Abstract
Introduction:
Ospreys (Pandion haliaetus) are well-known sentinels of aquatic ecosystem health and are indicators of both environmental contaminants and fish stocks. The Chesapeake Bay supports one of the largest osprey breeding populations in the world, but recent studies have documented declining reproductive performance and increasing food stress in some portions of the estuary.
Methods:
We monitored osprey nests (N = 571) throughout the Chesapeake Bay during the 2024 breeding season and compared breeding metrics between high-salinity (>10 parts per thousand [ppt]) and low-salinity (<5 ppt) study areas. We also compared contemporary breeding performance within four high-salinity sites to historical data collected during the 1980s.
Results:
Salinity was strongly associated with breeding performance and the likelihood that pairs achieved productivity levels required for population maintenance. All high-salinity study areas functioned as demographic sinks, whereas low-salinity areas functioned as demographic sources. Breeding metrics including the proportion of pairs breeding, clutch size, nesting failure, brood reduction, and nestling loss all suggested greater food stress within high-salinity areas. Temporal comparisons documented substantial declines in reproductive performance between the 1980s and 2024 within high-salinity study sites. High-salinity sites during the 1980s exhibited breeding performance comparable to low-salinity sites in 2024.
Discussion:
Although several factors may influence osprey productivity within the Chesapeake Bay, we suggest that reduced availability of Atlantic menhaden (Brevoortia tyrannus) is a primary driver of poor reproductive performance within high-salinity waters.
1 Introduction
Breeding ospreys (Pandion haliaetus) are considered among the most effective sentinels of aquatic ecosystem health on a global scale (; ). Ospreys have a large geographic range, occupy the top of the aquatic food web, are easy to observe and are tolerant of field research activities (; ). Ospreys both accumulate and are sensitive to a wide range of contaminants in the environment (e.g., ; ; ; ). Contaminants may be evaluated at the level of specific tissues (e.g., ; ; ; ; ; ) or in terms of population responses making it possible to link exposure to demographic consequences (; ; ). In addition to their usefulness as an indicator of contaminant exposure and adverse effects, ospreys have behavioral adaptations that adjust brood size (via brood reduction) to provisioning rate making brood size a general indicator of fish abundance (e.g., ; ; ; ; ;Watts et al., 2024). Because ospreys hatch asynchronously, insufficient provisioning leads to food competition and the formation of a stable dominance hierarchy within the brood maintained by sibling aggression (). Further reductions in provisioning lead to the sequential starvation of subordinate young (brood reduction) and ultimately nest failure ().
The Chesapeake Bay supports one of the largest osprey breeding populations in the world (; ). This population has been monitored for more than 50 years (Watts and Paxton, 2007). Coincident with the widespread use of DDT as a pesticide in the late 1940s, the population experienced reduced productivity (Wiemeyer, 1971) leading to an estimated 80% population decline (). The population reached a low of 1, 450 pairs in the early 1970s (; ). Following a cancellation order for DDT in 1972, the concentration of DDT (and metabolites) declined in osprey eggs () and reproductive rates recovered (Watts and Paxton, 2007). The Chesapeake Bay population responded with an exponential recovery reaching 3, 500 breeding pairs by the mid-1990s () and an estimated 10, 000 pairs today (Watts et al., 2024, unpublished data). The rate of recovery varied over an order of magnitude between geographic areas of the Bay (). Although the early recovery was centered within the salty main stem of the Bay (>10 parts per thousand (ppt) salinity), the more recent and ongoing recovery has been within the lower (<5 ppt) salinity reaches of the Bay where major tributaries receive large freshwater inputs (, Watts et al., 2024, unpublished data).
Researchers began to observe food stress (i.e., inadequate nutrition due to food scarcity or low quality) in Mobjack Bay, a sub-estuary within the Chesapeake Bay, during the late 1980s (; ). By the 2000s, food stress and associated brood reduction were widespread and the reproductive rate had fallen below the level required for population maintenance (). A forty-year retrospective evaluation (Watts et al., 2024) documented that the loss of young after hatching increased from 5% in the mid-1970s to more than 75% by the mid-2000s. In Mobjack Bay, the transition from a demographic source, where pairs were producing above maintenance, to a demographic sink was associated with a 74% decline in overall provisioning rate and a shift (67% to 25%) in diet composition away from Atlantic menhaden (Brevoortia tyrannus). The rate of menhaden delivery to nests declined by more than 80% over the study period whereas the delivery rate of other major fish species in the diet increased (; ; ). Because the energy density for menhaden is among the highest within the diet, the shift in diet composition away from menhaden resulted in a 50% decline in the overall energy content of the diet. The singular decline of menhaden in the diet reduced energetic provisioning below a threshold for demographic stability. A study in which nests were supplemented with menhaden demonstrated that returning the menhaden provisioning rates back to 1980s levels would successfully drive reproductive rates above maintenance levels ().
Patterns in food and demographic stress documented in Mobjack Bay invite questions about the geographic scope of deficits. Specifically, is Mobjack Bay an anomaly or are there other places throughout the Chesapeake Bay that are experiencing poor levels of breeding performance? During the 2024 breeding season, we increased the geographic footprint of osprey monitoring to other study areas throughout the Chesapeake Bay to examine broader patterns in reproductive performance and metrics that may help to explain drivers. Herein, we report and compare results from high salinity sites to reference sites within low salinity waters, as well as contemporary results to those from the 1980s.
