grove, R.A., Henny, C.J., Kaiser, J.L., 2009. Osprey: worldwide sentinel species for 
assessing and monitoring environmental contamination in rivers, lakes, reservoirs and 
estuaries. J. Toxicol. Environ. Health B. 12, 25-44.  
Guerra, P., Fernie, K., Jiménez, B., Pacepavicius, G., Shen, L., Reiner, E., Eljarrat, E., 
Barcelo, D., Alaee, M., 2012. Dechlorane plus and related compounds in peregrine falcon 
(Falco peregrins) eggs from Canada and Spain. Environ. Sci. Technol. Lett. 45, 1284-
1290.  
 
Harris, S.A., Jones, J.L., 2008. Fish consumption and PCB-associated health risks in 
recreational fishermen on the James River, Virginia. Environ. Res. 107, 254-263. 
 
Harris, M.L., Elliott, J.E., 2011. Effects of polychlorinated biphenyls, dibenzo-p-dioxins 
and  dibenzofurans, and polybrominated diphenyl ethers in wild birds. In: Beyer, 
W.N.,Meador, J.P. (Eds.). Environmental Contaminants in Biota. CRC Press. Boca 
Raton, FL.pp. 447-530.  
 
Haywood, H.C., Buchanan, C., 2007. Total maximum daily loads of polychlorinated 
biphenyls (PCBs) for tidal portions of the Potomac and Anacostia rivers in the District of 
Columbia, Maryland, and Virginia. Interstate Commission on the Potomac River Basin. 
ICPRB Report 07-7. Rockville, MD. 
 
He, P., He, W., Wang, A., Xia, T., Xu, B., Zhang, M., Chen, X., 2008. PBDE-47-induced 
oxidative stress, DNA damage and apoptosis in primary cultured rat hippocampal 
neurons. Neurotoxicology  29, 124-129. 
 
Heinz, G.H., Stebbins, K.R., Klimstra, J.D., Hoffman, D.J., 2009a. A simplified method 
for correcting contaminant concentrations in eggs for moisture loss. Environ. Toxicol. 
Chem. 28, 1425-1428.  
Helsel, D.R., 2005. Nondetects and Data Analysis Statistics for Censored Environ- 
 
 
 
151 
mental Data. John Wiley & Sons, Inc., Hoboken, NJ, USA. 
Helsel, D.R. 2009. Summing nondetects: Incorporating low-level contaminants in risk 
assessment. Integr.  Environ. Assess. Manag. 6, 361–366.   
Henny, C.J., Smith, M.M., Stotts, V.D., 1974., The 1973 distribution and abundance of 
breeding ospreys in the Chesapeake Bay. Chesap. Sci. 15, 125-133.  
Henny, C.J., Kaiser, J.L., 1996. Osprey population increase along the Willamette River, 
Oregon and the role of utility structures, 1976-1993. In: Bird, D.M., Varland, D.E., 
Negro, J.J. (Eds.), Raptors in Human Landscapes. Academic Press Ltd., London, pp. 97-
108.  
Henny, C.J., Kaiser, J.L., Grove, R.A., Bently, V.R., Elliott, J.E., 2003. Biomagnification 
factors (fish to osprey eggs from Willamette River, Oregon, U.S.A.) for PCDDs, PCDFS, 
PCBS and OC pesticides. Environ. Monit. Assess. 84, 275-315.  
Henny, C.J., Grove, R.A., Kaiser, J.L., Bentley, V.R., 2004. An evaluation of osprey eggs 
to determine spatial residue patterns and effects of contaminants along the lower 
Columbia River, U.S.A. In: Chancellor, R.D., Meyburg, B.U. (Eds.), Raptors Worldwide. 
WWGBP/MME, Budapest, Hungary, pp. 369-388.  
 
Henny, C.J., Kaiser, J.L., Grove, R.A., Johnson, B.L., Letcher, R.J., 2009. 
Polybrominated diphenyl ether flame retardants in eggs may reduce reproductive success 
of ospreys in Oregon and Washington, USA. Ecotoxicology. 18, 802-813.  
Henny, C.J., Grove, R.A., Kaiser, J.L., Johnson, B.L., 2010. North American osprey 
populations and contaminants: historic and contemporary perspectives. J. Toxicol. 
Environ. Health B. 13, 579-603.  
Henny, C.J., Grove, R.A., Kaiser, J.L., Johnson, B.L., Furl, C.V., Letcher, R.J., 2011. 
Wastewater dilution index partially explains observed polybrominated flame retardant 
concentrations in osprey eggs from Columbia River Basin, 2008-2009. Ecotoxicology. 
20, 682-697. 
Hernout, B.V., Arnold, K.E., McClean, C.J., Grimm, V., Boxall, A.B.A., 2011. 
Predicting the threats of chemicals to wildlife: What are the challenges? Integr. Environ. 
Assess. Manag. 7, 499–506. 
 
Hoffman, D.J., Melancon, M.J., Klein, P.N., Eisemann, J.D., Spann, J.W., 1998. 
Comparative developmental toxicity of planar polychlorinated biphenyl congeners in 
chickens, American kestrels, and common terns. Environ. Toxicol. Chem. 17, 747-757. 
 
