Content uploaded by Charles J Henny
Author content
All content in this area was uploaded by Charles J Henny
Content may be subject to copyright.
BIOMAGNIFICATION FACTORS (FISH TO OSPREY EGGS FROM
WILLAMETTE RIVER, OREGON, U.S.A.) FOR PCDDS, PCDFS, PCBS
AND OC PESTICIDES
CHARLES J. HENNY
1∗
, JAMES L. KAISER
1
, ROBERT A. GROVE
1
,
V. RAYMOND BENTLEY
1
and JOHN E. ELLIOTT
2
1
USGS, For est and Rangeland Ecosystem Science Center, Corvallis, Oregon, U.S.A.;
2
Canadian
Wildlife Service, Pacific Wildlife Research Centre, British Columbia, Canada
(
∗
author for correspondence, e-mail: charles_j_henny@usgs.gov)
(Received 26 October 2001; accepted 1 July 2002)
Abstract. A migratory population of 78 pairs of Osprey (Pandion haliaetus) nesting along the
Willamette River in western Oregon was studied in 1993. The study was designed to determine
contaminant concentrations in eggs, contaminant concentrations in fish species predominant in the
Ospreys diet, and Biomagnification Factors (BMFs) of contaminants from fish species eaten to
Osprey eggs. Ten Osprey eggs and 25 composite samples of fish (3 species) were used to evaluate or-
ganochlorine (OC) pesticides, polychlorinated biphenyls (PCBs), polychlorinated dibenzo-p-dioxins
(PCDDs), and polychlorinated dibenzofurans (PCDFs). Mercury was also analyzed in fish. Geo-
metric mean residues in Osprey eggs were judged low, e.g., DDE 2.3 µgg
−1
wet weight (ww),
PCBs 0.69 µgg
−1
, 2,3,7,8-TCDD 2.3 ng kg
−1
, and generally well below known threshold values
for adverse ef fects on productivity, and the population was increasing. Osprey egg residue data
presented by River Mile (RM) are discussed, e.g., higher PCDDs were generally found immediately
downstream of paper mills and eggs from the Willamette River had significantly elevated PCBs
and PCDDs compared to reference eggs collected nearby in the Cascade Mountains. Prey remains at
nest sites indicated that the Largescale Sucker (Catostomus macrocheilus) and Northern Pikeminnow
(Ptychocheilus or egonensis) accounted for an estimated 90.1% of the biomass in the Osprey diet, and
composite samples of these two species were collected from different sampling sites throughout the
study area for contaminant analyses. With the large percentage of the fish biomass in the Osprey diet
sampled for contaminants (and fish eaten by Ospreys similar in size to those chemically analyzed),
and fish contaminant concentrations weighted by biomass intake, a mean BMF was estimated from
fish to Osprey eggs for the large series of contaminants. BMFs ranged from no biomagnification
(0.42) for 2,3,7,8-TCDF to 174 for OCDD. Our findings for the migratory Osprey were compared
to BMFs for the resident Herring Gull (Larus argentatus), and differences are discussed. We believe
a BMF approach provides some basic understanding of relationships between contaminant burdens
in pre y species of fish-eating birds and contaminants incorporated into their eggs, and may prove
useful in understanding sources of contaminants in migratory species although additional studies are
needed.
Keyw ords: biomagnification factors, DDE, fish, mercury, Oregon, Osprey, Pandion haliaetus, PCBs,
PCDDs, PCDFs
1. Introduction
Major river systems, like the Columbia and its tributary the Willamette, drain
Environmental Monitoring and Assessment 84: 275–315, 2003.
© 2003 Kluwer Academic Publishers. Printed in the Netherlands.
276 C. J. HENNY ET AL.
vast and complex regions. These river systems receive municipal and industrial
wastes as well as runoff from agricultural lands – many which have been sprayed
with pesticides to protect farm crops. The Osprey (Pandion haliaetus), a large
(ca. 1.6 kg) piscivorous bird of prey with a nearly worldwide breeding distribution,
is found nesting throughout much of the Columbia River System. To expand a con-
taminant monitoring program in the region, the Osprey was chosen as a potential
indicator or sentinel species by both the Canadian Wildlife Service and the U.S.
Geological Surve y.
Sev eral Osprey life history traits make it a useful species for contaminant monit-
oring and research (see Elliott et al., 1998) including: (1) eat almost exclusively fish
(99+% of diet), which they capture within a relatively short distance from their nest
sites, (2) ha ve a long life span (up to 25 yr (Poole, 1989)) with high nest fidelity,
(3) construct large stick nests in very exposed sites that are readily detected from
both aerial and ground survey s, (4) readily use man-made structures for nesting
which facilitates access to nests for egg collection, (5) tolerate short-term nest
disturbance, (6) minimal short-term effect on the local population by sample egg
removal from a small subset of nests, and (7) sensiti ve to DDE-induced eggshell
thinning, and widely studied effects of other chlorinated hydrocarbon and mercury
pollutants. Furthermore, fish deliv eries and/or prey remains at nest sites can be
observed/collected to relate contaminants in prey to contaminants in eggs (e.g.,
mean Biomagnification Factors (BMFs) can be estimated). Osprey nests often are
spaced at fairly regular intervals along rivers, not clumped in colonies like herons,
cormorants, and egrets. The solitary and regular nest distribution permits eggs to be
collected at strategic or random sites, e.g., above and below known point sources
for contaminants or in various segments of a river.
Migratory Osprey arriv e in the Willamette Valley between 20 March and 15
April and spend about one month in the nesting area prior to egg laying. Thus,
the burden of some persistent lipophilic pollutants, such as DDE with a half-life in
large birds of about 400 days (Clark et al., 1987), could include contributions from
their wintering grounds. Recent satellite telemetry studies of 22 Ospreys trapped
and fitted with transmitters shows Ospreys nesting on the Columbia River along
the Oregon-Washington border and the Willamette River depart between 28 Au-
gust and 24 September, migrate quickly (mean migration period only 13 days),
and winter in less industrialized southern Mexico and northern Central America
(Martell et al., 1998, 2001). Thus, most industrial contaminants found in Osprey
eggs would be expected to originate from the breeding grounds.
The Osprey population in this study nests along the main stem Willamette
Riv er in northwestern Oregon between Eugene and Portland where it flows into
the Columbia River and includes the lower portions of its principal tributaries,
the McKenzie and Santiam Rivers. Historically, Gabrielson and Jewett (1940:199)
reported Osprey as ‘formerly common along the Columbia and Willamette Rivers...
must now be considered one of the rarer Oregon hawks’. More recently, this pop-
ulation of nesting Ospreys increased from an estimated 13 pairs in 1976 to 78
OSPREY BIOMAGNIFICATION FACTORS 277
pairs in 1993 (Henny and Kaiser, 1996). Therefore, the Osprey population in 1993
was increasing at a rapid rate when 10 eggs and 25 composite samples of fish
(Largescale Sucker (Catostomus macrocheilus), Mountain Whitefish (Prosopium
williamsoni) and Northern Pikeminnow (Ptychocheilus oregonensis)) were collec-
ted for contaminant evaluation. These fish species represent three basic trophic
groups of fish (based upon principal foraging strategy of adults (see Altman et al.,
1997)): omniv ore, insectivore, and piscivore, respectively.
The objectives of this study were: (1) ev aluate concentration patterns of organo-
chlorine (OC) pesticides, congener-specific polychlorinated biphenyls (PCBs), poly-
chlorinated dibenzo-p-dioxins (PCDDs), polychlorinated dibenzofurans (PCDFs)
and total mercury in composite samples of fish strategically collected above and
belo w major municipalities and other pollutant point sources along the W illamette
Riv er, (2) evaluate in a similar manner the same contaminants, except mercury, in
Osprey eggs, (3) ev aluate prey remains at Osprey nests to estimate the percentage
contrib ution (biomass basis) of each fish species to the overall diet, (4) estimate
BMFs for the v arious contaminants from fish (weighted by biomass in Osprey diet)
to Osprey eggs, and (5) ev aluate the present contaminant concentrations in Osprey
eggs to assess possible adverse effects on Osprey productivity.
2. Study Area and Methods
2.1. S
TUDY AREA
The Willamette River, located in northwestern Oregon, is the 13th largest river
in the conterminous United States in terms of total discharge (Kammerer, 1990)
and the largest tributary to the Columbia River below the Snake River (Parkhurst
et al., 1950). From the confluence of its Coast and Middle Forks near Eugene, the
main stem Willamette River flows 187 miles (301 km) northward through the broad
Willamette Valley floor before entering the Columbia Ri ver near Portland. Stream
gradient is relatively gentle, averaging less than 0.47 m km
−1
, including a single
drop of about 14 m at Willamette Falls near Oregon City, River Mile (RM) 26.5
(Altman et al., 1997) (Figure 1). The Willamette Riv er and its 13 major tributaries
drain a rectangular-shaped basin of approximately 30 000 km
2
that is bounded by
the Cascade and Coast mountain slopes and foothills on three sides and by the
Columbia River on the north. Land use in the Willamette basin is 70% forested
(primarily in tributary subbasins), 22% agriculture (primarily cropland on the val-
ley floor), and 5% urbanized (Bonn et al., 1995). The basin includes 11 of the 12
largest cities in Oregon, including the 5 largest (Center for Population Research and
Census 1992), and approximately 2 million people or 70% of Oregon’s population
(Oregon Department of Environmental Quality, 1988; Bonn et al., 1995). Previous
reports have described the basin’s general characteristics (Bonn et al., 1995), en-
vironmental setting (Uhrich and Wentz, 1999) and hydrology (McFarland, 1983;
Gonthier, 1985).
278 C. J. HE NNY ET AL.
F igure 1. Willamette River study area including Osprey nests with an egg collected and fish collection
sites in 1993 (U.S.G.S. map modified from Altman et al., 1997).
OSPREY BIOMAGNIFICATION FACTORS 279
For this study, we included the entire 187 riv er miles of the main stem Wil-
lamette River plus the lower 11.8 river miles (19 km) of the McKenzie Ri ver
(mouth at RM 3.0 due to changes in channel morphology). For simplicity in the
Results and Discussion section, we refer to the Willamette main stem plus the short
portion of the McKenzie River as the Willamette River. In addition, we surveyed
the 11.8 ri ver miles (19 km) of the main stem Santiam River, the lower 16.6 river
miles (27 km) of the North Santiam River, and lower 7.6 river miles (12 km) of the
South Santiam River (Figure 1). The Willamette Riv er was subdivided into four
reaches follo wing Rickert et al. (1975) and Gregory (1993) based on channel char-
acteristics: River Reach IV (Headwaters Reach) includes the river segment from
just above Eugene (RM 187) downstream to Corvallis (RM 131), River Reach III
(Salem Reach) extends from Corvallis downstream to just abov e Newberg (RM
60), River Reach II (Newberg Pool) extends from just abov e Newberg downstream
to Willamette Falls (RM 26.5) and, River Reach I (Tidal Reach) includes the river
segment from Willamette Falls to the Columbia River (RM 0.0) (Figure 1). The
physical descriptions and biological characterizations of each reach are described
in greater detail in Tetra Tech (1992), but briefly the physical structure of the main
stem varies from the upper reaches to the mouth. Reach IV is characterized by fast
moving currents that occur in shallow rocky channels, while Reach III is deeper
and the flow becomes more depositional in nature. Reach II is a large depositional
zone (channel bed contains more silt and sand due to decreased current speeds).
Below Willamette Falls (Reach I), the river enters the industrial Portland Harbor
which has tidal influence and the river has the greatest average depth and channel
width. Most of the data collected during this study pertains to River Reach III
and IV, since few Ospreys were nesting along the lower riv er in 1993 (Henny and
Kaiser, 1993).
The river banks in the study area primarily support Black Cottonwood (Populus
trichocarpa), with a few Douglas Fir (Pseudotsuga menziesii), true fir (Abies spp.)
and Bigleaf Maple (Acer macrophyllum). The cottonwoods, although mature-aged,
have inadequate branch structure at the tops to support Osprey nests. Tall snags and
broken-top live trees are scarce near the river banks. Additional Ospreys nest in the
Willamette Basin outside the study area and in other portions of western Oregon
(see Henny et al., 1978; Witt, 1990).
The U.S. Geological Survey (USGS) began to implement a nationwide National
Water-Quality Assessment (NAWQA) program in 1991 and the Willamette Basin
was among the first 20 NAWQA study units selected for investigation (Wentz and
McKenzie, 1991). Huff and Klingeman (1976) identified the Willamette River as
the largest riv er in the United States with restored water quality. Historically, high
loadings of organic wastes produced critically low dissolved oxygen concentra-
tions, floating and benthic sludge, and Sphaerotilus natans (filamentous bacteria
common in sewage and grossly polluted waters) beds that reduced salmon migra-
tion, recreational use, and aesthetic value. Water quality improv ed dramatically,
salmon runs returned, and recreational uses increased after low-flow augmentation
280 C. J. HE NNY ET AL.
from upstream reservoirs and basinwide secondary sewage treatment began in the
1950s (Hughs and Gammon, 1987). Huff and Klingeman (1976) and Hines et al.
(1977) documented improvements in water quality and Dimick and Merryfields
(1945) and Hughes and Gammon (1987) documented fish assemblages in 1944
and 1983, respectively. In all reaches of the river, Largescale Sucker and North-
ern Squawfish were usually the dominant species collected in 1983. Altman et al.
(1997) provide an excellent summary of information on aquatic biota, their habitats
and toxicological investigations in the Willamette Basin.
On 21 April 1993, we conducted an aerial surv ey from a Cessna 182 flown
about 150 m above the ground to locate occupied Osprey nests along the Wil-
lamette River and lower portions of the McKenzie and Santiam Rivers. The area
of coverage extended 2 km from the river channels. Once it became apparent that
many of the nests were built on electric power poles and transmission to wers, the
local utility companies were contacted to help identify additional Osprey nests. We
also conducted intensive ground searches for nests within the study area. All nests
were visited (not climbed) six to eight times during the nesting season to monitor
nest activity from distant vantage points on the ground using 15–45X spotting
scopes and to count the number of advanced-aged young produced at each nest.
The reproducti ve status of each nest (occupied–adult pair present and/or active–
eggs laid) was determined following definitions of Postupalsky (1977). Nestlings
were conspicuous when fish were delivered to each nest, therefore, the count of
advanced-age nestlings (nearly fledged) was made soon after a fish delivery.