2 Methods
2.1 Study area and sample sites
We conducted fieldwork with osprey within the Chesapeake Bay. We used the Chesapeake Bay Program’s salinity segmentation scheme (https://www.chesapeakebay.net/what/maps) to delineate waters and study areas within two salinity categories including high (average surface salinity ≥ 10 ppt) and low (average surface salinity <5 ppt) (Figure 1). Diet studies from the Chesapeake Bay show that diet composition shifts with salinity (; ) and that pairs nesting in higher salinity (≥10 ppt) areas appear to rely on Atlantic menhaden () but those nesting in low salinity waters depend primarily on catfish (Ictalurus spp.) and gizzard shad (Dorosoma cepedianum). We selected ten sample sites within high salinity waters and two reference sites within low salinity waters for comparison (Figure 1). We selected study sites based on ease of navigation and accessibility of osprey nests for monitoring. Each sample site supported 30 to 80 osprey pairs. All sample sites are microtidal (amplitude < 1 m) within two tide cycles-d. The osprey population nesting within the Chesapeake Bay is migratory. Osprey pairs return to breeding territories from South America during late February through late March, lay clutches in early April and fledge young from mid-June through late July (). Within the study area, ospreys primarily nest on navigational aids, offshore duck blinds, boat houses, abandoned docks and nesting platforms erected by property owners.
Figure 1
2.2 Osprey demography
The osprey is a long-lived species with associated high adult survival and relatively high reproductive potential (). Ospreys exhibit delayed onset to reproduction. For growing populations, recruitment into the breeding population typically occurs during the fourth year with a smaller number entering the population during their third or fifth year (). Age-at-first reproduction increases as populations reach carrying capacity (; ). used osprey band recoveries during two time periods to estimate juvenile and adult survival and then used a matrix population model to estimate the productivity required to sustain a stable population (). They estimated that first-year survival ranged from 42.7% to 48.5%, and survival for older age classes combined ranged from 81.5% to 83.8%. They estimated that 0.95 and 1.30 young/breeding pair/year would be required to offset mortality for the two time periods with a combined estimate of 1.22. More recent survival estimates include 60% for first-year birds and 85% for older age classes (). estimated that 1.15 young/breeding pair/year would be required to offset adult mortality within the Chesapeake Bay. extrapolated a break-even reproductive rate (0.8 young/breeding pair/year) in southern New England during a period of population expansion using a combination of population growth and young production. The break-even rate likely fluctuates as the population of floaters (birds of breeding age that do not hold territories) expands and contracts through time.
2.3 2024 field season
We monitored focal osprey nests (N = 571) at least four times during the 2024 breeding season (mid-March through mid-August) by boat to determine breeding performance. We used a telescopic mirror pole to facilitate the examination of nest contents for nests that were >2 m above the water line. The presence of adults and the number of eggs and young in the nest was recorded during each visit. We considered a breeding territory to be occupied if two adults were present and associated with a nest by 15 April. Pairs arriving after 15 April were considered to be juveniles prospecting for territories to be used in following years. Because these birds have not recruited into the breeding population, they were not included in metrics of breeding performance (). We considered a nest to be active if direct evidence (eggs or young detected in nest) of egg laying was recorded. We considered young that reach six weeks of age to be of “near fledging” age and used this as a milestone for fledging in place of directly observing first flights. A nest was deemed to be successful if a pair produced ≥1 young to fledging age. We considered productivity to be the number of young reaching fledging age.
Additional metrics were derived from observations including % of pairs that laid eggs, % of pairs that laid eggs that either were successful or failed to produce ≥1 young, reproductive rate, clutch size, % of eggs hatched, young loss rate, % of broods that were single young and the probability of young surviving to fledging age. We define reproductive rate to be the mean number of young reaching fledging age divided by the number of occupied territories (). We consider clutch size to be the highest number of eggs recorded within a nest. We estimated the % of eggs hatching where the outcome could be determined based on observations. Eggs within nests with young older than one week were considered to be addled. Most of these eggs were recorded in the nest for extended periods of time. We considered all eggs to have hatched when eggs laid equaled the highest number of young recorded. We considered hatching to be unknown for nests where the first observation of young was less than the clutch size. Because some young may have hatched and been lost prior to the next visit, our estimate of % hatching should be considered conservative. We determined the number of young lost that could be determined based on observations as the difference between the number of young known to hatch and the number of young fledged. The rate of young loss was estimated as the mean number of young lost between hatching and fledging per pair with known outcome. We estimated the probability of surviving the nestling period as the number of young fledged divided by the number of young known to hatch.
2.4 1980s field seasons
Four sample sites that were monitored during the 2024 nesting season were also monitored during the 1980s (1985, 1986, 1987, 1988). These sites include the York River, Mobjack Bay, Piankatank River, and Fleets Bay. Sampling schedule, approach and the metrics derived were consistent during the 1980s with the description above for the 2024 nesting season. Sample sizes within sampling sites were generally lower during the 1980s with mean ± SE including 20.0 ± 2.12 pairs for Fleets Bay, 20.0 ± 1.47 for Piankatank River, 32.8 ± 2.69 for Mobjack Bay and 14.3 ± 1.31 for the York River.