Hoffman, D.J., Eagles-Smith, C.A., Ackerman, J.T., Adelsbach, T.L., Stebbins, K.R. 
2011.  Oxidative stress response of Forster’s terns (Sterna forsteri) and Caspian terns 
(Hydroprogne caspia) to mercury and selenium bioaccumulation in kidney, liver, and 
brain. Environ. Toxicol. Chem. 30, 920-929.  
 
 
 
152 
 
Huggett, D.B., Cook, J.C., Ericson, J.F., Williams, R.T., 2003. A theoretical model for 
utilizing mammalian pharmacology and safety data to prioritize potential impacts of 
human pharmaceuticals to fish. Hum. Ecol. Risk Assess. 9, 1789–1799.   
 
Interstate Commission on the Potomac River Basin, 1997. Land use in the Potomac 
Basin. Available. http://www.potomacriver.org/cms/images/maps_images (accessed 
09.01.2014.). 
 
Iwanowicz, L.R., Blazer, V.S., Guy, C.P., Pinkney, A.E., Mullican, J.E., Alvarez, D.A.,  
2009. Reproductive health of bass in the Potomac, USA, drainage: Part 1. Exploring the 
effects of proximity to wastewater treatment plant discharge. Environ. Toxicol Chem. 28, 
1072-1083. 
 
Johnson, B.L., Kaiser, J.L., Henny, C.J., Grove, R.A., 2008. Prey of nesting ospreys on 
the Willamette and Columbia Rivers, Oregon and Washington. Northwest Sci. 82, 229-
236. 
 
Kelly, B.C., Ikonomou, M.G., Blair, J.D., Morin, A.E., Gobas, F.A.P.C., 2007. Food 
web- specific biomagnification of persistent organic pollutants. Science 317, 236-239.  
 
Kennedy, R.S., 1977. The status of the osprey in Tidewater, Virginia, 1970-71. In: 
Ogden, J.C. (Ed.), Transactions and Proceedings No. 2 of the North American Osprey 
Research Conference. College of William and Mary, Williamsburg, VA. pp.121-133.  
 
Kennedy, S.W., Lorenzen, A., Jones, S.P., Hahn, M.E., Stegeman, J.J., 
1996. Cytochrome P4501A induction in avian hepatocyte cultures: a promising approach 
for predicting the sensitivity of avian species to toxic effects of halogenated aromatic 
hydrocarbons. Toxicol. Appl Pharm. 141, 214-230. 
 
Kim, Y., Choi, K., Jung, J., Park, S., Kim, P.G., Park, J., 2007. Aquatic toxicity of 
acetaminophen, carbamazepine, cimetidine, diltiazem and six major sulfonamides, and 
their potential ecological risks in Korea. Environ. Int. 33, 370–375. 
 
Kim, W-K., Lee, S-K., Jung, J., 2010. Integrated assessment of biomarker response in 
common carp (Cyprinus carpio) exposed to perfluorinated organic compounds. J Hazard. 
Matr. 180, 395-400. 
 
Kinch, M.S., Haynesworth, A., Kinch, S.L., Hoyer, D., 2014. An overview of FDA-
approved new molecular entities: 1827-2013. Drug Disc. Today. 19, 1033-1039. 
 
 
Klosterhaus, S., 2007., The Bioavailability of brominated diphenyl ethers from urban 
estuarine sediments to deposit feeding invertebrates. Ph.D. Dissertation. University of 
Maryland. 
 
 
 
 
153 
Kolpin, D.W., Furlong, E.T., Meyer, M.T., Thurman, E.M., Zaugg, S.D., Barber, L.B., 
Buxton, H.T., 2002. Pharmaceuticals, hormones, and other wastewater contaminants in 
U.S. streams 1999-2000: A national reconnaissance. Environ. Sci. Technol. 36, 1202-
1211. 
 
Kolpin, D.W., Skopec, M., Meyer, M.T., Furlong, E.T., Zaugg, S.D., 2004. Urban 
contribution of pharmaceuticals and other organic wastewater contaminants to streams 
during differing flow conditions. Sci. Total Environ.  328, 119–130. 
 
Kubiak, T.J., Harris, H.J., Smith, L.M., Schwartz, T.R., Stalling, D.L., Trick, J.A., Sileo, 
L., Docherty, D.E., Erdman, T.C., 1989. Microcontaminants and reproductive impairment 
of the Forster’s tern on Green Bay, Lake Michigan-1983. Arch. Environ. Contam. 
Toxicol. 18, 705-727. 
 