2.2. O
SPREY EGG , W HOLE FISH A ND PR EY R EMAINS COLLEC TIONS
One partially incubated egg was collected from 10 Osprey nests along the Wil-
lamette River in 1993 to determine contaminant concentrations. Eggs were collec-
ted with the assistance of utility company equipment and personnel and nearest
RM was recorded for each egg collection site (Figure 1). An additional egg was
collected from three different nests near Crane Prairie Reservo ir which is located
about 160 km east of the Willamette River in the Cascade Mountains. These nests
were not associated with rivers or industrial sites and were considered reference
eggs, although forests in the area were sprayed with DDT prior to 1966 (see Henny
and Nelson, 1981). Egg contents were placed in chemically cleaned jars and frozen
for later analysis.
Twenty-five whole fish composite samples of three species (Largescale Sucker,
Mountain Whitefish and Northern Pikeminnow) were collected between 7 Septem-
ber and 8 October 1993 using boat electrofishing procedures at 11 strategically
selected sampling sites along different segments of the main stem W illamette River
and lower portions of the McKenzie and Santiam Rivers (Figure 1). Determination
of which fish species to collect was based on identification of fish species pre-
dominately represented in previously collected prey (fish) remains at Osprey nest
sites. Each composite sample usually consisted of 6 fish of the same species with
OSPREY BIOMAGNIFICATION FACTORS 281
weight (g) and fork length (cm) of each fish measured and recorded (Table I).
Fish were individually wrapped in aluminum foil (dull side of foil in contact with
fish), refrigerated during transport, then stored frozen (–20
◦
C) until chemically
analyzed.
Prey (fish) remains were periodically collected from below 24 sample nest sites
and nearby perches (including 8 of 10 egg collection sites) to identify species
and estimate size of fish eaten by Ospreys. Prey remains were placed in separate
ziplock bags and labeled for each nest site and collection date. Findings based
upon different collection dates were later combined for each nest site. A reference
collection of operculum (gill cover) bones was made from whole fish obtained in
the Willamette River to validate identification of species (Swenson 1978). Species
collected included Northern Pikeminnow, Largescale Sucker, Mountain White-
fish, Common Carp (Cyprinus carpio), Largemouth Bass (Micropterus salmoides),
Smallmouth Bass (M. dolomieui), White Crappie (Pomoxis annularis), Black Crap-
pie (Pomoxis nigr o maculatus), and bullheads (Ictalurus spp.) – not identified to
species. Determination of fish species occurrence in the diet of Osprey at each nest
site was made by examination of partial and whole operculums and, in the case
of b u llheads, by pectoral spines. Operculum shapes were sufficiently different to
allo w species identification. Right and left operculum covers were paired by size
and appearance (degree of bleaching/deterioration) and such pairs were treated as
one individual. Each unpaired operculum was treated as one individual. Other prey
remain items used to determine fish species occurrence included scales, fins, skulls,
inter-operculum arches, and pharyngeal bones. A fish collection at Oregon State
University was referenced to facilitate species identification of these prey remain
items.
We used reference fish to regress operculum length on fish weight for the North-
ern Pikeminnow and Largescale Sucker and, thus determine the weight of fish
captured by Osprey based on prey remains at nesting and perching sites. The oper-
culum model was constructed by measuring maximum operculum lengths from the
reference fish of known lengths and weights. Total fish weight was predicted for
operculum length using linear regression analysis (see Newsome, 1977; Van Daele
et al., 1980). For fish species with an insuf ficient number of reference samples to fit
an operculum/weight equation (all species of less importance in the Osprey diet),
fish information (length/weight) was obtained for the Willamette River Basin from
Oregon Department of Fish and Wildlife.
Frequency distributions were determined for each fish species from prey re-
mains found at Osprey nest sites and mean weights were used to adjust numbers in
diet (percent occurrence) to percent biomass. Ospreys are kno wn to be opportun-
istic foragers, therefore, we evaluated overall fish occurrence in the diet on a nest
by nest basis (N = 24 nests) and weighted each nest equally. Then, a mean weight
for each species was used to adjust percentage occurrence in the diet to percentage
biomass.
282 C. J. HE NNY ET AL.
TABLE I
Fish collected in the Willamette River system in 1993
Species of fish
Largescale Sucker Northern Pikeminnow Mountain Whitefish
Collection (RM) No. Water Weight Fork length No. Water Weight Fork length No. Water Weight Fork length
Date Site Location fish (%) (g)
a
(cm)
a
fish (%) (g)
a
(cm)
a
fish (%) (g)
a
(cm)
a
River Mile
(RM)
October 8 A McKenzie River
b
6 74.1 967 (850–1180) 43.5 (41.3–46.4) NA 6 66.5 205 (120–240) 28.5 (25.4–30.5)
September 9 B 166.3–167.5 6 77.1 742 (610–840) 40.6 (38.1–43.2) 7 75.0 259 (160–430) 29.0 (25.4–34.3) 6 69.7 264 (210–375) 29.7 (26.7–34.3)
September 7 C 152.1–152.6 7 73.9 728 (290–940) 40.6 (31.1–44.4) 6 70.9 180 (135–225) 25.9 (23.5–27.9) 6 65.9 277 (155–435) 29.5 (24.1–35.6)
September 7 D 146.5–147.0 6 75.8 777 (600–870) 42.2 (39.4–43.2) 6 74.1 310 (135–485) 30.7 (24.8–35.6) 6 66.0 221 (132–350) 26.7 (23.5–30.5)
September 7 E 136.8–137.4 6 73.7 537 (278–860) 35.8 (29.2–40.0) 6 73.4 261 (100–950) 28.4 (21.0–34.3) NA
September 9 F 122.5–123.8 6 79.8 698 (540–870) 40.1 (36.8–43.8) 7 72.8 246 (150–355) 27.9 (24.1–32.4) NA
October 8 G Santiam River
c
6 74.1 722 (610–830) 39.3 (36.2–41.9) NA 6 70.3 228 (180–300) 28.9 (26.7–32.4)
September 14 H 105.0–105.5 6 76.1 792 (710–930) 42.4 ( 40.6–43.8) 6 72.9 168 (105–255) 25.6 (22.2–28.6) NA
September 14 J 88.0–88.5 6 76.9 723 (580–830) 39.9 (36.2–41.9) 6 76.4 189 (115–250) 26.4 (23.5–29.2) NA
September 14 K 76.5–77.7 6 76.9 795 (630–970) 41.7 (38.1–44.4) 6 75.2 178 (125–225) 25.7 (23.5–27.9) NA
September 16 L 54.5–55.0 5 75.3 672 (460–890) 39.6 (34.9–43.2) 6 74.1 261 (185–350) 28.4 (26.0–30.5) NA
NA = Not available.
a
Mean and range.
b
McKenzie River (RM 14.0–14.8).
c
North Santiam Riv er (RM 12.5–14.0).
OSPREY BIOMAGNIFICATION FACTORS 283
Only three fish species (Largescale Sucker, Northern Pikeminno w and Moun-
tain Whitefish) were collected in the Willamette River for contaminant residue
studies and they represented 90.5% of the actual biomass calculated in the Os-
preys diet (see the Section Results), the other species in the diet were assigned to
one of the three species collected based upon diet similarity (see Li et al., 1987).
Common Carp (omniv o re) was assigned to the Largescale Sucker; Lar g emouth
Bass/Smallmouth Bass (piscivo re) to the Northern Pikeminnow; and bullheads,
Black Crappie, White Crappie, and Bluegill (Lepomis macrochirus) (insectivores)
to the Mountain Whitefish. Therefore, the contaminant intake of the Osprey was
determined by weighting residue concentrations by the adjusted biomass of the
three groups of fish species.
2.3. C
HEMICAL ANALYSES
Osprey egg and whole fish composite samples collected from the Willamette River
were sent to the Great Lakes Institute of Environmental Research (GLIER) at
the Univ ersity of Windsor, Windsor, Ontario, Canada and Zenon Environmental
Laboratories, Burnaby, British Columbia, Canada, for contract chemical analyses.
Organic chemical analyses for Osprey egg and fish samples were conducted using
methods of Lazar et al. (1992), which are described in detail in GLIER (1995). Os-
prey egg samples sent to Zenon Environmental Laboratories were analyzed under
their protocols. Analyses were conducted for 20 OC pesticides, 42 PCB congeners,
4 co-planar PCB congeners, 7 PCDDs, and 10 PCDFs. Quantification was accom-
plished by comparing sample-peak area against standard-peak area of 3 standards
supplied by the Canadian Wildlife Service. OC pesticides and PCB fractions were
analyzed separately on an electron-capture gas chromatograph (EC-GC). The de-
tection limit for OC pesticides and PCBs was 0.1 µgkg
−1
wet weight (ww). OC
pesticides and PCBs were confirmed using gas chromatography/mass spectrometry
(GC/MS). Co-planar PCBs, PCDDs, and PCDFs were analyzed by GC/MS. The
detection limit was 0.1 ng kg
−1
(ww). Methodology for extraction and cleanup
was checked by running sample blanks, replicate samples, and certified reference
samples provided by the Canadian Wildlife Service for OC pesticides and PCBs,
and a (
13
C)-surrogate spike for each sample ran for co-planar PCBs, PCDDs, and
PCDFs (GLIER, 1995). The 2,3,7,8-TCDD-toxic equivalent concentrations (TEQ)
were derived from toxic equivalency factors (TEF) suggested by Van den Berg et
al. (1998) for PCDDs, PCDFs, and PCBs.
Only fish were analyzed for total mercury by atomic absorption spectrophoto-
metry. The dry weight (dw) detection limit was 0.13 µgg
−1
.
2.4. S
TATISTICAL ANALYSES
Residue concentrations were presented as geometric means and log-transformed
for statistical analyses. The lower quantification limit was halved for samples in
which a contaminant was not detected. This v alue was used to calculate geometric
284 C. J. HE NNY ET AL.
means when ≥50% of the samples contained detectable residues. Because of un-
equal sample sizes, the General Linear Models Procedure (SAS Institute, 1996)
was used for analysis of variance. Tukey’s Studentized Range Test (P = 0.05)
was used to separate means. Unless otherwise noted, statistical significance was
P ≤ 0.05. We converted contents of eggs to an approximately fresh ww using egg
v olume (Stickel et al., 1973); all egg residues reported as fresh ww.
3. Results
3.1. C
ONTAMINANTS IN FISH
Of the three species of fish collected by electrofishing during this study, Larges-
cale Suckers were generally the largest (means of composites, 537 to 967 g), with
Northern Pikeminno ws (means 168 to 310 g) and Mountain Whitefish (means 205
to 277 g) similar in size (Table I). Mountain Whitefish were not found in the Wil-
lamette River below RM 146, and Largescale Sucker and Northern Pikeminnow
collections were not made below RM 54 as few Ospreys nested there in 1993
(Henny and Kaiser, 1996).
Sev eral OC pesticide patterns became apparent with the Largescale Sucker and
Northern Pikeminnow which were av ailable for collection at most sampling sites
(Table II): (1) concentrations of OCs generally increased from the headwaters to
the lower river, especially DDE, total chlordanes, dieldrin, and hexachlorobenzene
(HCB), and (2) concentrations were generally higher in the predaceous Northern
Pikeminnow than the omnivorous Largescale Sucker (e.g., geometric means DDE
73 vs. 22 µgkg
−1
(P = 0.008), total chlordanes 11 vs. 3.2 µgkg
−1
(P < 0.0001),
bu t not significantly different for dieldrin 1.2 vs. 0.55 µgkg
−1
(P = 0.24), HCBs
2.2 vs. 3.4 µgkg
−1
(P = 0.29) or the extremely low heptachlor epoxide 0.08 vs.
0.16 µgkg
−1
(P = 0.15)).
Sev e ral patterns became clear when reviewing PCBs in the various fish species
(Table III). PCBs were certainly present in the upper portions of the Willamette
Riv er, and depending upon fish species, PCBs in upper river sites (except Santiam
Riv e r) were sometimes among the highest. PCBs were detected in a relatively nar-
ro w range along much of the Willamette River, but with a consistently increasing
pattern downstream from Salem (RM 84). As with most of the OC pesticides,
the combined non-ortho PCBs (P < 0.001) and the combined remaining PCBs
(P = 0.008) were higher in the Northern Pikeminnow than the Largescale Sucker
(geometric means, 149 vs. 52 ng kg
−1
, and 103 vs. 58 µgkg
−1
, respectively). It
is note worthy that Mountain Whitefish (geo. mean 181 ng kg
−1
), only found in
the Willamette River above RM 146, contained higher concentrations of combined
non-ortho PCBs than upper river (same locations sampled for Mountain Whitefish)
Northern Pikeminnows (119 ng kg
−1
), or Largescale Suckers (48 ng kg
−1
)(P <
0.0001) (Table III), with all species significantly different from each other. The
OSPREY BIOMAGNIFICATION FACTORS 285
TABLE II
Organochlorine contaminants (µgkg
−1
, wet weight) and total mercury (µgg
−1
, dry weight) in composite samples of whole fish collected from the
Willamette River and tributaries in 1993 (N = 25)
Location
River Mile Total
a
Site (RM) QCB OCS HCB DDE Mirex b-HCH g-HCH chlordanes DDD DDT HE Dieldrin Mercury