2.5 Statistical analysis
G-tests were used to compare the proportion of pairs that failed to lay eggs, likelihood of eggs hatching, and likelihood of nesting success. We used generalized linear models (GLMs) to estimate clutch size, reproductive rate, chick loss, and nestling survival in Program R v4.4.1 (). For comparisons between salinity zones, we used only 2024 data and grouped sites by salinity category (high vs. low). For comparisons between study periods (1980s vs. 2024), only data from high salinity sites were available and, because all study sites were consistent across periods, we included study site as a predictor rather than grouping them to identify any site-specific trends. To account for overdispersion, we used a quasi-Poisson regression to estimate clutch size (total number of eggs laid). Poisson regressions were used to estimate reproductive rate (total fledglings produced) and chick loss (total nestlings lost). A multinomial log-linear regression was used to estimate frequency distributions of brood size. To estimate nestling survival likelihood, we used a beta-binomial regression and included brood size as a predictor to account for brood size-dependent survival and also used brood size as a weighting variable so that larger broods contributed more to the likelihood estimation than smaller broods. We used the Delta method to combine error estimates when averaging survival among different brood sizes and site-level reproductive metrics ().
3 Results
3.1 2024 season
Breeding performance varied with salinity during the 2024 breeding season (β = -1.24 ± 0.03, p < 0.001) (Table 1) with the magnitude of differences between salinity types increasing as birds progressed through the breeding cycle (Figure 2). Overall reproductive rate (young/pair) ranged from 0.08 ± 0.05 to 0.90 ± 0.19 within high salinity areas with all but two sites falling below 0.8 young/pair. Reproductive rate for both low salinity sites was above 1.3. Across all high salinity areas, over 66.9% of pairs produced no young compared to 34.5% in low salinity areas. Poor breeding performance within high salinity areas was in part driven by both the high percentage of pairs that were never documented to lay clutches and the high percentage of pairs that did lay clutches but failed to produce young. A total of 132 pairs (21.7% of 487) within high salinity sites were never documented to lay clutches compared to 3.6% (3 of 84) of pairs within low salinity sites (G-statistic = 23.5, df = 1, p < 0.001). Clutch size varied with salinity (β = -0.11 ± 0.03, p = 0.002) with pairs in low salinity areas laying larger clutches (2.70 ± 0.07) compared to pairs in high salinity areas (2.5 ± 0.04) (Table 2). For eggs of known outcome, the likelihood of hatching did not vary between high salinity (79.7%: 397 of 498) and low salinity (73.9%: 88 of 119) areas (G-statistic = 0.29, df = 1, p = 0.3) (Table 2). The Choptank River was an outlier hatching only 38.0% of eggs laid. We did not document the underlying factors contributing to low hatching but the fact that many eggs disappeared before term suggests that they may have been taken by egg predators. For pairs that laid clutches, the likelihood of nesting success was nearly twice as large in low salinity (65.5%: 55 of 81) compared to high salinity (33.1%: 161 of 355) study areas (G-statistic = 5.2, df = 1, p = 0.011) (Table 1).
Table 1
| Site | Salinity (ppt) | Pairs | Pairs lot laying (%) | Successful pairs (%) | Failed pairs (%) | Reproductive rate (SE) |
|---|---|---|---|---|---|---|
| High Salinity (>10 ppt) | ||||||
| Choptank River | 12.2 | 60 | 21.7 | 18.3 | 60.0 | 0.23 (0.07) |
| Patuxent River | 12.5 | 49 | 22.4 | 34.7 | 42.9 | 0.51 (0.11) |
| Fleets Bay | 15.2 | 38 | 57.9 | 7.9 | 34.2 | 0.08 (0.05) |
| Eastern Shore | 19.0 | 57 | 14.0 | 40.4 | 45.6 | 0.75 (0.13) |
| Piankatank River | 15.7 | 37 | 27.0 | 54.1 | 18.9 | 0.89 (0.16) |
| Mobjack Bay | 19.4 | 75 | 30.7 | 29.3 | 40.0 | 0.40 (0.08) |
| York River | 19.2 | 58 | 37.9 | 31.0 | 31.0 | 0.52 (0.12) |
| Poquoson River | 19.6 | 47 | 27.7 | 31.9 | 40.4 | 0.43 (0.10) |
| Elizabeth River | 17.8 | 36 | 27.8 | 47.2 | 25.0 | 0.69 (0.14) |
| Lynnhaven River | 17.8 | 30 | 0.0 | 50.0 | 50.0 | 0.90 (0.19) |
| High Salinity | 487 | 27.1 | 33.1 | 39.8 | 0.51 (0.04) | |
| Low Salinity (<5 ppt) | ||||||
| Rappahannock River | 4.1 | 33 | 0.0 | 63.6 | 36.4 | 1.31 (0.19) |
| James River | 0.2 | 51 | 5.9 | 66.7 | 27.5 | 1.39 (0.33) |
| Low Salinity | 84 | 3.6 | 65.5 | 31.0 | 1.36 (0.12) | |
Osprey breeding outcomes and reproductive performance for sample sites within the Chesapeake Bay (2024).
Reproductive rate is the mean number of young fledged per pair. Salinity (ppt) represents mean surface salinity during 1985-2018 () at the midpoint of each survey route. Bolded numbers represent overall percentages and means fro high and low salinity study areas.