Küster, A., Adler, N., 2014. Pharmaceuticals in the environment: scientific evidence of 
risks and its regulation. Philos. Trans. R. Soc. B. 369, doi: 10.1098/rstb.2013.0587 
 
La Guardia, M.J., Hale, R.C., Harvey, E., 2007. Evidence of debromination of deca- 
bromodiphenyl ether (BDE-209) in biota from a wastewater receiving stream. Environ. 
Sci. Technol. 41, 6663-6670.  
La Guardia, M.J., Hale, R.C., Harvey, E., Chen, D., 2010. Flame-retardants and other 
organohalogens detected in sewage sludge by electron capture negative ion mass 
spectrometry. Environ. Sci. Technol. 44, 4658-4664.  
La Guardia, M.J., Hale, R.C., Newman, B., 2013. Brominated flame-retardants in sub- 
Saharan Africa: burdens in inland and coastal sediments of the eThekwini metropolitan 
municipality, South Africa. Environ. Sci. Technol. 47, 9643-9650. 
Lajeunesse, A., Gagnon, C., Sauvé, S., 2008. Determination of basic antidepressants and 
their N-desmethyl metabolites in raw sewage and wastewater using solid-phase extraction 
and liquid chromatography-tandem mass spectrometry. Anal.  Chem.   80, 5325-5333.  
Langland, M.J., 1998. Importance of the Susquehanna River and Its Reservoir System to 
the Chesapeake Bay Fact Sheet 003-98. http://pa.water.usgs.gov/reports/fs003-98.html. 
 
Lazarus, R.S., Rattner, B.A., McGowan, P.C., Hale, R.C., Schultz, S.L., Karouna-Renier, 
N.K., Ottinger, M.A., 2015a. Decadal re-evaluation of contaminant exposure and 
productivity of ospreys (Pandion haliaetus) nesting in Chesapeake Bay regions of 
concern. Environ. Pollut. 205: 278-290. 
 
Lazarus, R.S., Rattner, B.A., Brooks, B.W., Du, B., McGowan, P.C., Blazer, V.S., 
Ottinger, M.A., 2015b. Exposure and food web transfer of pharmaceuticals in ospreys 
(Pandion haliaetus): predictive model and empirical data. Integ. Environ. Assess. Manag. 
11, 118-129.  
 
 
 
154 
Legendre, P., Anderson, M.J., 1999. Distance based redundancy analysis: testing 
multispecies responses in multifactorial ecological experiments. Ecol. Monograph 69, 1-
24.  
 
Lillicrap, A., Langford, K., Tollefsen, K.E., 2011. Bioconcentration of the intense 
sweetener sucralose in a multitrophic battery of aquatic organisms. Environ. Toxicol. 
Chem. 30, 673–681. 
Levenson, H., Koplin, J.R., 1984. Effects of human activity on productivity of nesting 
ospreys. J. Wild. Manage 48, 1374-1377.  
Lullen, D.R., Vadas, G.G., Unger, M.A., 2006. Kepone in James River fish: 1976-2002. 
Sci. Tot. Environ. 358, 286-297. 
 
Luloff, T.W., Wishart, A.E., Addison, S.M.F., MacDougall-Shackleton, S.A., Hill, K.A., 
2011. Radiation exposure differentially affects songbird 8-hydroxy-2’-deoxyguanosine 
plasma profiles: ionizing radiation damage response in songbirds. Environ. Mol. Mutagen 
52, 658-663.  
 
Machmer, M.M., Ydenberg, R.C., 1989. Weather and osprey foraging energetics. Can. J. 
Zool. 68, 40-43.  
Marnett, L.J., 1999. Lipid peroxidation-DNA damage by malondialdeyde. Mut. Res. 8, 
83-95.  
 
Marteinson, S.C., Kimmins, S., Letcher, R.J., Palace, V.P., Bird, D.M., Ritchie, I.J., 
Fernie, K.J., 2011. Diet exposure to technical hexabromocyclododecane (HBCD) affects 
testes and circulating testosterone and thyroxine levels in American Kestrels (Falco 
sparverius). Environ. Res. 111, 1116-1123.  
Marteinson, S.C., Bird, D.M., Letcher, R.J., Sullivan, K.M., Ritchie, I.J., Fernie, K.J., 
2012. Dietary exposure to technical hexabromocyclododecane (HBCD) alters court- ship, 
incubation and parental behaviors in American Kestrels (Falco sparverius). Chemosphere 
89, 1077-1083.  
Mauldin, R.P., Hard, R.J., Munoz, C.M., Rice, J. L.Z., Verostick, K., Potter, D.R., Dollar, 
N., 2012. Carbon and nitrogen stable isotope analysis of hunter-gathers from the 
Coleman site, a Late Prehistoric cemetery in Central Texas. J. Archaeol. Sci. 40, 1369-
1381. 
 
Mayfield, H., 1961. Nesting success calculated from exposure. Wilson Bull. 73, 255-261.  
 
McGee, B.L., Fisher, D.J., Yonkos, L.T., Ziegler, G.P., Turley, S., 1999. Assessment of 
sediment contamination, acute toxicity, and population viability of the estuarine 
amphipod Leptocheirus plumulosus in Baltimore Harbor, Maryland, USA. Enivron. 
Toxicol. Chem. 18, 2151-2160.  
 
 
 
 
155 
McKernan, M.A., Rattner, B.A., Hale, R.C., Ottinger, M.A., 2009. Toxicity of poly- 
brominated diphenyl ethers (DE-71) in chicken (Gallus gallus), mallard (Anas 
platyrhynchos), and American kestrel (Falco sparverius) embryos and hatchlings. 
Environ. Toxicol. Chem. 28, 1007-1017.  
 