Largescale Sucker
A McKenzie River
b
ND 0.54 1.25 9.13 ND ND ND 1.33 3.20 0.54 ND ND 0.24
B 166.3–167.5 ND ND 0.70 6.43 ND ND ND 1.40 1.32 0.43 ND ND 0.16
C 152.1–152.6 ND ND 1.19 18.22 ND ND ND 3.33 6.56 0.58 ND 0.41 0.28
D 146.5–147.0 0.32 1.65 14.83 19.09 ND ND ND 4.31 4.29 0.66 0.32 0.55 0.25
E 136.8–137.4 0.14 0.57 2.26 10.22 ND 0.15 0.07 2.47 2.52 0.62 0.48 0.70 0.14
F 122.5–123.8 0.23 3.33 5.59 23.11 0.19 ND ND 3.55 3.33 0.53 ND 1.28 0.84
G Santiam River
c
ND 0.55 1.41 24.18 ND ND ND 2.52 4.48 0.46 0.49 1.15 ND
H 105.0–105.5 0.37 2.73 8.14 26.42 0.20 ND ND 3.34 5.11 0.42 0.22 1.03 0.26
J 88.0–88.5 0.25 0.50 3.86 67.06 ND ND ND 4.60 5.51 0.59 0.42 0.83 0.35
K 76.5–77.7 0.26 0.58 5.67 35.59 ND ND 0.26 7.04 6.72 1.08 0.43 1.79 0.50
L 54.5–55.0 0.34 1.40 5.81 104.36 0.30 ND 0.20 4.50 15.05 0.41 0.38 3.35 0.30
Northern Pikeminnow
B 166.3–167.5 ND 0.25 0.75 52.10 0.49 ND ND 7.02 1.54 ND ND 0.83 1.29
C 152.1–152.6 0.07 0.52 1.70 38.31 0.23 ND 0.15 7.50 2.67 ND ND 1.55 0.93
D 146.5–147.0 0.09 0.43 1.90 49.35 0.31 ND ND 13.91 3.54 ND 0.21 2.02 1.57
E 136.8–137.4 ND 0.59 1.91 63.25 0.43 ND ND 11.36 3.92 ND ND 1.16 1.78
286 C. J. HE NNY ET AL.
TABLE II
(continued)
Location
River Mile Total
a
Site (RM) QCB OCS HCB DDE Mirex b-HCH g-HCH chlordanes DDD DDT HE Dieldrin Mercury
Northern Pikeminnow (continued)
F 122.5–123.8 0.10 0.46 2.27 26.66 0.25 ND 0.07 8.07 1.82 ND ND 1.24 0.84
H 105.0–105.5 0.17 2.02 1.37 318.36 0.32 ND 0.27 18.03 17.91 3.30 0.91 4.55 0.98
J 88.0–88.5 0.12 1.01 4.02 70.13 0.22 ND ND 9.47 4.84 1.05 ND 0.06 0.89
K 76.5–77.7 0.13 0.87 3.73 58.83 0.35 ND 0.22 9.60 4.22 1.15 ND 1.22 0.79
L 54.5–55.0 0.26 2.07 6.52 259.12 0.86 ND 0.22 17.13 14.45 0.83 ND 2.88 1.45
Mountain Whitefish
A McKenzie River
b
ND 0.91 2.40 56.04 0.32 ND ND 10.95 2.20 1.04 ND ND 0.14
B 166.3–167.5 ND 0.40 1.24 30.56 0.25 ND 0.24 7.99 2.31 1.26 ND ND 0.15
C 152.1–152.6 ND 0.36 1.98 27.99 ND ND ND 5.24 ND 0.91 ND ND ND
D 146.5–147.0 ND 2.36 4.20 37.85 ND ND ND 7.11 ND 1.01 ND ND ND
G Santiam River
c
ND 0.21 1.23 77.16 0.16 ND
d
0.99 13.14 2.53 3.31 0.69 4.62 0.15
QCB = pentachlorobenzene, OCS = octachlorostyrene, HCB = hexachlorobenzene, HCH = hexachlorocyclohexane, DDE = p,p
-DDE, DDD = p,p
-
DDD, DDT = p,p
-DDT, HE = heptachlor epoxide, ND = below detection limit.
a
Total chlordanes = sum of trans-nonachlor, cis-nonachlor, oxychlordane, trans-chlordane, cis-chlordane.
b
McKenzie River (RM 14.0–14.8).
c
North Santiam Riv er (RM 12.5–14.0).
d
Contained a-HCH, 0.82 µgkg
−1
.
OSPREY BIOMAGNIFICATION FACTORS 287
combined remaining PCBs showed a similar pattern from the upper portions of
the Willamette Ri ver: Mountain Whitefish (geo. mean 138 µgkg
−1
), Northern
Pikeminnows (82 µgkg
−1
), and Largescale Suckers (51 µgkg
−1
)(P = 0.005),
b ut only Mountain Whitefish and Largescale Suckers were significantly different
from each other.
The most toxic 2,3,7,8-TCDD (tetrachlorodibenzo-p-dioxin) was seldom detec-
ted in the Largescale Sucker except in the lower portion of the Willamette River,
whereas it was first detected in Northern Pikeminnow at RM 146 (waste w ater efflu-
ent site of pulp mill) and then at all sites farther downstream (Table IV). Mountain
Whitefish, contained 2,3,7,8-TCDD at all four sites sampled above Willamette RM
146 including the two highest concentrations reported during the study (Willamette
RM 146 and McKenzie RM 14). Most fish collected during this study contained
detectable concentrations of higher chlorinated PCDDs. The pattern was also relat-
ively consistent among sites, with OCDD > H
7
CDD > H
6
CDD (which corresponds
to octachlorodibenzo-p-dioxin, heptachlorodibenzo-p-dioxin, hexachlorodibenzo-
p-dioxin, respectively). OCDD and OCDF had several unique patterns (Tables IV
and V). OCDD concentrations were higher in the Largescale Sucker (geo. mean
8.5ngkg
−1
) than the Northern Pikeminno w (2.6 ng kg
−1
)(P = 0.001) with peak
concentrations at RM 166, RM 122, and RM 76 which were downstream from
Eugene, Albany, and Salem, respectively. The peak concentration pattern for suck-
ers did not hold for the Northern Pikeminnow. In the upper portion of the river
(abov e RM 146), where all three species were sampled, Largescale Suckers (geo.
mean 5.0 ng kg
−1
) contained higher concentrations of OCDD than both Northern
Pikeminnow (2.2 ng kg
−1
) and Mountain Whitefish (1.2 ng kg
−1
)(P = 0.0005),
with the latter two species not significantly different.
Mercury, which is known to biomagnify through food chains, was found at
higher concentrations throughout the study area in the predaceous Northern Pikemin-
now (piscivore) than the Lar g escale Sucker (omniv ore) (1.12 vs. 0.29 µgg
−1
, P <
0.0001) (Table II). In the upper portion of the river (above RM 146), where all three
species were sampled, mercury was significantly higher in Northern Pikeminnow
(geo. mean 1.2 µgg
−1
) than in Largescale Sucker (0.23) or Mountain Whitefish
(0.10) with all three species significantly different from each other.
3.2. C
ONTAMINANTS IN OSPREY EG GS
Ten eggs (1 per nest) strategically collected above and below major cities and/or
industrial sites along the Willamette River and 3 eggs from Crane Prairie Reservoir
were analyzed for OC pesticides (Table VI), PCBs (Table V II), PCDDs (Table VIII)
and PCDFs (Table IX). Eggs from the three nests at Crane Prairie Reserv oir were
used as reference eggs, i.e., outside the Willamette River system on the eastern
slope of the Cascade Mountains, but nearby at same general latitude. We anticip-
ated that patterns may emerge in relation to RM (from upper to lower river) and
known point sources of specific contaminants.
288 C. J. HE NNY ET AL.
TABLE III
Total PCBs and selected non-ortho, mono-ortho and di-ortho PCBs (wet weight) in composite samples of whole fish collected from the Willamette
River and tributaries in 1993 (N = 25)
Location Total PCBs (µgkg
−1
) Non-ortho PCBs (ng kg
−1
) Selected mono and di-ortho PCBs (µgkg
−1
)PCB
River Mile Aroclor
Site (RM) Congeners 1254:1260 77 81 126 169 99 118 153 105 138 182 183 180 TEQs
a
Largescale Sucker
A McKenzie River
b
70.42 114.88 27.04 2.60 10.13 1.40 3.16 5.35 7.39 0.99 8.50 2.73 1.30 4.88 2.78
B 166.3–167.5 31.47 49.51 26.12 3.57 9.14 3.50 1.43 2.84 3.73 0.44 3.66 1.24 0.66 1.56 2.65
C 152.1–152.6 45.04 72.98 31.00 2.76 10.43 2.87 2.45 4.34 4.80 0.52 5.40 2.00 1.21 1.45 2.97
D 146.5–147.0 68.04 101.85 50.38 3.48 11.71 1.71 3.02 6.15 7.38 0.76 7.54 2.91 1.44 2.57 4.18
E 136.8–137.4 30.56 31.79 34.84 1.50 6.29 0.94 1.92 1.89 1.97 0.21 2.35 0.87 0.37 1.72 2.56
F 122.5–123.8 61.73 78.76 34.07 3.25 11.24 3.75 2.94 4.67 5.53 0.67 5.83 2.39 0.83 3.57 3.27
G Santiam River
c
29.95 31.74 22.58 1.81 4.25 1.31 2.12 1.80 2.15 0.29 2.35 0.92 0.35 1.52 1.78
H 105.0–105.5 66.29 84.53 33.53 2.51 6.21 0.64 3.62 5.59 5.83 0.72 6.25 2.48 0.82 3.69 2.68
J 88.0–88.5 52.78 66.85 32.67 2.70 7.08 1.76 3.75 4.49 4.51 0.45 4.95 1.74 0.75 1.75 2.70
K 76.5–77.7 89.14 132.28 49.73 ND
a
9.53 1.24 4.78 7.59 8.84 0.88 9.79 4.27 1.82 2.99 3.61
L 54.5–55.0 99.38 143.82 80.10 4.90 13.19 1.87 5.15 7.21 10.83 1.02 1 0.64 4.80 1.66 6.17 5.99
Northern Pikeminnow
B 166.3–167.5 114.14 219.27 88.61 2.45 28.71 2.08 3.85 10.23 16.65 1.98 16.23 5.64 2.36 8.84 7.85
C 152.1–152.6 71.75 119.81 91.95 2.86 20.36 1.74 4.17 5.83 9.23 1.23 8.87 3.40 1.15 4.08 7.10
D 146.5–147.0 67.15 110.68 91.33 0.56 20.74 6.43 4.91 5.74 8.43 0.92 8.19 2.69 1.03 3.74 6.85
E 136.8–137.4 130.93 234.77 146.10 3.29 30.51 2.53 5.89 11.48 17.21 1.93 17.37 5.15 2.29 7.04 11.00
OSPREY BIOMAGNIFICATION FACTORS 289
TABLE III
(continued)
Location Total PCBs (µgkg
−1
) Non-ortho PCBs (ng kg
−1
) Selected mono and di-ortho PCBs (µgkg
−1
)PCB
River Mile Aroclor
Site (RM) Congeners 1254:1260 77 81 126 169 99 118 153 105 138 182 183 180 TEQs
a
Northern Pikeminnow (continued)
F 122.5–123.8 55.39 83.73 91.16 3.52 12.04 1.45 3.85 4.59 6.93 0.66 6.20 3.11 0.87 2.74 6.23
H 105.0–105.5 109.47 127.07 112. 00 3.65 16.27 1.87 8.78 7.60 9.34 1.23 9.40 3.51 1.29 15.96 7.79
J 88.0–88.5 91.67 122.69 97.46 2.76 14.34 1.41 4.92 7.35 9.56 1.08 9.08 3.36 1.03 13.82 6.77
K 76.5–77.7 128.79 182.19 159.00 4.26 21.80 1.65 5.72 11.04 13.18 1.47 13.48 4.36 1.65 18.49 10.82
L 54.5–55.0 254.77 355.80 305. 60 7.24 46.10 3.36 13.70 21.31 30.95 2.75 26.33 10.36 3.61 32.69 21.08
Mountain Whitefish
A McKenzie River
b
163.90 276.75 183.30 0.75 27.53 1.67 10.36 15.04 17.65 2.17 20.48 6.00 3.40 9.34 12.36
B 166.3–167.5 155.07 285.43 125.50 1.25 24.68 1.61 9.17 13.90 18.23 2.47 21.12 5.50 3.44 9.74 9.26
C 152.1–152.6 108.04 198.92 143.40 1.62 22.82 1.28 6.87 11.14 12.51 1.76 14.72 3.18 2.29 10.39 9.90
D 146.5–147.0 130.62 179.93 169.45 2.10 19.70 1.35 8.77 9.69 12.36 1.45 13.32 3.70 1.67 10.08 10.90
G Santiam River
c
40.73 61.90 32.84 0.92 5.87 1.00 2.17 3.28 4.36 0.62 4.58 1.72 1.07 2.10 2.42
ND = None detected (detection limit 0.53 ng kg
−1
).
a
Toxicity Equivalency Factor (TEF) from Van den Berg et al. (1998) for birds, multiply by concentration (= TEQ). TEQs for PCBs (77, 81, 126,
169, 118, 105).
b
McKenzie River (RM 14.0–14.8).
c
North Santiam Riv er (RM 12.5–14.0).
290 C. J. HE NNY ET AL.
TABLE IV
Polychlorinated dibenzo-p-dioxins (PCDD) (ng kg
−1
, wet weight) in composite samples of whole fish collected from the Willamette River
and tributaries in 1993 (N = 25)
Location 2378 12378 123478 123678 123789 H
6
CDD 1234678 H
7
CDD PCDD
Site River Mile (RM) TCDD P
5
CDD H
6
CDD H
6
CDD H
6
CDD Total H
7
CDD Total OCDD TEQs
a
Largescale Sucker
A McKenzie River
b
ND ND ND ND ND ND 2.52 4.43 8.03 <0.01
B 166.3–167.5 ND ND ND ND ND 1.28 3.05 5.58 21.95 <0.01
C 152.1–152.6 ND ND ND ND ND ND 2.28 3.88 8.29 <0.01
D 146.5–147.0 ND ND ND ND ND ND 1.42 2.36 5.14 <0.01
E 136.8–137.4 ND ND ND ND ND ND ND ND 2.01 <0.01
F 122.5–123.8 ND ND ND ND 0.59 0.59 3.68 5.48 17.72 0.06
G Santiam Riv er
c
ND ND ND ND ND 1.10 1.91 3.45 6.59 0.02
H 105.0–105.5 ND ND ND ND ND 1.58 2.64 4.46 9.66 <0.01
J 88.0–88.5 ND ND ND ND ND ND 1.93 3.50 7.14 <0.01
K 76.5–77.7 1.05 ND ND 2.06 1.90 3.96 3.57 3.57 12.01 1.27
L 54.5–55.0 0.50 ND ND ND 0.54 1.40 2.50 4.20 8.54 0.56
Northern Pikeminnow
B 166.3–167.5 ND ND ND 0.60 ND 0.60 1.01 1.01 1.00 <0.01
C 152.1–152.6 ND ND ND 0.94 ND 0.94 1.35 1.35 1.72 0.01
D 146.5–147.0 1.06 ND ND 1.23 0.76 1.99 2.37 2.51 6.00 1.15
E 136.8–137.4 2.25 ND ND 1.38 ND 1.38 1.46 1.46 2.03 2.27
OSPREY BIOMAGNIFICATION FACTORS 291
TABLE IV
(continued)
Location 2378 12378 123478 123678 123789 H
6
CDD 1234678 H
7
CDD PCDD
Site River Mile (RM) TCDD P
5
CDD H
6
CDD H
6
CDD H
6
CDD Total H
7
CDD Total OCDD TEQs
a
Northern Pikeminnow (continued)
F 122.5–123.8 0.82 ND ND 0.79 ND 0.79 1.17 1.17 2.08 0.83
H 105.0–105.5 0.71 ND ND 0.80 ND 0.80 ND ND 1.82 0.72
J 88.0–88.5 0.57 ND ND 0.84 ND 0.84 1.67 1.67 8.09 0.58
K 76.5–77.7 0.82 ND ND 0.46 ND 0.46 ND ND 2.04 0.83
L 54.5–55.0 2.00 ND ND 1.91 0.35 2.26 2.85 2.85 4.57 2.06
Mountain Whitefish
A McKenzie River
b
4.39 0.91 ND ND ND ND 0.87 0.87 0.96 5.31
B 166.3–167.5 0.59 ND ND ND ND ND 0.53 0.53 1.10 0.60
C 152.1–152.6 0.96 ND ND 0.69 ND 0.69 ND ND 1.01 1.03
D 146.5–147.0 2.81 ND ND 0.71 0.21 0.92 0.99 0.99 1.98 2.91
G Santiam Riv er
c
ND ND ND ND ND ND ND ND 0.64 <0.01
ND = below detection limit, detection limit determined on a per sample basis general range (0.1 to 0.6 ng kg
−1
).
a
Toxicity Equivalency Factor (TEF) from Van den Berg et al. (1998) for birds, multiply by concentration (= TEQ) for PCDDs only.
b
McKenzie River (RM 14.0–14.8).
c
North Santiam River (RM 12.5–14.0).