Figure 2
Table 2
| Site | Pairs | Clutch size mean (SE) | Eggs hatched (%) | Young loss rate (SE) | 1-young broods (%) | Survival probability (SE) |
|---|---|---|---|---|---|---|
| High Salinity (>10 ppt) | ||||||
| Choptank River | 60 | 2.7 (0.09) | 38.0 | 0.8 (0.14) | 72.7 | 0.23 (0.07) |
| Patuxent River | 49 | 2.3 (0.11) | 80.0 | 1.2 (0.18) | 58.8 | 0.44 (0.10) |
| Fleets Bay | 38 | 2.3 (0.14) | 81.8 | 1.2 (0.58) | 100.0 | 0.50 (0.22) |
| Eastern Shore | 57 | 2.7 (0.09) | 86.8 | 1.0 (0.22) | 44.0 | 0.52 (0.09) |
| Piankatank River | 37 | 2.4 (0.13) | 90.0 | 0.7 (0.17) | 45.0 | 0.68 (0.08) |
| Mobjack Bay | 75 | 2.4 (0.10) | 83.6 | 1.1 (0.15) | 68.2 | 0.40 (0.07) |
| York River | 58 | 2.5 (0.10) | 83.4 | 1.1 (0.19) | 50.0 | 0.50 (0.08) |
| Poquoson River | 47 | 2.3 (0.12) | 79.1 | 1.1 (0.18) | 66.6 | 0.46 (0.09) |
| Elizabeth River | 36 | 2.6 (0.14) | 91.1 | 1.4 (0.23) | 52.9 | 0.47 (0.08) |
| Lynnhaven River | 30 | 2.9 (0.11) | 81.3 | 1.5 (0.29) | 33.3 | 0.44 (0.10) |
| High Salinity | 487 | 2.5 (0.04) | 79.7 | 1.1 (0.06) | 54.6 | 0.45 (0.03) |
| Low Salinity (<5 ppt) | ||||||
| Rappahannock River | 33 | 2.6 (0.10) | 81.4 | 0.4 (0.17) | 14.3 | 0.84 (0.07) |
| James River | 51 | 2.7 (0.35) | 69.7 | 0.3 (0.09) | 20.6 | 0.83 (0.06) |
| Low Salinity | 84 | 2.7 (0.07) | 73.9 | 0.3 (0.08) | 17.4 | 0.84 (0.04) |
Osprey reproductive performance for sample sites within the Chesapeake Bay (2024).
Young loss rate is the mean number of young lost per pair between hatching and fledging. Bolded numbers represent overall percentages and means fro high and low salinity study areas.
The difference in breeding performance between high and low salinity areas was primarily driven by brood reduction (Table 2). Brood reduction was widespread throughout high salinity areas with 64.1% (152 of 237) of pairs losing at least one young after hatching compared to 25.0% (14 of 56) of pairs in low salinity areas (G-statistic = 11.8, df = 1, p < 0.001). The proportion of young lost between hatching and fledging ranged from 33.3% to 80.0% (eight of the ten study areas above 50%) compared to 15.2% and 17.0% for the two low salinity sites, respectively. The number of young lost per pair was greater in high salinity sites (1.01 ± 0.11; universal mean) compared to low salinity (0.32 ± 0.08) (β = 1.14 ± 0.26, p < 0.001). The likelihood of survival was inversely related to initial brood size, such that survival was lower with increasing brood size (β = -0.51 ± 0.17, p = 0.003). Overall mean survival of nestlings was less in high salinity (0.47 ± 0.03; universal mean) compared to low salinity (0.84 ± 0.07) sites (β = -2.61 ± 0.53, p < 0.001) and this difference was consistent across all brood sizes (Figure 3). The result of lost young is that brood size was lower in high salinity (1.51 ± 0.08) compared to low salinity (2.40 ± 0.09) sites (β = -0.11 ± 0.03, p = 0.002). One-chick broods were more common within high salinity (54.6%: 89 of 163) compared to low salinity sites (17.4%: 25 of 143) (G-statistic = 35.3, df = 1, p < 0.001).
Figure 3
3.2 2024 vs 1980s comparison
Reproductive performance was lower in 2024 (0.46 ± 0.06) compared to the 1980s (1.57 ± 0.13) for the four study areas examined (β = -1.24 ± 0.11, p < 0.001) (Table 3). The decline in reproductive metrics was consistent among sites, though variation was observed with the largest differences being between the site with the lowest productivity (i.e., Fleets Bay), and both the lower York and Piankatank Rivers. The likelihood that a pair would not lay a clutch was higher in 2024 (41.8%: 87 of 208) compared to the 1980s (1.8%; 6 of 348) (G-statistic = 164.6, p < 0.001). For pairs that did attempt breeding (eggs or young observed), the likelihood of failure was greater during 2024 compared to the 1980s for all study areas (all G-statistics > 19.0, p < 0.001) (Table 4). Combined, the likelihood of failure was 56.2% in 2024 compared to 23.8% during the 1980s (G-statistic = 58.4, p < 0.001). Mean clutch size was smaller in 2024 (2.41 ± 0.07) compared to the 1980s (2.88 ± 0.07) (β = -0.18 ± 0.03, p < 0.001). The difference in breeding performance between the two time periods was mostly a reflection of young lost after hatching. Mean number of young lost per pair was greater in 2024 (1.00 ± 0.20) compared to the 1980s (0.32 ± 0.07) (β = 1.12 ± 0.17, p < 0.001). Mean survival of hatched young was significantly (β = -2.61 ± 0.53, p < 0.001) higher during the 1980s (0.84 ± 0.04) compared to 2024 (0.47 ± 0.16). Mean brood size was lower (β = -0.32 ± 0.06, p < 0.001) in 2024 (1.49 ± 0.10) compared to the 1980s (2.06 ± 0.09).