McKernan, M.A., Rattner, B.A., Hatfield, J.S., Hale, R.C., Ottinger,  M.A., 2010. 
Absorption and biotransformation of polybrominated diphenyl ethers DE-71 and DE-70 
in chicken (Gallus gallus), mallard (Anas platyrhynchos), American kestrel (Falco 
sparverius) and black-crowned night-heron (Nycticorax nycticorax) eggs. Chemosphere. 
79, 100-109.  
 
Meylan, W.M., Howard, P.H., Boethling, R.S., Aronson, D., Printup, H., Gouchie, S., 
1999. Improved method for estimating bioconcentration/bioaccumulation factor from 
octanol/water partition coefficient. Environ. Toxicol. Chem. 18, 664-672.  
(MDE) Maryland Department of the Environment, 2009. Guidelines for Recreationally 
Caught Fish Species in Maryland. Available. http://www.mde.state.md.us/assets/do 
cument/fish/ advisory_summary.pdf (accessed 03.12.15.).  
 
(MDE) Maryland Department of the Environment, 2014. Guidelines for Recreationally 
Caught Fish Species in Maryland. Available. http://www.mde.state.md. 
us/assets/document/fish/advisory_summary.pdf (accessed 03.02.15.).  
Misra, R.R., Smith, G.T., Waalkes, M.P., 1998. Toxicology 126, 103-114.  
Mitchelmore, C.L., Chipman, J.K., 1998. DNA strand breakage in aquatic organisms and 
the potential value of the comet assay in environmental monitoring. Mutat. Res. Fund. 
Mol. Mech. Mut. 399, 135-147. 
 
Morose, G., 2006. An Overview of Alternatives to Tetrabromobisphenol A (TBBPA) and 
Hexabromocyclododecane (HBCD). A Publication of the Lowell Center for Sustainable 
Production University of Massachusetts, Lowell, MA. Available: 
http://www.chemicalspolicy.org/downlo ads/AternativestoTBBPAandHBCD.pdf 
(accessed 02.19.2014.). 
 
Nagy, K.A,. 1987. Field metabolic rate and food requirement scaling in mammals and 
birds. Ecol. Monogr . 57, 111–128. 
 
Nikolaou, A., Meric, S., Fatta, D., 2007. Occurrence patterns of pharmaceuticals in water 
and wastewater environments. Anal. Bioanal. Chem. 387, 1225-1234. 
 
Norstrom R.J., Simon M., Moisey J., Wakeford B., Chip Weseloh, D.V., 2002. 
Geographical distribution (2000) and temporal trends (1981-2000) of brominated 
diphenyl ethers in Great Lakes Herring Gull eggs. Environ. Sci. Technol. 36: 4783-4789. 
 
 
 
 
156 
Oakes, K.D., Van Der Kraak, G.J., 2003. Utility of the TBARS assay in detecting 
oxidative stress in white sucker (Catostomus commersoni) populations exposed to pulp 
mill effluent. Aquat. Toxicol. 63, 447-463. 
 
Oaks, J.L., Watson, R.T., 2011. South Asian vultures in crisis: Environmental 
contamination with a pharmaceutical. In: Elliott JE, Bishop CA, Morrissey CA (Eds.). 
Wildlife Ecotoxicology Forensic Approaches. Springer. New York, NY. pp. 413–441. 
 
Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’Hara, R.B., 
Simpson, G.L., Solymos, P., Stevens, M.H.H., Wagner, H., 2015. Vegan: Community 
Ecology Package. R package version 2.3-0. http://CRAN.R-project.org/package=vegan 
 
Park, J.S., Holdern, A., Chu, C., Kim, M., Rhee, A., Patel, P., Shi, Y., Linthicum, S., 
Walton, B.J., McKeown, K., Jewell, N.P, Hooper, K., 2009. Time-trends and congener 
profiles of PBDEs and PCBs in California Peregrine Falcons (Falco peregrinus). 
Environ. Sci. Technol. 43, 8744-51. 
 
Pastor, N., Baos, R., Lopez-Lazaro, M., Jovani, R., Tella, J.L., Hajji, N., Hiraldo, F., 
Cortes, F., 2004. A 4 year follow-up analysis of genotoxic damage in birds of the Doñana 
area (south west Spain) in the wake of the 1998 mining waste spill. Mutagenesis 19, 61-
65.  
 
Pellacani, C., Buschini, A., Galati, S., Mussi, F., Franzoni, S., Costa, L.G., 2012. 
Evaluation of DNA damage induced by 2 polybrominated diphenyl ether flame retardants 
(BDE-47 and BDE-209) in SK-N-MC cells. Internal. J. Toxicol. 31, 372-379. 
 
Peng, H., Saunders, D.M., Sun, J., Codling, G., Wiseman, S., Jones, P.D., Giesy, J.P., 
2015. Detection, identification, and quantification of hydroxylated bis(2-ethylhexyl)-
tetrabromophthalate isomers in house dust. Environ. Sci. Technol. doi: 
10.1021/es505743d 
 
Phillips S.W. ed., 2007. Synthesis of U.S. Geological Survey science for the Chesapeake 
Bay ecosystem and implications for environmental management: U.S. Geological Survey 
Circular 1316, 63p. 
 