292 C. J. HE NNY ET AL.
TABLE V
Polychlorinated dibenzofurans (PCDF) (ng kg
−1
, wet weight) in composite samples of whole fish collected from the Willamette River and tributaries in
1993 (N = 25)
Location
River Mile 2378 23478 123478 234678 123678 123789 Total 1234678 1234789 Total OCDF PCDF
a
Combined
b
Site (RM) TCDF P
5
CDF H
6
CDF H
6
CDF H
6
CDF H
6
CDF H
6
CDF H
7
CDF H
7
CDF H
7
CDF TEQs
Largescale Sucker
A McKenzie River
c
ND ND ND 1.10 ND ND 1.95 ND 0.15 1.32 1.48 0.11 2.89
B 166.3–167.5 ND 0.95 ND 1.10 ND ND 2.32 ND ND 1.39 2.39 1.06 3.72
C 152.1–152.6 ND ND ND 1.03 ND ND 1.03 ND ND 0.57 1.42 0.10 3.07
D 146.5–147.0 0.88 ND ND 1.01 ND ND 1.01 ND ND ND 0.80 0.98 5.16
E 136.8–137.4 1.38 ND ND 0.76 ND ND 0.76 ND ND ND ND 1.46 4.02
F 122.5–123.8 1.05 1.54 0.82 1.01 0.43 ND 2.86 ND 0.39 1.58 8.39 2.82 6.16
G Santiam River
d
ND 2.01 0.56 0.89 ND ND 2.04 ND 0.11 0.71 1.10 2.16 3.96
H 105.0–105.5 1.31 ND 0.69 1.09 ND ND 2.50 ND 0.19 0.95 1.84 1.49 4.17
J 88.0–88.5 ND ND 0.47 1.02 0.29 ND 2.35 ND ND ND 1.11 0.18 2.88
K 76.5–77.7 1.46 ND ND ND ND ND ND 6.73 4.19 10.92 8.86 1.50 6.37
L 54.5–55.0 1.66 1.79 0.60 1.06 0.33 ND 2.59 ND ND 0.60 1.39 3.65 10.20
Northern Pikeminnow
B 166.3–167.5 1.90 ND ND 0.78 ND ND 0.94 ND ND ND ND 1.98 9.83
C 152.1–152.6 2.25 ND ND 0.95 ND ND 1.42 ND ND ND ND 2.35 9.46
D 146.5–147.0 3.04 0.60 0.72 1.48 0.58 0.59 2.25 1.26 0.67 1.93 3.81 4.00 12.00
E 136.8–137.4 6.59 0.43 ND 0.81 ND ND 1.33 ND ND ND ND 7.10 20.36
OSPREY BIOMAGNIFICATION FACTORS 293
TABLE V
(continued)
Location
River Mile 2378 23478 123478 234678 123678 123789 Total 1234678 1234789 Total OCDF PCDF
a
Combined
b
Site (RM) TCDF P
5
CDF H
6
CDF H
6
CDF H
6
CDF H
6
CDF H
6
CDF H
7
CDF H
7
CDF H
7
CDF TEQs
Northern Pikeminnow (continued)
F 122.5–123.8 2.64 ND ND 0.81 ND ND 0.81 ND ND ND ND 2.72 9.78
H 105.0–105.5 2.48 ND ND 0.76 ND ND 0.76 ND ND ND ND 2.56 11.07
J 88.0–88.5 2.12 ND ND 0.87 ND ND 1.41 0.14 ND 0.14 3.28 2.21 9.56
K 76.5 -77.7 3.24 ND ND 0.76 ND ND 0.76 ND ND ND ND 3.32 14.96
L 54.5–55.0 7.15 ND 0.13 0.89 ND ND 1.02 0.21 ND 0.21 ND 7.25 30.39
Mountain Whitefish
A McKenzie River
c
12.09 0.41 ND 0.71 ND ND 0.71 ND ND ND ND 12.57 30.24
B 166.3–167.5 4.21 ND ND 0.76 ND ND 1.08 ND ND ND 0.22 4.29 14.14
C 152.1–152.6 5.05 ND ND 0.71 ND ND 1.25 ND ND ND ND 5.12 16.05
D 146.5–147.0 10.24 ND ND 0.77 ND ND 1.50 ND ND ND 0.34 10.32 24.13
G Santiam River
d
1.97 ND ND 0.91 ND ND 0.91 ND ND ND ND 2.06 4.48
ND = below the detection limit which determined on a per sample basis, general range (0.10 and 0.60 ng kg
−1
).
a
Toxicity Equivalancy Factor (TEF) from Van den Berg et al. (1998) for birds, multiply by concentration (= TEQ) for PCDFs only.
b
TEQs for PCBs (77, 81, 126, 169, 118, 105) (Table III), PCDDs (Table IV) and PCDFs.
c
McKenzie River (RM 14.0–14.8).
d
North Santiam River (RM 12.5–14.0).
294 C. J. HE NNY ET AL.
TABLE VI
Organochlorine contaminants (µgkg
−1
, wet weight) in Osprey eggs collected along the Willamette River in 1993 (N = 10) and a reference area (N
= 3). Principal municipal and/or industrial discharge sites denoted by bold type
Nest Location Total
a
No. River Mile (RM) QCB OCS HCB DDE Mirex a-HCH b-HCH chlordanes DDD DDT HE Dieldrin
43 McKenzie River
b
0.4 0.4 4.0 12054 4.8 ND 4.5 32.3 272.7 40.4 10.1 4.8
Harrisburg
40 RM 159 1.0 ND 2.3 632 12.0 ND ND 12.9 13.5 9.7 2.4 6.0
Halsey
29A RM 147 ND ND 1.9 2331 3.2 ND 1.2 15.1 25.6 12.9 2.8 3.8
29 RM 145 ND ND 3.5 241 1.3 ND ND 10.8 11.0 1.7 4.1 0.7
26 RM 137 ND 0.7 9.1 1781 1.0 ND 4.6 6.9 79.1 52.0 1.8 1.8
Corvallis
22 RM 124 ND ND 1.2 5595 11.3 ND 1.2 24.8 706.4 2811.7 3.4 9.4
Albany
16 RM 111 0.6 ND 11.1 3620 0.9 0.9 44.9 19.9 212.0 ND 2.4 2.4
14 RM 104 ND 0.5 2.2 3660 1.0 ND 11.0 11.3 444.2 78.9 3.7 5.3
11 RM 89 0.5 ND 7.2 2230 6.0 ND ND 32.5 68.7 136.5 5.1 1.7
Salem
3A RM 68 0.5 0.7 5.0 4067 2.1 0.7 10.0 19.9 229.7 198.0 3.4 33.9
CP1 Crane Prairie
c
0.4 ND 1.1 6672 2.0 ND 27.7 16.3 650.5 70.4 14.5 3.8
CP2 Crane Prairie
c
ND ND 0.8 4541 1.4 ND 7.8 10.2 104.7 14.7 11.8 1.4
CP3 Crane Prairie
c
0.9 1.1 1.4 7143 0.8 0.7 148.6 15.0 122.2 36.9 11.4 55.1
QCB = pentachlorobenzene, OCS = octachlorostyrene, HCB = hexachlorobenzene, DDE = p,p
-DDE, HCH = he xachloroc y clohexane, DDD =
p,p
-DDD, DDT = p,p
-DDT, HE = heptachlor epoxide, ND = below detection limit, photomirex and g-HCH were not detected in any eggs.
a
Total
chlordanes = sum of trans-nonachlor , cis-nonachlor , oxychlordane, trans-chlordane, cis-chlordane.
b
McKenzie River (RM 14).
c
Reservoir in
Cascade Mountain Range near Bend, Oregon.
OSPREY BIOMAGNIFICATION FACTORS 295
TABLE VII
Total PCBs and selected non-ortho, mono-ortho and di-ortho PCBs (wet weight) in Osprey eggs collected along the Willamette River in 1993 (N = 10)
and a reference area (N = 3). Principal municipal and/or industrial effluent discharge sites denoted by bold type
Location Total PCBs (µgkg
−1
) Non-ortho PCBs (ng kg
−1
) Selected mono and di-ortho PCBs (µgkg
−1
)
Nest River Mile Aroclor PCB
No. (RM) Congeners 1254:1260 77 126 169 99 118 153 105 138 182 183 180 TEQs
a
43 McKenzie River
b
3675.1 9843.3 308.7 481.2 43.6 171.5 491.8 532.0 96.7 728.4 108.1 55.8 229.7 78.19
Harrisburg
40 RM 159 566.9 1303.8 146.7 192.6 20.2 19.3 51.3 91.7 8.5 96.5 31.3 13.5 57.2 27.98
Halsey
29A RM 147 720.4 1869.2 79.2 144.0 15.3 22.3 64.3 127.8 12.4 138.3 44.6 22.4 103.5 20.26
29 RM 145 347.7 684.8 103.7 122.6 20.7 17.4 31.3 55.5 4.8 50.6 22.1 8.1 27.3 18.26
26 RM 137 361.5 769.3 140.6 178.0 18.7 13.8 31.3 52.9 4.5 57.0 19.2 8.8 32.9 25.61
Corvallis
22 RM 124 1787.6 3332.2 63.7 172.9 17.3 29.5 65.9 326.9 11.3 246.6 156.0 68.6 281.2 22.28
Albany
16 RM 111 357.0 597.9 227.7 86.4 14.1 12.2 19.8 49.6 1.3 44.3 24.3 9.7 41.6 20.37
14 RM 104 374.9 638.7 120.8 84.5 8.4 17.6 31.6 45.6 6.1 47.3 15.8 6.4 24.6 15.42
11 RM 89 966.7 2150.0 109.6 200.9 30.1 41.4 83.4 146.4 24.5 159.1 55.4 24.2 90.3 28.88
Salem
3A RM 68 540.9 837.3 870.5 112.3 8.4 23.3 45.9 52.0 10.0 62.0 17.5 7.8 29.0 56.22
CP1 Crane Prairie
c
193.6 422.5 106.0 106.0 17.1 8.5 15.2 31.2 2.0 31.3 11.4 5.2 18.7 16.27
CP2 Crane Prairie
c
95.3 221.0 32.8 41.9 9.1 4.0 6.7 15.2 0.8 16.4 5.5 2.2 9.5 5.99
CP3 Crane Prairie
c
153.4 271.3 96.4 58.7 8.8 5.8 9.1 19.1 1.0 20.0 7.0 2.4 10.0 10.89
a
Toxicity Equivalency Factor (TEF) from Van den Berg et al. (1998) for birds, multiply by concentration (= TEQ). TEQs for PCBs (77, 126, 169,
118, 105).
b
McKenzie Riv er (RM 14).
c
Reservoir in Cascade Mountain Range near Bend, Oregon.
296 C. J. HE NNY ET AL.
TABLE VIII
Polychlorinated dibenzo-p-dioxins (PCDD) (ng kg
−1
, wet weight) in Osprey eggs collected along the Willamette River in 1993 (N = 10) and
a reference area (N = 3). Principal municipal and/or industrial effluent discharge sites denoted by bold type
Nest Location 2378 12378 123478 123678 123789 H
6
CDD 1234678 H
7
CDD PCDD
No. River Mile (RM) TCDD P
5
CDD H
6
CDD H
6
CDD H
6
CDD total H
7
CDD total OCDD TEQs
a
43 McK enzie River
b
2.6 7.7 6.3 63.6 15.4 90.8 408 427 1362 13.34
Harrisburg
40 RM 159 2.2 6.3 5.9 22.0 11.0 41.3 138 156 1009 10.35
Halsey
29 A RM 147 2.5 10.8 12.6 76.5 36.0 135 639 657 2880 19.22
29 RM 145 2.5 7.7 9.4 39.6 21.7 74.5 547 575 3489 14.13
26 RM 137 2.1 4.6 3.7 28.1 7.2 42.2 141 150 647 8.09
Corvallis
22 RM 124 1.5 6.0 5.8 41.0 12.7 63.7 127 127 255 9.62
Albany
16 RM 111 2.0 5.3 6.2 22.8 12.6 54.2 377 416 2120 9.69
14 RM 104 1.7 6.2 8.5 31.6 25.1 69.7 307 322 2323 11.69
11 RM 89 5.6 12.8 13.7 54.8 37.4 110 347 356 1004 23.82
Salem
3A RM 68 1.8 4.4 3.8 12.2 9.4 26.2 159 168 1217 7.73
CP1 Crane Prairie
c
0.4 1.5 ND 1.5 0.7 2.4 11.4 13.0 64.4 2.00
CP2 Crane Prairie
c
0.4 1.5 0.7 4.5 1.7 7.4 38.3 41.9 219 2.21
CP3 Crane Prairie
c
0.6 1.8 ND 1.9 0.4 2.5 3.4 3.7 10.5 2.46
ND = below detection limit (0.35–0.70 ng kg
−1
).
a
Toxicity Equivalency Factor (TEF) from Van den Berg et al. (1998), multiply by concentration (= TEQ) for PCDDs only.
b
McKenzie River
(RM 14).
c
Reservoir in Cascade Mountain Range near Bend, Oregon.