Table 3
| Site | York river | Mobjack bay | Piankatank river | Fleets bay | Total |
|---|---|---|---|---|---|
| 1985-1988 | |||||
| Pairs | 57 | 131 | 80 | 80 | 348 |
| Reproductive rate (young/pr) | 1.95 (0.17) | 1.42 (0.10) | 1.75 (0.14) | 1.27 (0.12) | 1.57 (0.13) |
| Pairs not laying (%) | 0.0 | 0.0 | 6.3 | 1.3 | 1.8 |
| Successful pairs (%) | 86.0 | 72.5 | 77.7 | 70.9 | 76.9 |
| Failed pairs (%) | 14.0 | 27.5 | 16.0 | 27.8 | 23.8 |
| Clutch size | 2.99 (0.08) | 2.86 (0.05) | 2.90 (0.07) | 2.76 (0.07) | 2.88 (0.07) |
| Young lost (%) | 3.4 | 22.4 | 2.4 | 17.2 | 13.2 |
| Young loss rate (young/pr) | 0.29 (0.06) | 0.43 (0.07) | 0.20 (0.05) | 0.42 (0.09) | 0.32 (0.07) |
| 1-chick broods (%) | 20.4 | 30.5 | 22.6 | 29.8 | 26.6 |
| Brood size | 2.32 (0.11) | 1.95 (0.07) | 2.05 (0.09) | 1.94 (0.10) | 2.06 (0.09) |
| 2024 | |||||
| Pairs | 58 | 75 | 37 | 38 | 208 |
| Reproductive rate (young/pr) | 0.57 (0.07) | 0.41 (0.05) | 0.51 (0.06) | 0.37 (0.05) | 0.46 (0.06) |
| Pairs not laying (%) | 37.9 | 30.7 | 27.0 | 57.9 | 41.8 |
| Successful pairs (%) | 31.0 | 29.3 | 54.1 | 7.9 | 30.2 |
| Failed pairs (%) | 31.1 | 40.0 | 18.9 | 34.2 | 56.2 |
| Clutch size | 2.50 (0.07) | 2.40 (0.06) | 2.90 (0.07) | 2.31 (0.08) | 2.41 (0.07) |
| Young lost (%) | 50.0 | 60.0 | 33.3 | 66.7 | 50.1 |
| Young loss rate (young/pr) | 0.89 (0.16) | 1.33 (0.18) | 0.60 (0.14) | 1.29 (0.32) | 1.00 (0.20) |
| 1-chick broods (%) | 50.0 | 68.2 | 45.0 | 100.0 | 57.1 |
| Brood size | 1.68(0.11) | 1.41 (0.09) | 1.48 (0.10) | 1.40 (0.11) | 1.49 (0.10) |
Osprey breeding performance (± SE) during the 1980s (1985, 1986, 1987, 1988) and 2024 within four high salinity (>10 ppt) study areas in the Chesapeake Bay.
Table 4
| Metric | Site | Estimate | SE | P-value |
|---|---|---|---|---|
| Clutch Size | Mobjack | 0.04 | 0.03 | 0.216 |
| Piankatank River | 0.05 | 0.03 | 0.132 | |
| York River | 0.08 | 0.04 | 0.023 | |
| Brood Size | Mobjack | 0.00 | 0.06 | 0.971 |
| Piankatank River | 0.06 | 0.07 | 0.971 | |
| York River | 0.18 | 0.07 | 0.008 | |
| Reproductive Rate | Mobjack | 0.12 | 0.12 | 0.319 |
| Piankatank River | 0.32 | 0.12 | 0.009 | |
| York River | 0.43 | 0.13 | 0.001 | |
| Chick Loss | Mobjack | 0.03 | 0.25 | 0.898 |
| Piankatank River | –0.77 | 0.33 | 0.019 | |
| York River | –0.37 | 0.29 | 0.195 | |
| Survival | Mobjack | 0.44 | 0.34 | 0.188 |
| Piankatank River | 1.61 | 0.44 | <0.001 | |
| York River | 1.00 | 0.39 | 0.010 |
Site-specific coefficient estimates from generalized linear models (GLMs) comparing 2024 to the 1980s for each reproductive index.
Fleets Bay served as the reference site. Estimates represent the difference from Fleets Bay within each model. P-values for sites that differed from Fleets Bay (p < 0.05) are shown in bold.
4 Discussion
Recent observations of low reproductive rates for ospreys within Mobjack Bay (; ; Watts et al., 2024) do not appear to be an anomaly. None of the ten study areas within the high salinity waters reached the demographic threshold (1.15 young/pair) believed to be required for population maintenance within the Chesapeake Bay (; ; Watts et al., 2024). Only two of the study areas (Lynnhaven River: 0.90, Piankatank River: 0.89) reached the lower reproductive threshold (0.80 young/pair) extrapolated by . These results suggest that ospreys throughout the main stem of the Chesapeake Bay are currently experiencing a demographic deficit that threatens population stability. In contrast, reproductive rates for both study areas within low salinity waters were above reproductive thresholds for population maintenance. In the 2024 breeding season, reproductive rates within high salinity sites were similar to or below those recorded (0.46 to 0.78 young/pair) within the Chesapeake Bay during the height of the DDT era of the 1960s (; ; Wiemeyer, 1971; ).