Pinkney, A.E., Harshbarger, J.C., May, E.B., Melancon, M.J., 2001. Tumor prevalence 
and biomarkers of exposure in brown bullhead (Ameiurus nebulosus) from the tidal 
Potomac River, USA, watershed. Environ. Toxicol. Chem. 20, 1196-1205. 
 
Pinkney, A.E., Harshbarger, J.C., May, E.B., Reicher, W.L., 2004a. Tumor prevalence 
and biomarkers of exposure and response in brown bullhead (Ameiurus nebulosus) from 
the Anacostia River, Washington, DC and Tuckahoe river, Maryland, USA. Environ. 
Toxicol. Chem. 23, 638-647. 
 
Pinkney, A.E., Harshbarger, J.C., May, E.B. Melancon, M.J., 2004b.Tumor prevalence 
and biomarkers of exposure in brown bullhead (Ameiurus nebulosus) from Back River, 
 
 
 
157 
Furnace Creek, and Tuckahoe River, Maryland. Arch. Environ. Toxicol. Chem. 46, 492-
501. 
 
Poole, A.F., 1985. Courtship feeding and osprey reproduction. The Auk, 102, 479-492.  
 
Poole, A.F., 1989. Ospreys: a natural and unnatural history. Cambridge University Press: 
New York, NY, USA. 
 
Postupalsky, S., 1977. A critical review of problems in calculating osprey reproductive 
success. In: Ogden, J.C. (Ed.), Transactions North American Osprey Research 
Conference, U.S. Department of the Interior National Park Service Transactions and 
Proceedings Series No. 2. College of William and Mary, Williamsburg, VA, pp. 1-11, 
1972.
 
Potter, K.E., Watts, B.D., La Guardia, M.J., Harvey, E.P., Hale, R.C., 2009. 
Polybrominated diphenylether flame-retardants in Chesapeake Bay region, USA 
peregrine falcon (Falco peregrinus) eggs: urban/rural trends. Environ. Toxicol. Chem. 
28, 5973-5981. 
 
Quinn, G.P., Keough, M.J., 2002. Experimental Design and Data Analysis for 
Biologists.Cambridge University Press. New York, NY, USA. 
 
Ramirez, A.J., Brain, R.A., Usenko, S., Mottaleb, M.A., O’Donnell, J.G., Stahl, L.L., 
Wathen, J.B., Snyder, B.D., Pitt, J.L., Perez Hurtado, P., Dobbins, L.L., Brooks, B.W., 
Chambliss, C.K., 2009. Occurrence of pharmaceuticals and personal care products in 
fish: results of a national pilot study in the United States. Environ. Toxicol. Chem. 28, 
2587-2597. 
 
Rattner, B.A., Melancon, M.J., Custer, T.W., Hothem, R.L., King, K.A., Le Captain, L.J., 
Span, J.W., Woodin, B.R., Stegman, J.J., 1993. Biomonitoring environmental 
contamination with pipping black-crowned night heron embryos: induction of 
cytochrome P450. Environ. Toxicol. Chem. 12, 1719-1732.  
 
Rattner, B.A., McGowan, P.C., Golden, N.H., Hatfield, J.S., Toschik, P.C., Lukei, R.F., 
Hale R.C., Schmitz-Afonso I., Rice, C.P., 2004. Contaminant exposure and reproductive 
success of ospreys (Pandion haliaetus) nesting in Chesapeake Bay regions of concern. 
Arch. Environ. Contam. Toxicol. 47, 126-140. 
 
Rattner, B.A., McGowan, P.C., 2007. Potential hazards of environmental contaminants to 
avifauna residing in the Chesapeake Bay estuary. Waterbirds 30 (Sp 1), 63-81.  
Rattner, B.A., Golden, N.H., Toschik, P.C., McGowan, P.C., Custer, 2008., 
Concentrations of metals in blood and feathers of nestling Ospreys (Pandion haliaetus) in 
Chesapeake and Delaware Bays. Arch. Environ. Contam. Toxicol. 54, 114-122. 
 
Rattner, B.A., Lazarus, R.S., Heinz, G.H., Karouna-Renier, N.K., Schultz, S.L., Hale, 
 
 
 
158 
R.C., 2013. Comparative embryotoxicity of a pentabrominated diphenyl ether mixture to 
common terns (Sterna hirundo) and American kestrels (Falco sparverius). Chemosphere 
92, 441-447.  
Rendal, C., Kusk, K.O., Trapp, S., 2011. Optimal choice of pH for toxicity and 
bioaccumulation studies of ionizing organic chemicals. Environ. Toxicol. Chem. 30, 
2395-2406.  
Rice, C.P., O'Keefe, P.W., Kubiak, T.J., 2003. Sources, pathways and effects of PCBs 
dioxins and dibenzofurans. In: Hoffman, D.J., Rattner, B.A., Bruton Jr., G.A., Cairns Jr., 
J. (Eds.), Handbook of Ecotoxicology. Lewis Publishers, Boca Raton, FL, pp. 501-573.  
Roosens, L., Abdallah, M.A., Harrad, S., Neels, H., Covaci, A., 2009. Exposure to 
Hexabromocyclododecanes (HBCDs) via dust ingestion, but not diet, correlates with 
concentration in human serum—preliminary results. Environ Heal Persp. 117, 1707-02.   
 