OSPREY BIOMAGNIFICATION FACTORS 297
OC pesticides were generally low in Osprey eggs collected along the Willamette
Riv er (Table VI). DDE was the highest, but the geometric mean was only 2.3 µg
g
−1
with the highest concentration 12.1 µgg
−1
. Three e ggs collected at the refer-
ence site (geo. mean 6.0 µgg
−1
, range 4.5 to 7.1) had higher DDE concentrations
than those from the Willamette River (2.3 µgg
−1
, range 0.24 to 12.1), although the
difference was not statistically significant (P = 0.19). To further evaluate potential
sources of DDE, the ratios of DDE:DDT were computed for each egg; all eggs
were in excess of 16:1 and generally much higher except one collected at RM
124 between Corvallis and Albany (2:1). The egg at RM 124 with the low ratio
had above average DDE (5.6 µgg
−1
) and high DDT (2.8 µgg
−1
). A pattern of
increasing DDE concentrations downstream was apparent (geo. mean Reach IV
(n = 5) 1.43 µgg
−1
vs. Reach III (n = 5) 3.68 µgg
−1
), especially if two eggs
with high DDE concentrations (RM 189 and RM 124) could be explained (see the
Section Discussion). All other OC pesticides were at extremely low concentrations
(Table VI).
Although many PCB congeners were analyzed (Table VII), patterns in Osprey
eggs in relation to RM were reviewed by evaluating the combined non-ortho PCBs
(PCB 77, 126, 169) in ng kg
−1
and the remaining combined PCBs in µgkg
−1
.
Reference area non-ortho PCBs were significantly lower than from the Willamette
River (geo. mean 147 vs. 359 ng kg
−1
)(P = 0.02) which implies PCB sources
abov e background in the Willamette River. The same pattern held for the remain-
ing combined PCBs (141 vs. 688 µgkg
−1
)(P = 0.007). No overall pattern was
apparent between RM and residue concentrations for PCBs, although the highest
concentration was found in an egg from the farthest upstream site near Springfield
(Table VII).
Osprey eggs from along the Willamette River contained 2,3,7,8-TCDD at low
concentrations and within a narrow range (geometric mean 2.3 ng kg
−1
, range
1.5 to 5.6) (Table VIII), although e ggs from the reference area contained lower
concentrations (0.46 ng kg
−1
)(P = 0.001). As with the fish, concentrations of
OCDD > H
7
CDD > H
6
CDD > H
5
CDD with the following geometric means, 1299,
287, 64 and 6.8 ng kg
−1
, respecti vely. The reference eggs contained 53, 13, 3.5
and 1.6 ng kg
−1
, respectively, and were significantly lower (P = 0.0004, all others
P < 0.0001). No general pattern of PCDD increase was apparent from upper to
lower river. PCDFs did not follow the pattern observed for PCDDs in Osprey eggs,
instead geometric means for OCDF < H
7
CDF > H
6
CDF < H
5
CDF > 2,3,7,8-TCDF
with the following geometric means, 6.2, 26, 18, 32, 0.37 ng kg
−1
, respectively
(Table IX). The concentrations for reference area eggs were, 0.36, 1.7, 3.9, 11,
0.28 ng kg
−1
, respectively, and always lower–OCDF (P = 0.007), H
7
CDF (P =
0.002), H
6
CDF (P = 0.0006), H
5
CDF (P = 0.01), and 2,3,7,8-TCDF (P = 0.54).
298 C. J. HE NNY ET AL.
TABLE IX
Polychlorinated dibenzofurans (PCDF) (ng kg
−1
, wet weight) in Osprey eggs collected along the Willamette River in 1993 (N = 10) and a
reference area (N = 3). Principal municipal and/or industrial effluent discharge sites denoted by bold type
Nest Location 2378 Total 23478 Total Total 1234678 Total PCDF Combined
No. River Mile (RM) TCDF TCDF P
5
CDF P
5
CDF H
6
CDF H
7
CDF H
7
CDF OCDF TEQs
a
TEQs
b
43 McKenzie River
c
ND 5.7 7.4 48.1 28.1 10.9 11.8 1.8 7.51 99.04
Harrisburg
40 RM 159 0.4 4.7 5.0 38.5 18.3 23.8 25.7 4.4 5.64 43.97
Halsey
29A RM 147 0.4 5.9 ND 58.5 29.7 54.0 58.5 13.5 0.94 40.42
29 RM 145 0.4 3.9 ND
d
21.7 14.1 70.7 76.4 26.4 1.11 33.50
26 RM 137 0.6 4.4 ND 26.2 15.0 8.9 9.4 1.7 0.69 34.39
Corvallis
22 RM 124 ND 3.5 ND
d
28.2 19.1 5.0 5.4 1.8 0.05 32.48
Albany
16 RM 111 0.6 5.1 ND 23.6 15.7 56.5 61.2 9.4 1.17 31.23
14 RM 104 0.9 3.5 3.3 27.9 10.2 51.1 55.7 11.1 4.71 31.82
11 RM 89 ND 6.3 ND 56.6 31.0 17.3 19.2 8.9 0.17 52.87
Salem
3A RM 68 0.7 2.1 2.6 19.7 8.9 30.0 31.8 10.3 3.33 67.28
CP1 Crane Prairie
e
0.3 5.8 ND 32.6 8.2 2.0 2.1 0.6 0.32 18.59
CP2 Crane Prairie
e
ND 0.9 1.4 6.7 2.4 5.7 6.2 0.5 1.46 9.66
CP3 Crane Prairie
e
0.5 2.2 ND
d
5.8 3.0 0.4 0.4 ND 0.50 13.05
ND = below detection limit (0.21–0.39 ng kg
−1
) except as noted in superscript d.
a
Toxicity Equivalency Factor (TEF) from Van den Berg et al.
(1998) multiply by concentration (= TEQ). Note: If value is below detection limit, 0 used as concentration in calculation.
b
TEQs for PCBs
(77, 126, 169, 118, 105) (Table VII), PCDDs (Table VIII) and PCDFs.
c
McKenzie Riv er RM 14.
d
Possible diphenyl ether interference; higher
than normal detection limit (2.2–6.7 ng kg
−1
).
e
Reservoir in Cascade Mountain Range near Bend, Oregon.
OSPREY BIOMAGNIFICATION FACTORS 299
3.3. F
ISHINTHEOSPR EY D IET
Fish species eaten by Osprey were determined by identifying prey remains peri-
odically collected from below 24 nests and nearby perches (Table X). Largescale
Sucker were identified in the prey remains at all 24 nest sites sampled and were
usually the dominant prey species, comprising 68.6% of the fish and 82.8% of
estimated biomass in the Osprey diet. The Largescale Sucker in the optimum size
class for Osprey (600–900 g) is ab undant throughout the Willamette River (Tetra-
Tech 1992). Like all suckers, it lives on the bottom in generally shallow water
(Wydoski and Whitney 1979), and because of its slo w movement is especially
vulnerable to aerial strikes by Osprey. Prey remains of Northern Pikeminnow were
identified at three-fourths of the nests and comprised 13.2% of the fish in the
Ospreys diet, but, because of their smaller size, only 7.3% of the total biomass.
Common Carp remains were found at one-fourth of the nest sites sampled and
accounted for 8.2% of the fish and 6.4% of the biomass. The Common Carp in
the smaller size range, confirmed by prey remains at nest sites, were most likely
captured by Ospreys in sloughs or backwaters as carp of this size/age class were
rarely encountered during multiple electrofishing and other fish investigations us-
ing multiple sampling gear in the main stem Willamette River (R. Wildman, pers.
commun.). Sloughs and backwaters were not uniformly distributed near each Os-
prey nest. All other fish species identified in the prey remains at Osprey nest sites
accounted for only 3.5% of the biomass.
The fish captured by Ospreys were similar in size to those collected and ana-
lyzed for contaminants. The mean weight of Largescale Suckers eaten by Ospreys
(based upon prey remains collected) and direct observations was 624 g (Table X),
and the mean weights of whole suckers collected at various river sites for chem-
ical analyses ranged from 537 to 967 g (Table I). The mean weight of North-
ern Pikeminnow eaten by Ospreys was 286 g (Table X), while mean weights of
those collected for chemical analyses ranged from 168 to 310 g (Table I). The
mean weight of Mountain Whitefish eaten by Osprey was 447 g (Table X), while
those collected for chemical analyses were smaller (means ranging from 205 to
277 g) (Table I); however, Mountain Whitefish represented a small percentage of
the Osprey diet.
3.4. B
IOMAGNIFICATION FACTORS (FISH TO OSPR EY EG GS)
To estimate Biomagnification Factors (BMFs) from fish to Osprey eggs, we must
first understand the diet of the fish-eating Osprey. We estimated the contribution
of each fish species to the diet (biomass basis) based upon prey remains collected
belo w 24 Osprey nests and perches (including 8 of 10 sites with eggs collected)
(Table X). Although at least nine fish species were eaten by Ospreys (Table X),
only three fish species were collected for residue analyses, but they represented
three different feeding strategies (Table I). Species of whole fish collected and their
feeding strategies were: (1) Largescale Sucker (omnivore), (2) Northern Pikemin-
300 C. J. HE NNY ET AL.
TABLE X
Percent composition of fish species in Osprey diet (based upon pre y remains) at 24 nests studied
along the Willamette River, 1993
Fish species (Percent composition)
Nest No. Sucker Pikeminnow Carp Bass Bullhead Whitefish Bluegill Other
a
N
2A 72.7 22.7 4.5 22
2B 100.0 29
3A
b
50.0 6.1 37.9 6.1 66
3D 90.9 9.1 11
6 100.0 5
7 60.0 20.0 20.0 5
8 88.9 11.1 9
10 71.0 16.1 9.7 3.2 31
10A 80.0 20.0 5
11
b
62.5 21.9 15.6 32
12 75.0 16.7 8.3 12
13 77.8 11.1 5.6 5.6 18
14
b
61.5 30.8 7.7 13
15 75.0 25.0 4
17 75.0 25.0 4
21 88.9 11.1 9
22
b
60.0 20.0 6.7 6.7 6.7 15
26
b
72.7 9.1 9.1 9.1 11
27 50.0 50.0 12
29
b
44.4 11.1 33.3 11.1 9
29A
b
32.0 4.0 60.0 4.0 25
36 50.0 50.0 4
37 83.3 16.7 6
40
b
25.0 12.5 37.5 12.5 12.5 16
Occurrence 68.6 13.2 8.2 2.6 3.5 0.5 2.1 1.3
(%)
Mean wt (g) 624 286 402 298 146 447 76 101
Biomass (%) 82.8 7.3 6.4 1.5 1.0 0.4 0.3 0.3
Sucker = Largescale Sucker, pikeminnow = Northern Pikeminno w, carp = Common Carp, bass =
Largemouth and/or Smallmouth Bass, bullhead = Brown Bullhead and/or Yellow Bullhead, white-
fish = Mountain Whitefish.
a
Black Crappie, White Crappie and Yellow Perch.
b
Nest with egg collected.
OSPREY BIOMAGNIFICATION FACTORS 301
no w (pisciv o re), and (3) Mountain Whitefish (insectivore). The three fish species
collected represented 82.8, 7.3 and 0.4%, respectively, of the fish biomass eaten
by Ospreys (total 90.5%). The two most important species in the Osprey diet (in
terms of biomass) were represented by similar-sized fish collected for residue ana-
lysis. Furthermore, the size of the fish collected for residue analyses was consistent
among the collection sites (Table I).
Contaminants in the remaining 9.5% of the diet were estimated by assigning fish
species not analyzed for contaminants to a species collected based upon diet simil-
arity (see Li et al., 1987 and the Section Methods). Contaminant intake of Osprey
was estimated by weighting residue concentrations by the adjusted fish biomass
with three types of feeding strategies: (1) omnivore (Largescale Sucker, Common
Carp) 89.2%, (2) pisciv o re (Northern Pikeminno w, Largemouth/Smallmouth Bass)
8.8%, and (3) insectiv ore (Mountain Whitefish, Black Crappie, White Crappie,
bu llhead species, Bluegill) 2.0%.
Contaminant concentrations are notorious for the lack of normal distributions,
therefore, geometric means were used for fish and egg concentrations when calcu-
lating BMFs for Ospreys nesting along the Willamette River (Tables XI and XII).
BMFs for the v arious OC pesticides (based upon wet weight concentrations in
fish and eggs) ranged from extremely lo w (1.2) for HCB to high (87) for DDE.
DDT (47) and DDD (23) were also among the higher BMFs (Table XI). For PCBs,
the tetrachlorobiphenyls had relatively low BMFs (3.8 to 4.4), the pentachlorobi-
phenyls (5.9 to 13), the hexachlorobiphen yls (5.0 to 17), the heptachlorobiphenyls
(5.8 to 19), and the octachlorobipheny ls (15 to 18). BMFs for non-ortho PCBs
also varied: PCB 77 (3.3), PCB 126 (15), and PCB 169 (12) (Table XII). BMFs for
PCDDs were generally much higher than for PCBs, but also variable: OCDD (174),
Total H
7
CDD (110), Total H
6
CDD (125) and 2,3,7,8-TCDD (10). The BMFs for
PCDFs were generally much lower than for PCDDs: OCDF (5.7), Total H
6
CDF
(15) and 2,3,7,8-TCDF showed no biomagnification (0.42).