All of the breeding metrics associated with food stress [percentage of pairs documented to make breeding attempts, mean clutch size, percentage of pairs failing, frequency of pairs experiencing brood reduction and the mean number of young lost between hatching and fledging (brood reduction) suggest that high and low salinity areas differ in their level of food stress. These results are consistent with work by who compared osprey breeding performance between high and low salinity sites within the lower Chesapeake Bay during the 2006 and 2007 seasons. In addition to the metrics recorded during this study, Glass showed that provisioning rates (number of fish, mass of fish, energy content of diet) and both the growth rate of young and asymptotic (fledging) weight of young were greater in low compared to high salinity areas. These metrics support the observation that food stress in high salinity areas is greater than in low salinity areas.
Breeding performance within high salinity study areas with comparable data was greater during the 1980s than recorded during 2024. During the 1980s, reproductive rates were above target maintenance levels for all four study areas examined. This result is consistent with the forty-year retrospective within the Mobjack Bay study area (Watts et al., 2024) and suggests that the decline in productivity is more widespread than previously documented. Differences in productivity between 2024 and the 1980s were comparable to the findings between high and low salinity areas in 2024. All of the breeding metrics associated with food stress declined between the 1980s and 2024. Both low salinity sites in 2024 and high salinity sites in the 1980s exhibit little indication of food stress.
A large percentage (27.1) of pairs within high salinity study sites were not documented to make breeding attempts in 2024 in contrast to 3.6% of pairs in low salinity study sites and 1.8% of pairs in high salinity sites during the 1980s. Nonbreeding pairs arrived from wintering grounds during the normal time (late February – early March), remained resident throughout the nesting season and defended territories. Unlike the other metrics that have been associated with food stress, the documentation of such a large number of pairs that did not attempt breeding is novel. The influence of food limitation on breeding is well-studied in birds (; ). One common finding in food supplementation studies has been that females provided additional food lay clutches earlier than controls, suggesting that insufficient food may constrain a female’s ability to reach breeding condition. observed that clutch initiation dates were significantly earlier (mean difference was 5 days in 2006 and 12 days in 2007) in low salinity sites compared to high salinity sites. Foregoing an entire breeding season was not anticipated and may suggest severe food limitation. Additional investigation and experimentation may help to clarify the role of food in this behavior ().
Food stress and subsequent brood reduction in ospreys has been well documented. monitored the osprey population on Hyatt Lake in Oregon before and after treatment with Rotenone (used to eliminate the fish community) and restocking. Treatment resulted in a reduction in fish abundance, osprey provisioning rates, osprey clutch size and osprey productivity and an increase in osprey brood reduction. documented a 58% decline in the osprey breeding population within Florida Bay and suggest that this was driven by reduced fish abundance. This contention was supported by documentation of low provisioning rates, low productivity, sibling aggression and brood reduction (; ). found that brood reduction and low productivity were associated with low provisioning rates for ospreys on lakes of southwest Sweden. working within four study sites in Montana (one segment of river and three reservoirs) found that the site with the lowest provisioning rate (grams of fish/hour) had the lowest productivity. found that a decline in cutthroat trout (Oncorhynchus clarkii) in Yellowstone Lake was coincident with a decline in productivity, an increase in osprey foraging outside the lake, and a population decline on the lake. Responses by breeding ospreys to declines in primary prey species within other study areas are consistent with observations within high salinity areas of the Chesapeake Bay during 2024.
4.1 Potential mechanisms driving poor reproductive performance of ospreys
Several documented and suspected factors other than food stress could be contributing to nest failures within the Chesapeake Bay including stressors that influence clutch/brood survival (nest predators, weather, contaminants) and those that may disrupt breeding (nest competitors, disease) (; ; Watts and Paxton, 2007; ). The role of mammals in nest failure has declined over the past several decades as nesting ospreys have shifted substrate use (; ; Watts and Paxton, 2007). Prior to the 1950s nearly all osprey nests were built on dead and live trees. By the 1990s overwater structures (e.g., navigational aids, duck blinds, osprey platforms) accounted for more than 90% of nesting substrates (). A shift to overwater structures has been associated with a decline in mammal predation (). During 2024 we documented (using nest cameras) two (0.4%) nests predated by mammals including a three-egg clutch taken by a raccoon (Procyon lotor) and one egg and two young just after hatching taken by an American mink (Neogale vison). Fish crows (Corvus ossifragus) have increased significantly throughout the Bay region over the past fifty years () and specialize on bird eggs and broods (). Fish crows were common to abundant within most study areas during the 2024 nesting season and may have accounted for the disappearance of some eggs and small young. Under normal circumstances, crows have not been considered a threat to nesting osprey (). We observed two instances of crow predation on clutches during the 2024 season (Watts et al., 2024, unpublished data). Both occurrences followed females leaving nests unguarded after receiving no fish from the male for more than twenty-four hours, suggesting an interaction between food stress and vulnerability to crows. Great horned owls (Bubo virginianus) are well-known predators of osprey broods () and occasionally kill adults (). This nocturnal predator has been documented to take osprey broods within the Chesapeake Bay () and within some of the study areas used in 2024 (Watts et al., 2024, unpublished data). The influence of owls on nesting osprey is generally limited to their nesting territory, and although they may take broods in successive years from the same nest, their impact is typically local rather than widespread. The most common pattern of brood loss observed in 2024 was also not consistent with owl predation. Owls generally take an entire brood over a short period () compared to brood reduction where young are lost sequentially over a longer period. We also have no reason to suspect that owl predation would be more acute within high compared to low salinity study areas.