Roper, T.A., Sykes, R., Gray, C., 1993. Fatal diltiazem overdose: report of four cases and 
review of the literature. Clin.Toxicol. 69, 474-476. 
 
Rose, W.L., French, B.L., Reichert, W.L., Faisal, M., 2000. DNA adducts in 
hematopoietic tissues and blood of the mummichog (Fundulus heteroclitus) from a 
creosote-contaminated site in the Elizabeth River, Virginia. Mar. Environ. Res. 50, 581-
589. 
 
RxList. 2008., Internet drug index from WebMD. Available. http://www.rxlist.com/ 
(accessed 02.19.2014.).  
 
Schaadt, C.P., Bird, D.M., 1993. Sex-specific growth in ospreys: the role of sexual size 
dimorphism. Auk 110, 900-910.  
Schultz, M.M., Furlong, E.T., Kolpin, D.W., Werner, S.L., Schoenfuss, H.L., Barber, 
L.B., Blazer, V.S., Norris, D.O., Vajda, A.M.,  2010. Antidepressant pharmaceuticals in 2 
U.S. effluent-impacted streams: Occurrence and fate in water, sediment, and selective 
uptake in fish neural tissue. Environ. Sci. Technol. 44, 1918–1925. 
 
Schwab, B.W., Hayes, E.P., Fiori, J.M., Mastrocco, F.J., Roden, N.M., Cragin, D., 
Meyerhoff, R.D., D'Aco, V.J., Anderson, P.D., 2005. Human pharmaceuticals in US 
surface waters: A human health risk assessment. Regul. Toxicol. Pharmacol. 42, 296–
312. 
 
She J., Holden, A., Adelsbach, T.L., Tanner, M., Schwarzbach, S.E., 2008. 
Concentrations and time trends of polybrominated diphenyl ethers (PBDEs) and 
polychlorinated biphenyls (PCBs) in aquatic bird eggs from San Francisco Bay, CA 
2000-2003. Chemosphere. 73, S201-S209. 
 
 
 
 
159 
Shore, R.F., Taggart, M.A., Smits, J., Mateo, R., Richards, N.L., Fryday, S., 2014. 
Detection and drivers of exposure and effects of pharmaceuticals in higher vertebrates. 
Phil. Trans. R. Soc. B. 369, 2013.0570. 
 
Singh N.P., McCoy M.T., Tice R.R., Schneider, E.L., 1988. A simple technique for 
quantitation of low levels of DNA damage in individual cells. Exp. Cell Res. 175, 184-
191.  
 
Soh, L., Connors, K.A., Brooks, B.W., Zimmerman, J.B., 2011. Fate of sucralose through 
environmental and water treatment processes and impact on plant indicator species. 
Environ. Sci. Technol. 45, 1363–1369. 
Spitzer, P.R., 1980. Dynamics of a Discrete Coastal Breeding Population of Ospreys in 
the Northeastern United States during the Period of Decline and Recovery, 1969-1978 
(Ph.D. thesis). Cornell Univ., Ithaca, NY.  
Spitzer, P.R., Poole, A.F., Scheibel, M., 1983. Initial recovery of breeding ospreys in the 
region between New York City and Boston. In: Bird, D.M. (Ed.), Biology and 
Management of Bald Eagles and Ospreys. Harpell Press, Ste. Anne de Bellevue, Quebec, 
pp. 231-241.  
Stapleton, H.M., Klosterhaus, S., Eagle, S., Fuh, J., Meeker, J.D., Blum, A., Webster, 
T.F., 2009. Detection of organophosphate flame retardants in furniture foam and U.S. 
house dust. Environ. Sci. Tch. 43, 7490-7495.  
Steenhof, K., Newton, I., 2007. Assessing nesting success and productivity. In: Bird, 
D.M., Bildstein, K.L. (Eds.), Raptor Research and Management Techniques. Hancock 
House, Surrey, British Columbia, Canada, pp. 181-192.  
Stieger, G., Scheringer, M., Ng, C.A., Hungerbühler, K., 2014. Assessing the persis- 
tence, bioaccumulation potential and toxicity of brominated flame retardants: data 
availability and quality for 36 alternative brominated flame retardants. Chemosphere 116, 
118-123.  
Stinson, C.H., Lauthner, J., Ray, R.T., 1987. The effect of weather conditions on 
behavior of ospreys in northwestern Washington. Can. J. Zool. 65, 2116-2118.  
Su, H., Wu, F., Zhang, R., Zhao, X., Mu, Y., Feng, C., Giesy, J.P., 2014. Toxicity 
reference values for protecting aquatic birds in China from the effects of polychlorinated 
biphenyls. Rev. Environ. Contam. Toxicol. 230, 59-82.  
Susquehanna River Basin Commission., 2006. Susquehanna River Basin Fact Sheet. 
http://www.srbc.net/pubinfo/docs/Susq%20River%20Basin%20General%20(11_06).pdf 
 
Thomann, R.V., Connolly, J.P., Parkerton T.F., 1992. An equilibrium model of organic 
chemical accumulation in aquatic food webs with sediment interaction. Environ. Toxicol. 
Chem. 11, 615-629.  
 