4. Discussion
4.1. C
ONTAMINANT PATTERNS IN FISH AND OSPR EY EGG S
4.1.1. Fish Characteristics
When evaluating contaminant patterns in fish by RM or among fish species, and
their relationship to contaminant burdens in eggs of Ospreys, two important re-
quirements related to fish must be met: (1) fish species collected must represent
the fish species and size eaten by Ospreys, and (2) fish seasonal movement patterns
up and down the river should be understood and taken into consideration. The
size of fish collected for contaminant analysis, especially the two most important
species (Largescale Sucker and Northern Pikeminnow), which represented 90.1%
of the estimated biomass of fish eaten, was in good agreement with the size of fish
302 C. J. HE NNY ET AL.
TABLE XI
Contaminant concentrations (geometric means, µgkg
−1
, wet weight) and biomagnification factors
(BMFs) for PCB congeners and organochlorines in Osprey eggs and fish (whole body) collected
along the Willamette River in 1993, and compared with Herring Gull eggs and Alewife from Lake
Ontario in 1985 (see Braune and Norstrom, 1989)
BMFs
b
Osprey egg Sucker Pikeminnow Whitefish Osprey egg/ Gull egg/
Category (10)
a
(10)
a
(9)
a
(4)
a
Fish Alewife
% Lipid
c
4.3 5.1 4.6 4.6 4.3/5.0 7.7/2.8
Tetrachlorobiphenyls
PCB 60
d
6.2 1.3 2.2 2.9 4.4 (7.9)
e
8.1
PCB 66+95 6.4 1.6 2.7 3.0 3.8 (6.8) 16
PCB 74 4.5 1.1 1.4 2.3 3.8 (6.8) 30
Pentachlorobiphenyls
PCB 99 25 3.0 5.7 8.7 7.4 (13) 28
PCB 101 35 3.2 5.9 6.9 10 (18) 7.9
PCB 105 9.3 0.61 1.4 1.9 13 (23) 20
PCB 110 16 2.5 4.2 6.7 5.9 (11) 6.9
PCB 118 55 4.7 8.5 12.3 11 (20) 31
Hexachlorobiphenyls
PCB 138 104 6.0 11.6 17.1 16 (29) 42
PCB 141 11 0.67 1.2 2.1 15 (27) 6.7
PCB 146 16 0.88 1.7 1.5 17 (30) 39
PCB 149 18 3.4 4.9 8.0 5.0 (9.0) 15
PCB 153 100 5.5 12.1 14.9 16 (29) 48
Heptachlorobiphenyls
PCB 170+190 28 1.4 2.4 3.4 19 (34) 41
PCB 171 9.1 0.56 1.5 2.3 13 (23) 40
PCB 172 4.2 0.21 0.25 0.45 19 (34) 55
PCB 174 5.2 0.95 1.0 2.2 5.8 (10) 5
PCB 180 63 2.7 8.9 9.9 19 (34) 53
PCB 182+187 36 2.3 4.2 4.4 14 (25) 43
PCB 183 16 0.98 1.5 2.6 15 (27) 43
OSPREY BIOMAGNIFICATION FACTORS 303
TABLE XI
(continued)
BMFs
b
Osprey egg Sucker Pikeminnow Whitefish Osprey egg/ Gull egg/
Category (10)
a
(10)
a
(9)
a
(4)
a
Fish Alewife
Octachlorobiphenyls
PCB 194 9.5 0.61 0.85 1.2 15 (27) 50
PCB 200 3.5 0.17 0.36 0.51 18 (32) 27
PCB 201 14 0.80 1.1 1.5 17 (30) 39
PCB 203 8.0 0.51 0.82 1.1 15 (27) 48
Total PCBs () 688 58 103 138 11 (20) 32
Organochlorines
Total chlordanes 17 3.2 11 7.6 4.3 (7.7) N A
HCB 3.8 3.4 2.2 2.2 1.2 (2.1) 20
DDE 2347 22 73 37 87 (156) 34
Mirex 2.8 NC 0.35 0.12 35 (63) 30
Photomirex ND ND ND ND ND 34
B-HCH 1.3 NC ND ND NC 10
Octachlorostyrene NC 0.58 0.73 0.75 NC 8
Oxychlordane 6.0 0.23 0.92 0.42 21 (38) 60
Trans-nonachlor 3.7 1.6 5.9 6.3 1.8 (3.2) 2.6
Cis-nonachlor 5.7 0.36 2.2 0.22 11 (20) 5.0
DDT 25.3 0.56 0.22 1.1 47 (84) 2.0
HE 3.5 0.16 0.08 ND 25
f
(45) 30
Dieldrin 4.0 0.55 1.2 ND 6.7
f
(12) 7.0
DDD 98.5 4.4 4.3 0.34 23 (41) NA
Cis-chlordane 1.1 0.25 0.55 NC 4.1
f
(7.3) NA
ND = None Detected, NA = Not Available, NC = Not Calculated (>50% samples below detection
limit).
a
Sample size ( ): Number of Osprey eggs (1 egg/nest collected) or number of composite samples
of fish, sucker = Largescale Sucker, pikeminnow = Northern Pikeminnow, whitefish = Mountain
Whitefish. Fish collected in Santiam River excluded from calculations (only Willamette River fish).
b
BMFs for this study determined by weighting fish residue concentrations by percentage of each
fish species biomass in the Osprey diet (see text).
c
For % lipid, arithmetic mean for egg and for fish.
d
IUPAC Numbers for PCB congeners, after Ballschmiter and Zell (1980).
e
The second BMF ( ) = original BMF X 1.79 to normalize to lipid content of Herring Gull eggs (see
the Section Discussion).
f
BMF estimated using one-half the detection limit value when ‘only one’ fish cell with NC or ND.
304 C. J. HE NNY ET AL.
TABLE XII
Contaminant concentrations (geometric means, ng kg
−1
, wet weight) and biomagnification factors
(BMFs) for PCDDs and PCDFs in Osprey eggs and fish (whole body) collected along the Willamette
Riv er in 1993, and compared with Herring Gull eggs and Alewife from Lake Ontario in 1985 (see
Braune and Norstrom, 1989)
Osprey BMFs
b
egg Suck er Pikeminnow Whitefish Osprey egg/ Gull egg/
Category (10)
a
(10)
a
(9)
a
(4)
a
Fish
d
Alewife
% Lipid
c
4.3 5.1 4.6 4.6 4.3:5.0 7.7:2.8
PCB 77
d
156 38 120 154 3.3 (5.9)
e
NA
PCB 81 NA 2.2 2.9 1.3 NC NA
PCB 126 156 9.2 22 24 15 (27) NA
PCB 169 18 1.7 2.2 1.5 12 (21) NA
2378 TCDD 2.3 NC 0.68 1.6 10
f
(18) 21
12378 P
5
CDD 6.8 ND ND NC NC 9.7
123478 H
6
CDD 6.9 ND ND ND NC NA
123678 H
6
CDD 34 NC 0.91 0.32 154
f
(276) 16
123789 H
6
CDD 16 NC NC NC NC NA
Total H
6
CDD 64 0.47 0.98 0.35 125 (224) NA
1234678 H
7
CDD 272 1.9 0.94 0.51 154 (276) NA
Total H
7
CDD 287 2.9 0.95 0.51 110 (197) NA
OCDD 1299 8.5 2.6 1.2 174 (311) NA
2378 TCDF 0.37 0.53 3.1 6.7 0.42 (0.75) NA
Total TCDF 4.3 0.54 3.1 7.2 5.7 (10) NA
Total P
5
CDF 32 NC NC NC NC NA
Total H
6
CDF 18 1.4 1.1 1.1 15 (27) NA
1234678 H
7
CDF 24 NC NC ND NC NA
Total H
7
CDF 26 0.71 NC ND NC NA
OCDF 6.2 1.6 NC 0.21 5.7
f
(10) NA
ND = None Detected, N A = Not Available, NC = Not Calculated (>50% samples below detection
limit).
a
Sample size ( ): Number of Osprey eggs (1 egg/nest collected) or number of composite samples
of fish, sucker = Largescale Sucker, pikeminnow = Northern Pikeminnow, whitefish = Mountain
Whitefish. Fish collected in Santiam River excluded from calculations (only Willamette River fi sh).
b
BMFs for this study determined by weighting fish residue concentrations by percentage of each
fish species biomass in the Osprey diet (see text).
c
For % lipid, arithmetic mean for egg and for fish.
d
IUPAC Numbers for PCB congeners, after Ballschmiter and Zell (1980).
e
The second BMF ( ) = original BMF X 1.79 to normalize to lipid content of Herring Gull eggs (see
the Section Discussion).
f
BMF estimated using one-half the detection limit value when ‘only one’ fi sh cell with NC or ND.
OSPREY BIOMAGNIFICATION FACTORS 305
actually eaten by Ospreys based on prey remains collections and direct foraging
observations. Percent composition of fish species delivered to nests based upon
direct observations also agreed with the collected prey remains. Dauble (1986)
reported that some male Largescale Suckers (on Columbia River) matured at 5–6 yr
of age (32–34 cm fork-length), but most matured at 6–7 yr of age (40 cm), while
females matured at 6–9 yr of age (40–45 cm). Therefore, many Largescale Suckers
eaten by Ospreys and collected for residue analysis (Tables I and X) were sexually
mature. Dauble (1986) reported upstream movement belie ved to be associated with
spawning (primarily in May and June). Tagged Largescale Suckers in the Columbia
Riv er were recovered as far as 14 km upstream and 60 km downstream from
release sites. We assume fish returned to their original locals by the September-
October fish collection period. Ospreys nesting along the Willamette River (part of
the Columbia River system) may hav e been eating Largescale Suckers in May and
June that lived much of their life farther downstream. Ospreys along the Willamette
Riv er arrive in spring during the last 10 days of March; and most lay their eggs dur-
ing the last 10 days of April (presumably before the beginning of major Largescale
Sucker spawning movements). Northern Pikeminnow spawning occurs at a similar
time period (late May–June) as the suckers (Wydoski and Whitney 1979). Mature
Northern Pikeminnows implanted with radio-transmitters in the Columbia River
moved an average of 19.5 km from the release site (Martinelli and Shiveley, 1997).
The upstream movement (apparently to spawn) occurred in May and June and
do wnstream back to their original location in July and August. Reservoir -released
Pikeminnows made most of the long-distance movements to tailraces of dams.
With the mainstem Willamette not dammed, these longer distance movements may
not be required by the Pikeminnows. Thus, based upon data from the Columbia
Riv er, the two principal Osprey prey species were not believed migrating during
late March to early May when fish eaten would potentially influence Osprey egg
b urdens, i.e., fish should reflect local Willamette River contaminant exposure. Little
or no information was available on movements of the other fish species in the W il-
lamette River which comprise a comparatively low proportion of the Osprey diet.
We therefore expected contaminant concentration patterns found in fish collected
at v arious locations to represent the fish in the diet of Osprey at the various nest
locations.
4.1.2. Osprey Characteristics
The 10 Osprey eggs sampled in 1993, although not a large sample size, provide a
means of making a preliminary evaluation of residue patterns in Ospreys. DDE
was the most prominent OC pesticide in both fish and Osprey eggs, although
most OC pesticides were extremely low. DDE generally increased in fish from
the upper to lower river; however, the pattern was less clear for Osprey eggs. We
initially belie ved that three Osprey-related factors could influence a direct relation-
ship between geographical residue patterns in whole fish collected from the river
and residue patterns in Osprey eggs: (1) some Osprey may forage in ponds or lakes
306 C. J. HE NNY ET AL.
nearby in addition to foraging in the Willamette Riv er, (2) some Osprey may oppor-
tunistically capture different fish species because of local availability/vulnerability
(e.g., predaceous species with presumed higher contaminant exposure), and (3) age
of breeding Osprey females, i.e., years nesting along Willamette River. We have in-
adequate Osprey dietary data on a nest-by-nest basis to even attempt a preliminary
contaminant evaluation of the influence of a particular fish species in the diet. Os-
preys are opportunistic foragers, and if a local situation resulted in a species being
ab undant/vulnerable to capture, we believe Ospreys would seize the opportunity.
With respect to age of the female Ospreys, a review of the literature (Anderson
and Hickey, 1976; Ewins et al., 1999) indicates that DDE and PCB concentrations
(in addition to other contaminants) soon reach equilibrium and age is not a factor,
e.g., 25 kno wn-aged Osprey females from Michigan laid eggs with the following
DDE concentrations: 3–4 yr old (geo. means 1.1 µgg
−1
), 5–9 yr (1.1 µgg
−1
), and
10–15 yr (1.2 µgg
−1
).
The ev aluation that follows concerning Osprey foraging location (river vs. ponds
and lakes) must be considered exploratory because of limited data. We recorded
observ ations of foraging sites (i.e., main stem Willamette River or nearby ponds
and lakes) frequented by adult male Osprey at nest sites where an egg was collec-
ted. All Ospreys foraged in the Willamette River, but at three sites, some foraging
also occurred away from the Willamette River. Adjacent lakes and ponds were not
present near all nest sites. Perhaps the best approach to evaluate the potential effect
of feeding a portion of the time (the percentage of time not known with precision)
away from the Willamette River is to revie w industrial contaminants. Contamin-
ants like PCBs, OCDD and TEQs (PCDDs and PCDFs only) were expected to
be at lo wer concentrations in water bodies associated with cropland away from
the Willamette River. The residue concentrations in eggs from each sampled pair
were scored based upon how they compared to the sampled population (nests with
eggs collected). Each egg residue concentration was evaluated and assigned a score
of 0 (value ±20% of overall geometric mean), + 1 (value >20% of mean), or –1
(value <20% of mean). Do eggs from the three nests where males foraged away
from the river consistently have lower residue concentrations? Eggs from the three
nests had PCBs, TEQs (PCDDs and PCDFs), and OCDD scored with nine possible
outcomes. The scores were +, –, –, –, 0, –, –, + and 0 which provided an overall
score of –3 suggesting some reduced industrial contaminant uptake by Ospreys
catching a portion of their fish away from the Willamette River (Table XIII).
OCDD concentrations in Osprey eggs collected near the Willamette Ri ver (geo.
mean 1299 ng kg
−1
) were exceptionally high compared to other locations in the
region (geo. means 13 to 134 ng kg
−1
) (Elliott et al., 1998). The presence of high
concentrations of OCDD in eggs of Ospreys demonstrates that it is able to bioac-
cumulate in some food chains. Furthermore, Largescale Sucker, the dominant prey
species of the Osprey in the W illamette Ri ver, consistently had the highest OCDD
concentrations among the three species of fish sampled. The origin of industrial
sources of high OCDD concentrations in Largescale Suckers from the Willamette
OSPREY BIOMAGNIFICATION FACTORS 307
TABLE XIII
Residues in Osprey eggs along the Willamette River in 1993 with respect to Osprey foraging
locations and location of paper mills
Nest River Mile Osprey eggs Fishing
Number Nest Paper Mill DDE
a
PCBs
a
TEQ
b
OCDD
c
sites
d
43
e
14
e
14.7
e
12.1 + 3.7 + 20.9 + 1362 0 W
40 159 0.6 – 0.6 0 16.0 0 1009 – W
29A 147 147.4 2.3 0 0.7 0 20.2 + 2880 + W
29 145 0.2 – 0.3 – 15.2 0 3489 + W
26 137 1.8 – 0.4 – 8.8 – 647 – W
Reach IV Geo. mean 1.4 0.7 15.5 1550
22 124 5.6 + 1.8 + 9.7 – 255 – W + O
16 111 116.5 3.6 + 0.4 – 10.9 – 2120 + W
14 104 3.7 + 0.4 – 16.4 0 2323 + W + O
11 89 89.0
f
2.2 0 1.0 + 24.0 + 1004 – W
3A 68 4.1 + 0.5 – 11.1 – 1217 0 W + O
Reach III Geo. mean 3.7 0.7 13.6 1090
Scoring System: 0 = ±20% of geo. mean; + = >20% above mean; – = >20% below mean.
a
µgg
−1
ww.
b
For PCDDs and P CDFs only.
c
ng kg
−1
ww.
d
W = Willamette River, O = Other lakes and ponds.
e
McKenzie River: RM 14 for Osprey egg, RM 14.7 for Paper Mill.
f
Old Paper Mill at Salem.