The bald eagle (Haliaeetus leucocephalus) population within the Chesapeake Bay has recovered since the 1960s with a high average annual rate of increase (8-12%; Watts et al., 2007; ). As eagle populations have recovered throughout their range, researchers have documented adverse effects on other waterbirds (e.g., ; ; ), including osprey (; ). observed that osprey nesting within Florida Bay temporarily relocated to other keys following colonization by bald eagles and attributed these responses to aggressive territorial behavior when eagles had eggs or young. evaluated the potential influence of factors associated with a decline (>20 to <5 pairs) in nesting osprey within Voyageurs National Park (1988-2012). Several bottom-up factors (e.g., fish abundance, weather, foraging habitat) and one primary top-down factor (bald eagle nests) were evaluated. The increase in bald eagles was the best predictor of declines in nesting osprey and their productivity, though the specific mechanisms of the interaction were not assessed. However, there was no indication of competition for prey suggesting some alternative mechanism (e.g., territoriality, predation). Bald eagles are known osprey brood predators () and have been documented to take both eggs and broods within the Chesapeake Bay (Watts, Center for Conservation Biology, W&M, 2021-2024, unpublished data). No studies have examined the impact of brood predation by eagles on osprey reproductive rates. We deployed at least four motion-triggered cameras on nests within all study areas (N > 50 nests monitored with cameras) (Watts, Center for Conservation Biology, W&M, 2021-2024, unpublished data). Cameras captured more than one million images during the nesting season. Bald eagles did not appear in a single image suggesting that eagle predation is relatively rare. This is consistent with observations made during fieldwork in 2024 and over past decades. We estimate that high salinity waters of the Chesapeake Bay support more than 5, 000 osprey pairs suggesting that pairs would have lost in the range of 5, 500 young during the 2024 season. Given the infrequency of bald eagle predation events, consuming this number of young is implausible. If brood predation reflects eagle density, then we would expect ospreys within low salinity waters to suffer the greatest loss to predation. Bald eagle population recovery within the Chesapeake Bay has varied with salinity (Watts et al., 2006) such that by 2002 eagle breeding density was four-fold greater within low salinity compared to high salinity waters. The divergence in density has continued such that by 2016, low salinity waters supported fifteen-fold greater bald eagle density. Despite dramatically greater eagle densities, osprey productivity within low salinity waters was more than double that recorded within high salinity waters. Bald eagles are well-known kleptoparasites on many waterbird species (), other eagles () and ospreys (). Prevost made detailed observations of eagle interactions with foraging ospreys in Nova Scotia and found that the consequences of food-robbing were minimal. Of 1, 793 fish captures made by ospreys, only two were taken by eagles which is less than the number of fish dropped accidentally. Thus, evidence does not support the notion that bald eagles are the primary driver of poor osprey breeding performance in 2024.
Adverse weather has been shown to affect osprey breeding performance in many populations (; ; ) and within the Chesapeake Bay (). Osprey broods within the first two weeks after hatching are particularly vulnerable to exposure during extended periods of cool temperatures and rain. However, working in Idaho and Washington examined the frequency of rainstorms during the nesting season and found that years with late storms that included >1.3 cm in rainfall adversely affected productivity presumably due to impaired foraging conditions. This result is in contrast to who found that weather conditions had no influence on hunting behavior or foraging success in northwestern Washington. Within the Chesapeake Bay, weather was responsible for 5% of nest failures (N = 1, 761) between 1966 and 1979 (, ). Wind and rain were responsible for 50% of weather-related egg losses and 66% of nestling losses (). Impacts of weather may be episodic. In 1972, tropical storm Agnes killed 18% of nestlings produced in the Chesapeake Bay. mentioned several nestling ospreys apparently suffering from heat stress in the Chesapeake Bay. Periods of extreme heat are common throughout the Bay region during the late summer. During the 2024 nesting season the Chesapeake Bay did not experience exceptional weather events that would be expected to drive broad-scale brood failure.
Several other bird species utilize osprey nests for breeding and may usurp them from osprey pairs (). Most of these species are resident and initiate nesting before osprey return to breeding grounds. Within the Chesapeake Bay, these species include bald eagles, red-tailed hawks (Buteo jamaicensis), great horned owls, peregrine falcons (Falco peregrinus), great blue herons (Ardea herodias), great black-backed gulls (Larus marinus), and Canada geese (Branta canadensis) (Watts et al., 2024, unpublished data). Only gulls and geese used osprey nests within study areas during 2024. A single nest was used by a pair of great black-backed gulls and only late in the season after the osprey pair had abandoned the nest. Canada geese used osprey nests within all study areas (e.g., ). The resident Canada goose population has experienced a 15% annual rate of increase since the 1970s (; ). Geese now usurp osprey nests throughout much of North America (e.g., ; ; ). Although geese may disrupt breeding for some pairs, their influence at the population level is not clear. Canada geese often initiate nesting before osprey return and early nesting geese may conclude nesting in time for osprey to use the same nest leading to no disruption. Later nesting pairs of geese may disrupt osprey nesting. Disruption typically occurs during the first years after colonization by geese when ospreys are delayed as they build new nests. Pairs that build alternate nests in the first year often do not experience disruptions in following years. Pairs with high site fidelity may experience disruptions for several years. We documented 84 geese using osprey nests within our study area representing 10% of nest structures examined during the initial establishment of study areas. Because our focus was osprey pairs, goose nests were not included in our samples and so had no direct influence on reproductive rates. However, it is possible that pairs displaced by geese were included in our sample and may have suffered breeding delays or failure to nest. Within study areas that we have monitored for years, goose nests have been observed for some time and osprey pairs have moved on to nest on other structures. Because several study areas were new in 2024, we are unable to fully evaluate potential disruption by geese. However, given that we did not include goose nests in our sample, we believe that the impact of recently disrupted nests on our broader results is likely minimal.