 
 
160 
 
Tillitt ,D., Giesy, J.P., 2012.  Ecotoxicology of organochlorine chemicals in birds of the 
Great Lakes. Environ. Toxicol. Chem. 32, 490-492.   
 
Toschik, P.C., Rattner, B.A., McGowan, P.C., Christman, M.C., Carter, D.B., Hale, R.C., 
Matson, C.W., Ottinger, M.A., 2005. Effects of contaminant exposure on repro- ductive 
success of ospreys (Pandion haliaetus) nesting in Delaware River and Bay, USA. 
Environ. Toxicol. Chem. 24, 617-628.  
Turle, R., Norstrom, R.J., Collins, B.,1991. Comparison of PCB quantification methods: 
re-analysis of archived specimens of herring gull eggs from the Great Lakes. 
Chemosphere 22, 201-213.  
(US EPA) U.S. Environmental Protection Agency.,1993. Wildlife exposure factor 
handbook. Washington, DC: USEPA. EPA/600/R-98-187a. 
1994. Chesapeake Bay Basinwide Toxics Reduction Strategy Reevaluation 
Report. Annapolis, MD. U.S. EPA CBP/TRS 117/94. p. 192.  
 
2009. Pharmaceuticals and Personal Care Products (PPCPs). Available. 
http://www.epa.gov/ppcp/ (accessed 02.19.2015).  
 
2010. An exposure assessment of polybrominated diphenyl ethers. p. 110.
 EPA/600/R-09/086F. Washington, D.C. 
 
2012a. Toxic Contaminants in the Chesapeake Bay and its Watershed: Extent and 
Severity of Occurrence and Potential Biological Effects. U.S. EPA Chesapeake 
Bay Program Office, Annapolis, MD. p. 175.  
 
2012b. Total Maximum Daily Loads (303d). Overview of Impaired Waters and 
Total Maximum Daily Loads Program.  Available. http://water.epa.gov/lawsregs 
/lawsguid ance/cwa/tm dl/overview.cfm (accessed 02.19.2014.).  
 
2013. Polychlorinated Biphenyls (PCBs) Available. http://www.epa.gov
 /wastes/hazard/tsd/pcbs/about.htm (accessed 09.24.15).  
2014a. Polybrominated Diphenylethers (PBDEs) Significant New Use Rules.
 Available. http://www.epa. gov/oppt/existingchemicals/pubs/qanda.html
 (accessed 03.02.15.).  
2014b. DecaBDE Phase-out Initiative. Available. http://www.epa.gov/oppt/exis 
tingchemi cals/pubs/actionplans/deccadbe.html (accessed 03.02.15.).  
2015.  Polybrominated diphenylethers (PBDEs) Significant New Use Rules
 (SNUR). Available. http://www.epa.gov/oppt/existingchemicals/pubs/qanda.html
 (accessed 09.02.15.). 
 
 
 
161 
Valavandis, A., Vlachogianna, T., Fiotakis, C., 2009. 8-hydroxy-2’-deoxyguanosine (8-
OHdG): A critical biomarker of oxidative stress and carcinogenesis. J. Environ. Sci. 
Health C 27, 120-139. 
 
Valenti, T.W., Perez-Hurtado, P., Chambliss, C.K., Brooks, B.W., 2009. Aquatic toxicity 
of sertraline to Pimephales promelas at environmentally relevant surface water pH. 
Environ. Toxicol. Chem. 28, 2685–2694. 
Valenti, T.W., Taylor, J.M., Back, J.A., King, R.S., Brooks, B.W., 2011. Influence of 
drought and total phosphorous on diel pH in wadeable streams: Implications for 
ecological risk assessment of ionizable contaminants. Integr. Environ. Assess. Manag. 7, 
636–647. 
Valenti, T.W., Gould, G.G., Berninger, J.P., Connors, K.A., Keele, N.P., Prosser, K.N., 
Brooks, B.W.,  2012. Human therapeutic plasma levels of the selective serotonin 
reuptake inhibitor (SSRI) sertraline decrease serotonin reuptake transporter binding and 
shelter seeking behavior in adult male fathead minnows. Environ. Sci. Technol. 46, 
2427–2435. 
van den Berg, M., Birnbaum, L., Bosveld, A.T.C., Brunström, B., Cook, P., Feeley, M., 
Giesy, J.P., Hanberg, A., Hasegawa, R., Kennedy, S.W., Kubiak, T., Larsen, J.C., Rolaf 
van Leeuwen, F.X., Djien Liem, A.K., Nolt, C., Peterson, R.E., Poellinger, L., Safe, S., 
Schrenk, D., Tillitt, D., Tysklind, M., Younes, M., Waern, F., Zacharewski, T., 1998. 
Toxic equivalency factors (TEFs) for PCBs, PCDDs, PCDFs for humans and wildlife. 
Environ. Health Perspect. 106, 775-792.  
 
van den Berg, M., Birnbaum, L.S., Denison, M., De Vito, M., Farland, W., Feeley, M., 
Fiedler, H., Hakansson, H., Hanberg, A., Haws, L., Rose, M., Safe, S., Schrenk, D., 
Tohyama, C., Tritscher, A., Tuomisto, J., Tysklind, M., Walker, N., Peterson, R.E., 2006. 
The 2005 world health organization reevaluation of human and mammalian toxic 
equivalency factors for dioxins and dioxin-like compounds. Toxicol. Sci. 93, 223-241.  
 