Riv er is uncertain, although the use of pentachlorophenol (PCP) as a wood pre-
servative and subsequent contamination of local benthic food chains appeared to
be the main source of OCDDs in Ospreys from British Columbia and along the
Columbia River (Elliott et al., 1998). PCP was reported in water (0.10–0.70 µg
L
−1
) from the Willamette River (Buhler et al., 1973). Adult Largescale Suckers
are opportunistic omniv ores and primarily feed on a variety of benthic or ganisms
including aquatic insect larvae, earthworms, snails and detritus.
4.1.3. Wintering Ground and Breeding Ground Point Sources of Contaminants
Additional accumulation of contaminant residues by some Osprey, but perhaps not
all, on wintering grounds seems plausible. Although the general pattern of increas-
ing DDE egg residues to w ard the mouth of the river was observed (Table XIII),
several seemingly random spikes (the e gg from the McKenzie River with high
DDE; the egg at RM 124 with a unique DDE:DDT ratio (2:1) which implied
recent use of DDT (see Henny et al., 1982)) support additional wintering ground
DDE:DDT sources. With the 400 day half-life of DDE in large birds (Clark et al..
1987), wintering grounds exposure may especially influence DDE concentrations
308 C. J. HE NNY ET AL.
in Osprey eggs. Elliott et al. (2000) implied that some DDE exposure occurred out-
side the breeding area for Ospreys nesting along the Columbia and Fraser Rivers.
Ospreys from the Pacific Northwest winter over a wide region in southern Mexico
and Central America (Martell et al., 1998, 2001) where DDE:DDT exposure could
vary .
Paper mills located along the Willamette River were potential sources of PCDDs
and PCDFs (TEQs) and OCDD. Four of the 10 nests were located within 4 km
downstream of mills and TEQs and OCDD scores with 8 possible outcomes (same
scoringprocedureasusedabove)wereasfollows:+,0,+,+,0,+,+,and–foran
ov erall score of +4 (Table XIII). Thus, known point sources seemed to influence
PCDDs and PCDFs in Osprey eggs, with extremely low concentrations found in
reference area eggs from the Cascade Mountains (Table VIII).
4.2. F
ISH IN DIET AND BIOMAGNIFICATION FACTORS
Analysis of prey use based exclusively on prey remains without observations of
prey deliveries to nests has certain biases. Generally, small soft-boned fish (e.g.,
trout) are under-represented, and large bony fish (e.g., carp and bullheads) are over-
represented in prey remains. In this study, the species composition represented in
the prey remains found under nest and nearby perch sites was similar to the percent
composition of fish species we observed delivered to the same nests and supports
our belief that the diet of the sampled Ospreys was accurately estimated. The two
most important fish species in the Osprey diet (Largescale Sucker and Northern
Pikeminnow) were taken by Ospreys in sizes comparable to those collected by
electrofishing for contaminant evaluation, which is important for BMF calcula-
tions. Fish movement could influence contaminant uptake by Ospreys. However,
the BMFs we calculated are based upon means for the whole river (both fish con-
centrations and Osprey egg concentrations) which should minimize any influence
of local fish movement. The geographical pattern reported earlier of higher TEQs
and OCDD scores in Osprey eggs downstream of paper mills (a source) further
implies that fish mov ement is not a major issue.
BMFs were based upon geometric means of contaminants in fish and Osprey
eggs (on a wet weight basis) from the Willamette River (excluding Santiam Riv er
fish where no Osprey eggs were collected). Although some data limitations were
mentioned earlier and residue concentrations showed variability among locations,
we believe BMFs (fish to Osprey eggs) are instructive and can be used to assess
potential hazards for Osprey and other fish-eating birds. BMFs for the various OC
pesticides, PCBs, PCDDs and PCDFs (ranging from no biomagnification (0.42) for
2,3,7,8-TCDF to a high (174) for OCDD) help us understand basic relationships
between contaminant burdens in prey species of fish-eating birds and contaminants
eventually incorporated into fish-eating bird eggs. We belie ve BMFs can be used to
evaluate the potential hazard of contaminants in fish to those fish-eating birds liv ing
in the area; howe ver, knowledge is also required about the birds diet and sensitivity
OSPREY BIOMAGNIFICATION FACTORS 309
to contaminants. Conversely, the Osprey can be used as an indicator/sentinel spe-
cies (an integrator of contaminants) by collecting only sample eggs, and projecting
via BMFs estimated contamination in fish communities represented in the diet from
v arious large rivers, bays and estuaries.
BMFs for migratory Ospreys were compared to findings for a non-migratory
Herring Gull (Larus argentatus) population from Lake Ontario in 1985 (Braune
and Norstrom, 1989). Some concern was noted by Braune and Norstrom (1989)
that Alewife (Alosa pseudoharengus), the only prey species sampled, represented
only a portion of the gulls diet, but they reported findings that Rainbow Smelt
(Osmerus mordax), the other major fish species in the diet, had similar lev els of
organochlorines. Ryckman et al. (1997) summarized 25 yr of Herring Gull studies
from the Great Lakes and reported Alewife (26%) and Rainbow Smelt (32%) as the
most important species in the diet, but other fish (15%), birds (9%), small mammals
(7%), garbage (4.5%), and miscellaneous items (6.5%) were also consumed. The
BMFs for the Herring Gull (fish to egg) were not calculated by the original authors;
only BMFs from fish to whole body (gull) were made, although data were available
for us to calculate fish to egg BMFs.
Three potential issues may confound a direct comparison of BMFs for Herring
Gulls and Ospreys: (1) a few Ospreys on the nesting grounds sometimes foraged
at ponds or lakes away from the Willamette River (all fish were sampled from the
river, see Section 4.1.2) which yielded lower concentrations of industrial contamin-
ants in their eggs than expected, (2) the Herring Gull is a year-around resident, but
the migrant Osprey only spends about 1 month at the nesting area (where all fish
were sampled) before laying eggs; egg concentrations could be influenced by the
presence or lack of persistent contaminants on wintering grounds thousands of kms
away, and (3) lipid content of eggs, if considerably different between species,
will influence concentrations of lipophilic contaminants when presented on a wet
weight basis. Eggs of Ospreys contained 4.3% lipid, while Herring Gulls contained
7.7% lipid, thus, the gulls had 1.79 times as much lipid which is considerable. To
adjust for this difference, a second BMF value was presented (in parentheses) for
the Osprey in Tables XI and XII (the original BMF X 1.79 to normalize to lipid
content of Herring Gull eggs).
The higher BMF (87) calculated for DDE in Osprey eggs in this study (based
only on fish from Willamette River) dif fers considerably from the BMF (34) for
Herring Gulls (Table XI) and the BMF (31) for Brown Pelicans (Pelecanus oc-
cidentalis)(Bluset al., 1977). Additionally, a BMF of 47 for the parent DDT as
opposed to a BMF of only 2 for Herring Gulls again differs considerably. The
lipid adjusted BMFs for Ospreys are even higher for DDE and DDT (156 and
84). These differences can be accounted for by higher DDE/DDT in fish captured
by some Ospreys on their wintering grounds (and its long half-life in birds) as
suggested by Elliott et al. (2000). The non-lipid adjusted BMFs for several other
organochlorine pesticides and their metabolites from this study and the Herring
Gull study were similar: mirex 35 (adjusted 63) vs. 30, HE 25 (45) vs. 30, dieldrin
310 C. J. HE NNY ET AL.
6.7 (12) vs. 7.0, trans-nonachlor 1.8 (3.2) vs. 2.6, and cis-nonachlor 11 (20) vs. 5,
bu t with substantial differences for oxychlordane 21 (38) vs. 60 and HCB 1.2 (2.1)
vs. 20. Limited data from the gulls were available to compare BMFs for PCDDs
and PCDFs.
Non-lipid adjusted BMFs for PCB congeners from this study were consistently
lo wer than calculated for the Herring Gull, and often by a factor of 2 or more, but
the lipid adjustment made the values somewhat closer. Howe ver, the BMFs still
remained lower than for gulls. Possible reasons for lower BMFs for PCBs (e ven
with the lipid adjustment) in Osprey include a combination of (1) some Ospreys
foraging in ponds or lakes away from the Willamette River (see discussion above),
and (2) residual ef fects on Osprey body burdens from lower PCBs in fish taken on
the wintering grounds in southern Mexico and northern Central America where
sparse industrial activity should result in less PCB contamination. The Osprey
migration period would have a minimal effect because it is very short (mean 13
days).
We believe general residue patterns for Osprey eggs collected along the river
and the calculation of BMFs can be useful and instructive. For migratory birds
and some persistent contaminants with different use patterns on the breeding and
wintering grounds, calculated BMFs may possibly provide insightful information
about relative sources of contaminants found in eggs. This concept needs to be
more fully explored and additional research is planned. Our Osprey population
increased to 234 nesting pairs in 2001 (Henny and Kaiser, unpubl. data) including
many pairs now nesting in Reaches I and II (Tidal Reach and Newberg Pool) which
are the most contaminated (e.g., Curtis et al., 1993; Bonn, 1998). BMFs will be
evaluated in the future for Ospreys nesting along five segments of the river with
differing local exposure to contaminants.
4.3. C
ONTAMINANTS AND OSPREY PRODUCTIVITY
Before evaluating effects of contaminants on Osprey productivity two points need
to be emphasized: (1) the Osprey population nesting along the Willamette Riv er
increased in recent years at a rapid rate (from 13 pairs in 1976 to 78 pairs in
1993) (Henny and Kaiser, 1996), and (2) the observed production rate in 1993
(1.64 young/active nest) was about twice the estimated rate required to main-
tain a stable population (0.80 young/active nest) in the northeastern United States
(Spitzer, 1980; Spitzer et al., 1983). More recent information mentioned abo ve
showed that the population continued to increase (234 nesting pairs in 2001), thus,
aprioricontaminant effects were minimal. The three nests with an egg sampled at
Crane Prairie Reservoir (the reference area) produced 1.67 young/active nest.
Wiemeyer et al. (1988) reported that 4.2 µgg
−1
DDE was associated with 15%
shell thinning in Osprey eggs and 8.7 µgg
−1
DDE with 20% shell thinning, and
Lincer (1975) pointed out that not one North American raptor population exhibiting
18% or more eggshell thinning has been able to maintain a stable self-perpetuating
OSPREY BIOMAGNIFICATION FACTORS 311
population. Therefore, mean DDE concentrations in eggs somewhere between 4.2
and 8.7 µgg
−1
would be expected to result in a declining Osprey population.
Clearly, the Willamette River population with a geometric mean of 2.3 µgg
−1
DDE was below the critical mean. Only 2 of 10 Willamette River nests with an egg
collected in 1993 contained DDE concentrations ≥4.2 µgg
−1
. A nest with 5.6 µg
g
−1
DDE in the sample egg at RM 124 produced 2 young and the nest with 12.1 µg
g
−1
DDE at RM 189 failed. DDE may be adversely impacting a few individuals in
this generally thriving Osprey population. The Crane Prairie Reservoir pairs with
an egg collected had 6.7, 4.5, and 7.1 µgg
−1
DDE and all pairs nested successfully,
fledging 2, 2 and 1 young, respectively. All other OC pesticides were judged low
and well below known effect concentrations.
Concentrations of 2,3,7,8-TCDD in Osprey eggs were generally low and within
a narro w range (geo. mean 2.3 ng kg
−1
, range 1.5 to 5.6), and not significantly
different from eggs collected along the lower Columbia River (below Bonneville
Dam) in 1995–1996 (geo. mean 4.3 ng kg
−1
, range 1.9 to 26) (Elliott et al., 1998).
However, Osprey eggs from both the Willamette River and the Columbia Riv er
had significantly higher 2,3,7,8-TCDD than eggs from the Crane Prairie Reservo ir
reference area (0.46 ng kg
−1
, range 0.4 to 0.6). OCDD (TEF = 0.0001) is not
very toxic when compared to 2,3,7,8-TCDD (TEF = 1), but concentrations among
Osprey eggs from the same three Osprey populations in the Pacific Northwest were
quite different: W illamette Riv er (geo. mean 1299 ng kg
−1
, range 255 to 3489) was
significantly higher than the Columbia River (18 ng kg
−1
, range 1.5 to 79) (Elliott
et al., 1998) or Crane Prairie Reservoir (53 ng kg
−1
, range 10.5 to 219). The latter
two locations were not significantly different.
Woodford et al. (1998) reported much higher 2,3,7,8-TCDD in Osprey eggs
(geo. mean 78 ng kg
−1
, range 29 to 162) collected in 1992–1996 downstream
from two bleached-kraft mill facilities on the Castle Rock and Petenwell Rivers,
Wisconsin (Woodford et al., 1998). In the Wisconsin study, no association between
egg-hatching or chick fledgling rates was observ ed between their two study areas;
however, Woodford et al. (1998) concluded that current exposure levels may hav e
affected normal development of Osprey chicks downstream of the paper mills. The
no-observed-adv erse-effect level (NOAEL) for Osprey eggs hatching was ≥136 ng
kg
−1
combined TEQs (PCDDs, PCDFs, PCBs) (Woodward et al., 1998). Findings
for Osprey eggs collected on the Fraser and Columbia River in 1995 and 1996
and artifically incubated (Elliott et al., 2001) support this NOAEL. Elliott et al.
(2001) reported enzyme induction (lowest observed effect le vel) in Osprey chicks
(hepatic CYPIA) at 130 ng kg
−1
TEQs with the NOAEL at 37 ng kg
−1
. Combined
TEQs for the Osprey eggs along the Willamette River in this study (geo. mean
43.32 ng kg
−1
(range 31.23 to 99.04)) were much lower than the NOAEL for egg
hatching and combined TEQs were ev en lower in our reference area (geo. mean
13.28 ng kg
−1
(range 9.66 to 18.59) (Table IX). Therefore, with the exception of a
small portion of the population affected by DDE (probably from wintering ground
sources), no evidence was found for other contaminants adversely affecting the
312 C. J. HE NNY ET AL.
Osprey population nesting along the Willamette Riv er. The general release from
DDE problems in the 1950s–1970s, following the DDT ban in 1972 was one of
the primary reasons the Willamette River population has increased at a rapid rate
(Henny and Kaiser , 1996), as well as Osprey populations throughout the United
States (Houghton and Rymon, 1997).
Acknowledgements
We appreciate the assistance of R. Goggans, C. Heath and J. Pesek of the Oregon
Department of Fish and W ildlife, and thank J. Gunter for providing background
data on Ospreys nesting in the study area. G. Foster from the Oregon Department
of Environmental Quality assisted with electrofishing. R. Wildman, Oregon State
University, provided information on recent fish distribution in the Willamette River.