Vanderford, B.J., Snyder, S.A., 2006. Analysis of pharmaceuticals in water by isotope 
dilution liquid chromatography/tandem mass spectrometry. Environ. Sci. Technol. 40, 
7312–7320. 
Velinsky, D.J., Riedel, G.F., Ashley, J.T.F., Cornwell, J.C., 2011. Historical 
contamination of the Anacostia river, Washington, D.C. Environ. Monit. Assess. 183, 
307-328.  
 
Verreault, J., Gebbink, W.A., Gauthier, L.T., Gabrielsen, G.W., Letcher, R.J., 2007. 
Brominated flame retardants in glaucous gulls from the Norwegian Artic: more than just 
an issue of polybrominated diphenyl ethers. Environ. Sci. Technol. 41, 4925-4931. 
  
(VDH) Virginia Department of Health, 2013. Consumption Advisories and Re- strictions 
in Effect of Virginia Waterways. Available. http://www.vdh.state.va. 
us/epidemiology/DEE/PublichealthToxicology/Advisories (accessed 03.02.15.).  
 
 
 
162 
 
Viverette, C. B., G. C. Garman, S. P. McIninch, A. C. Markham, B. D. Watts, Macko, S. 
A., 2007. Finfish-waterbird trophic interactions in tidal freshwater tributaries of the 
Chesapeake Bay. Waterbirds. 30 (Special Publication), 39-49. 
 
Watts, B.D., Byrd, M.A., Watts, M.U., 2004. Status and distribution of breeding ospreys 
in Chesapeake Bay: 1995-1996. J. Raptor Res. 38, 47-54.
 
Watts, B.D., Paxton, B.J., 2007. Ospreys of the Chesapeake Bay: population recovery, 
ecological requirements, and current threats. Waterbirds 30 (Special Publication 1), 39-
49.  
Weinberg, H., 2010. Chemical contaminants (2009) impairments illustrated using the 
Chesapeake Bay segmentation scheme. Chesapeake Bay Program. Available. 
http://www.chesapeakebay.net/maps. aspx?me nuitem=16834 (accessed 02.09.2014.).  
 
Wiemeyer, S.N., 1971. Reproductive success of Potomac River ospreys, 1970. Chesap. 
Sci. 12, 278-280.
 
Wiemeyer, S.N., Spitzer, P.R., Krantz, W.C., Lamon T.G., Cromartie, E., 1975. Effects 
of environmental pollutants on Connecticut and Maryland ospreys. J. Wild. Manage. 39, 
124-139. 
 
Wiemeyer, S.N., Bunk, C.M., Krynitsky, A.J., 1988. Organochlorine pesticides, poly- 
chlorinated biphenyls, and mercury in osprey eggs -1970-79-and their relationships to 
shell thinning and productivity. Arch. Environ. Contam. Toxicol. 17, 767-787.  
Wishart, D.S., Knox, C., Guo, A.C., Cheng, D., Shrivastava, S., Tzur, D., Gautam, B., 
Hassanali, M., 2008. DrugBank: A knowledgebase for drugs, drug actions and drug 
targets. Nucleic Acids Res. 36, 901–906. 
 
Wolf, J., 2009. Point Sources and Priority Watersheds, Chesapeake Bay watershed within 
Virginia. Chesapeake Bay Program. Available. http://www.chesapeakebay.net/maps. 
aspx?menuitem=168 (accessed 02.19.2014).  
 
Woodford, J.E., Karasov, W.H., Meyer, M.W., Chambers, L., 1998. Impact of 2,3,7,8-
TCDD exposure on survival, growth, and behavior of ospreys breeding in Wisconsin, 
USA. Environ. Toxicol. Chem. 17, 1323-1331. 
 
Wu, J.P., Guan, Y.T., Zhang, Y., Luo, X.J., Zhi, H., Chen, S.J., Mai, B.X., 2011. Several 
current use non-PBDE brominated flame retardants are highly bioaccumulative: evidence 
from field determined bioaccumulation factors. Environ. Int. 37, 210-215.  
 
Zhang, R., Guo, J., Wu, F., Mu, Y., Giesy, J.P., Chang, H., Zhao, X., Feng, C., 2014. 
Toxicity reference values for polybrominated diphenyl ethers: risk assessment for 
 
 
 
163 
predatory birds and mammals from two Chinese lakes. Rev. Environ. Contam. Toxicol. 
229, 111-137.