Logistical support and additional information were provided by personnel from
Consumers Power, Philomath; Emerald Peoples Utility District, Eugene; Eugene
Water and Electric, Eugene; Pacific Power and Light, Albany and Independence;
Portland General Electric, Salem; and Salem Electric, Salem. The lando wners were
most helpful; we appreciate access to their land and the historical information about
occupied Osprey nests. J. Anderson assisted with egg collections at Crane Prairie
Reservoir. G. Lienkaemper, U.S. Geological Survey, helped prepare the map.
References
Altman, B., Henson, C. M. and Waite, I. R.: 1997, ‘Summary of information on aquatic biota and
their habitats in the Willamette Basin, Oregon, through 1995’, U .S. Geol. Surv. Water-Resour.
Invest. Rept., 97–4023, Portland, OR 174 pp.
Anderson, D. W. and Hickey, J. J.: 1976, ‘Dynamics and storage of organochlorines in Herring Gulls’,
Environ. Pollut. 10, 183–200.
Ballschmiter, K. and Zell, M.: 1980, ‘Analysis of polychlorinated biphenyls (PCB) by glass capillary
gas chromatography. Composition of technical Aroclor-and Clophen-PCB mixtures’, Fres. Z.
Anal. Chem. 302, 20–31.
Blus, L. J., Neely Jr., B. S., Lamont, T. G. and Mulhern, B.: 1977, ‘Residues of organochlorines and
heavy metals in tissues and eggs of Brown Pelicans, 1969–1973’, Pestic. Monit. J. 11, 40–53.
Bonn, B. A.: 1998, ‘Dioxins and Furans in Bed Sediment and Fish Tissue of the Willamette Basin,
Oregon, 1992–1995’, U.S. Geol. Surv. Water-Resour. Invest. Rept. 97-4082-D.
Bonn, B. A., Hinkle, S. R., Wentz, D. A. and Uhrich, M. A.: 1995, ‘Analysis of Nutrient and Ancillary
Water-quality Data for Surface and Ground Water of the Willamette Basin, Oregon, 1980–1990’,
U.S. Geol. Surv. Water-Resour. Invest. Rept. 95–4036, 88 pp.
Braune, B. M. and Norstrom, R. J.: 1989, ‘Dynamics of organochlorine compounds in Herring Gulls:
III. Tissue distribution and bioaccumulation i n Lake Ontario gulls’, Environ. Toxicol. Chem. 8,
957–968.
Buhler, D. R., Rasmusson, M. F. and Nakaue, H. S.: 1973, ‘Occurrence of hexachlorophene and
pentachlorophenol in sewage and water’, Environ. Sci. Tech. 7, 929–934.
Center for Population Research and Census: 1992, Population Estimates for Oregon 1980–1981,
Portland State University, Portland, Oregon, 22 pp.
OSPREY BIOMAGNIFICATION FACTORS 313
Clark, T. P., Norstrom, R. J., Fox, G. A. and Won, H. T.: 1987, ‘Dynamics of organochlorine
compounds in Herring Gulls (Larus argentatus). II. A two-compartment model and data for ten
compounds’, Environ. Toxicol. Chem. 6, 547–559.
Curtis, L. R., Carpenter, H. M., Donohoe, R. M., Williams, D. E., Hedstrom, O. R., Deinzer, M.L.,
Beilstein, M. A., Foster, E. and Gates, R.: 1993, ‘Sensitivity of cytochrome P450-1A1 induction
in fi sh as a biomarker for distribution of TCDD and TCDF in the Willamette River, Oregon’,
Environ. Sci. Tech. 27, 2149–2157.
Dauble, D. D.: 1986, ‘Life history and ecology of the Largescale Sucker (Catostomus macrocheilus)
in the Columbia River’, Amer. Midl. Nat. 116, 356–367.
Dimick, R. D. and Merryfield, F.: 1945, ‘The fishes of the Willamette River system in relation to
pollution’, Engin. Expt. Sta. Bull. Ser. 20, 7–55 (Oregon State College, Corvallis, OR).
Elliott, J. E., Machmer, M. M., Henny, C. J., Wilson, L. K. and 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., Machmer, M. M., Wilson, L. K. and Henny, C. J.: 2000, ‘Contaminants in Ospreys from
the Pacific Northwest: II. Organochlorine pesticides, polychlorinated biphenyls and mercury,
1991–1997’, 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. and Cheng,
K. M.: 2001, ‘Assessment of biological effects of chlorinated hydrocarbons in Osprey chicks’,
Environ. Toxicol. Chem. 20, 866–879.
Ewins, P. J., Postupalsky, S., Hughes, K. D. and 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.
Gabrielson, I. N. and Jewett, S. G.: 1940, The Birds of Oregon, Oregon State College Press, Corvallis,
650 pp.
GLIER: 1995, Methods and Pr ocedures Quality Manual, 1st ed., O Revision. Great Lakes Institute
for Env ironmental Research, University of Windsor, Windsor, Ontario, Canada (July 1995).
Gonthier, J. B.: 1985, ‘Oregon Ground-water Resources’, in U.S. Geological Survey, National
Water Summary 1984 – Hydrologic Events, Selected Water-quality Trends, and Ground-water
Resour ces, U.S. Geol. Surv. Water-Supply Paper 2275. pp. 355–360.
Gregory, S. V.: 1993, ‘Willamette River Basin S tudy – Periphyton Algal Dynamics’, Final Report,
prepared for Oregon Dept. Environ. Quality, and Dept. Fish and Wildlife. Oregon State Univ.
Corvallis, 112 pp.
Henny, C. J. and Kaiser, J. L.: 1996, ‘Osprey Population Increase Along the Willamette River, Ore-
gon, and the Role of Utility St ructures, 1976–1993’, in D. M. Bird, D. E. Varland and J. J. Negro
(eds), Raptors in Human Landscapes, Academic Press, Ltd., London, pp. 97–108.
Henny, C. J. and Nelson, M. W.: 1981, ‘Decline and present status of breeding Peregrine Falcons in
Oregon’, Murrelet 62, 43–53.
Henny, C. J., Collins, J. A. and Deibert, W. J.: 1978, ‘Osprey distribution, abundance, and status in
western North America: II, The Oregon Population’, Murrelet 59, 14–25.
Henny, C. J., Maser, C., Whitaker Jr., J. O. and Kaiser, T. E.: 1982, ‘Organochlorine residues in bats
after a forest spraying with DDT’, Northwest Sci. 56, 329–337.
Hines, W. G., McKenzie, S. W., Rickert, D. A. and Rinella, F. A.: 1977, ‘Dissolved-oxygen regime of
the Willamette River, Oregon, under conditions of basin-wide secondary treatment’, U.S. Geol.
Surv. Circ. 715–1.
Houghton, L. M. and Rymon, L. M.: 1997, ‘Nesting distribution and population status of U.S.
Ospreys, 1994’, J. Raptor Res. 31, 44–53.
Huff, E. S. and Klingeman, P. C.: 1976, ‘Restoring the Willamette River: Costs and impacts of water
quality control’, J. Water Pollut. Control Fed. 48, 2410–2415.
Hughes, R. M. and Gammon, J. R.: 1987, ‘Longitudinal changes in fish assemblages and water
quality in the Willamette River, Oregon’, Trans. Am. Fish. Soc. 116, 196–209.
314 C. J. HE NNY ET AL.
Kammerer, J. C.: 1990, ‘Largest Rivers in the United States’, U.S. Geol. Surv. Open File Rept.,
87–242, 2 pp.
Lazar, R., Edwards, R. C., Metcalfe, C. D., Metcalfe, T., Gobas, F. A. P. C. and Haffner, G. D.:
1992, ‘A simple, novel method for the quantitative analysis of coplanar (non-ortho substituted)
polychlorinated biphenyls in environmental samples’, Chemosphere 25, 493–504.
Li, H. W., Schreck, C. B., Bond, C. E. and Rextad, E.: 1987, ‘Factors Influencing Changes in Fish
Assemblages of Pacific Northwest Streams’, in W. J. Matthews and D. C. Heins (eds), Community
and Evolutionary Ecology of North American Stream Fishes, Univ. Oklahoma Press, Norman,
pp. 193–202.
Lincer, J. L.: 1975, ‘DDE-induced eggshell thinning in the American Kestrel: A comparison of the
field situation with laboratory results’, J. Appl. Ecol. 12, 781–793.
Martell, M. S., Kennedy, M. J., Henny, C. J. and Nye, P. E.: 1998, ‘Highway to the Tropics: Us-
ing Satellite Telemetry and the Internet to Track Ospreys in the Western Hemisphere’, in Y.
Leshem. E. Lachman and B. Berthold (eds), Migratory Birds Know no Boundaries: Scientific
and Educational Aspects of Migrating Bird Conservation, Tel Aviv, Israel, pp. 163–172.
Martell, M. S., Henny, C. J., Nye, P. E. and Solensky, M. J.: 2001, ‘Fall migration routes, timing, and
wintering sites of North American Ospreys as determined by satellite telemetry’, Condor 103,
715–724.
Martinelli, T. L. and Shively, R. S.: 1997, ‘Seasonal distribution, movements and habitat associations
of Northern Squawfish in two lower Columbia River reservoirs’, Regul. Rivers Res. Manage. 13,
543–556.
McFarland, W. D.: 1983, ‘A Description of Aquifer Units in Western Oregon’, U.S. Geol. Surv.
Open-File Report, 82–165, 35 pp.
Newsome, G. E.: 1977, ‘Use of opercular bones to identify and estimate lengths of prey consumed
by piscivores’, Can. J . Zool. 55, 733–736.
Oregon Department of Environmental Quality: 1988, Oregon Environmental Atlas, Portland, Oregon,
64 pp.
Parkhurst, Z. E., Bryant, F. G. and Nielson, R. S.: 1950, ‘Survey of the Columbia River and its
Tributaries, Part 3’, U.S. Fish and Wildlife Service, Special Science R eport – Fisheries, No. 36,
103 pp.
Poole, A. F.: 1989, Ospr eys – A Natural and Unnatural History, Cambridge University Press,
Cambridge, 246 pp.
Postupalsky, S.: 1977, ‘A Critical Review of Problems in Calculating Osprey Reproductive Success’,
in J. C. Ogden (ed.), Transactions No. American Osprey Research Conf., Trans. and Proc. Series,
No. 2, Natl. Park Serv., Washington, D.C., pp. 1–11.
Rickert, D. A., Hines, W. G. and McKenzie, S. W.: 1975, ‘Methods and data requirements for river-
quality assessment’, Water Resour. Bull. 11, 1013–1039.
Ryckman, D. P., Weseloh, D. V. and Bishop, C. A.: 1997, Contaminants in Herring Gull Eggs from
the Great Lakes: 25 yr of Monitoring Levels and Effects, Great Lakes Fact Sheet, Environment
Canada, Burlington, Ontario.
SAS Institute: 1996, ‘SAS User’s Guide: Statistics’, Version 6.12 Edition, SAS Inst., Inc, Cary, NC.
Spitzer, P. R.: 1980, ‘Dynamics of a Discrete Coastal Breeding Population of Ospreys in the North-
eastern United States During the Period of Decline and Recovery, 1969–1978’, Ph.D. Thesis,
Cornell University, Ithaca, NY.
Spitzer, P. R., Poole, A. F. and Scheibel, M.: 1983, ‘Initial Recovery of Breeding Ospreys in the
Region between New York City and Boston’, in D. M. Bird (chief ed.), Biology and Management
of Bald Eagles and Ospr eys, Harpell Press, Ste. Anne de Bellevue, Quebec, pp. 231–241.
Stickel, L. F., Wiemeyer, S. N. and Blus, L. J.: 1973, ‘Pesticide residues in eggs of wild birds:
Adjustment for loss of moisture and lipid’, Bull. Environ. Contam. Toxicol. 9, 193–196.
Swenson, J. E.: 1978, ‘Prey and foraging behavior of Ospreys on Yellowstone Lake, Wyoming’, J.
Wildl. Manage. 42, 87–90.
OSPREY BIOMAGNIFICATION FACTORS 315
Tetra Tech.: 1992, Characterization of Willamette River Main-stem Ecoregions – Component 9,Tetra
Tech, Inc., Bellevue, WA., 27 pp (prepared for Oregon Dept. Environmental Quality, Portland,
OR).
Uhrich, M. A. and Wentz, D. A.: 1999, ‘Environmental setting of the Willamette Basin, Oregon’,
U.S. Geol. Surv. Water-Resour. Invest. Rept., 97-4082-A.
Van Daele, L. J., Van Daele, H. A. and Johnson, D. R.: 1980, ‘The Status and Management of Ospreys
Nesting in Long Valley, Idaho’, Univ. Idaho, Dept. Biological Sciences, Moscow, ID. 49 pp.
Van den Berg, M., Birnbaum, L., Bosveld, A. T., Brunstrom, B., Cook, P., Feeley, M., Giesy, J.
P., Hanberg, A., Hasegaw a, R., Kennedy, S. W., Kubiak, T., Larsen, J. C., v an Leeuwen, F. X.,
Liem, A. K., Nolt, C., Peterson, R. E., Poellinger, L., Safe, S., Schrenk, D., Tillitt, D., Tysklind,
M., Younes, M., Waern, F. and Zacharewski, T.: 1998, ‘Toxic equivalency factors (TEFs) for
PCBs, PCDDs, PCDFs for humans and wildlife’, Environ. Health Perspect. 106, 775–792.
Wentz, D. A. and McKenzie, S. W.: 1991, ‘National Water-Quality Assessment Program – The
Willamette Basin, Oregon’, U.S. Geol. Surv. Open-File Rept., 91–167, 2 pp.
W iemeyer, S. N., Bunck, C. M. and Krynitsky, A. J.: 1988, ‘Organochlorine pesticides, polychlor-
inated biphenyls, and mercury in Osprey eggs – 1970–1979 – and their relationships to shell
thinning and productivity’, Ar ch. Envir on. Contam. Toxicol. 17, 767–787.
Witt, J. W.: 1990, Productivity and management of Osprey along the Umpqua River, Oregon’,
Northwestern Nat. 71, 14–19.
Woodford, J. E., Karasov, W. H., Meyer, M. W. and Chambers, L.: 1998, ‘Impact of 2,3,7,8-TCDD
exposure on survival, growth, and behavior of Ospreys breeding in Wisconsin, U.S.A.’, Environ.
Toxicol. Chem. 17, 1323–1331.
Wydoski, R. S. and Whitney, R. R.: 1979, Inland Fishes of Washington, Univ. Washington Press,
Seattle, 220 pp.