Due to their position as a high trophic level predator, ospreys will always be exposed and potentially vulnerable to contaminants that are introduced into the aquatic food web. Ospreys have been subjected to a wide range of contaminants throughout their North American range () and within the Chesapeake Bay (Wiemeyer, 1971; ; ; ). Legacy pesticides and metabolites, including p, p’-DDE associated with eggshell thinning, and other persistent organic pollutants (e.g., polychlorinated biphenyls) that were detected in both adult ospreys and their eggs (Wiemeyer et al., 1987) have declined over the decades, and are not currently believed to suppress reproduction (). Concentrations of heavy metals continue to be greater within industrialized portions of the Bay when compared to more rural areas but are not believed to be affecting reproduction at current levels (; ). Other emerging and more contemporary contaminants (e.g., polybrominated diphenyl ether flame retardants, perfluorinated compounds) and pharmaceuticals that move through the food chain have been detected in ospreys but have not been documented to affect health or reproduction (, ). Reproductive rates recorded during the 2024 nesting season are below those reported during the height of the DDT era. In contrast to the 1970s (Wiemeyer, 1971; Wiemeyer et al., 1987, Wiemeyer et al., 1988) when poor reproductive success was due to impaired hatching success, the low productivity recorded during 2024 was principally due to brood reduction.
Numerous pathogens and parasites have been identified in ospreys breeding in the Chesapeake Bay (). Most of these do not have the potential to cause population-level effects. In 1994, an outbreak of avian cholera (Pasteurella multocida) occurred in the Chesapeake Bay and killed an estimated 36, 700 birds of 57 species, including ospreys (). West Nile virus reached the Western Hemisphere in 1999 and has affected many bird populations (). Ospreys were not listed among 22 raptor species examined in western North America that tested positive for West Nile () and had one of the lowest positivity rates among raptors tested in Georgia (). The emergence of the most recent strain (H5N1, clade 2.3.4.4b) of highly pathogenic avian influenza (HPAI) in 2021 has spread rapidly on a global scale and has caused very high mortality in some bird groups (; ). Vulnerability to avian influenza appears to vary between ecological groups (). There has been no evidence that HPAI is affecting osprey populations based on a surveillance program in New York and Minnesota with no positive cases involving ospreys (N = 15 ospreys; ; ). From 2020 to 2025, there were 210 cases of HPAI in wild birds of Maryland and Virginia (33 involved bald eagles), but none documented infection in ospreys (). It is noteworthy that the Virginia Department of Wildlife Resources submitted three dead ospreys collected from the Chesapeake Bay in August 2024 to the University of Georgia College of Veterinary Medicine for analysis, and these tested negative for both HPAI and West Nile virus (). In addition, accession records were obtained for 159 ospreys admitted to five rehabilitation facilities in the vicinity of the Chesapeake Bay in 2024 [written communication from: Leigh-Ann Horne, Wildlife Center of Virginia, Waynesboro, VA on January 12, 2025 (Wildlife Center of Virginia, 2024); Malia Hale, Owl Moon Raptor Center, Boyds, MD on December 14, 2024; Andrea Howey-Newcomb, Tri-State Bird Rescue & Research, Inc., Newark, DE on December 13, 2024; Gay Frazee, Wildlife ER, Jamesville, VA on December 30, 2024; Jen Riley, Blue Ridge Wildlife Center, Boyce, VA on December 16, 2024). In these records, no infectious diseases were diagnosed. Notably, 53 of the 159 accession records provide a diagnosis of “emaciated”, “debilitated”, “thin” and/or “starvation”. In addition, recurring harmful algal blooms have affected large numbers of waterbirds in portions of the Chesapeake Bay (), but have not been linked to mortality events in ospreys (). There is no indication that disease affected osprey reproductive rates during the 2024 breeding season or had an influence on the observed disparity between high and low salinity sites.
4.2 Potential role of menhaden for nesting ospreys in high salinity portions of the Chesapeake
Nesting ospreys within the higher salinity reaches of the Chesapeake Bay are believed to be menhaden-dependent (; ; ; Watts et al., 2024). Menhaden are a schooling fish with a very high energy density making them efficient to capture and ideal for brood-rearing. Broad patterns in food-stress metrics and reproductive deficits in 2024 suggest that menhaden abundance was inadequate to support a stable population. This finding is consistent with previous years (; ; Watts et al., 2024). Observations for ospreys are consistent with trends reported for the commercial menhaden bait harvest (). Landings in 2024 (313, 921 kg) were the lowest in 61 years, were only 34% of the previous low set in 2023 (930, 326 kg), and were less than 10% of the long-term average (3, 188, 005 kg). There is also the possibility of a food web cascade effect related to inadequate menhaden abundance. Notably, many species of fish in the Chesapeake Bay are dependent on menhaden as prey (; ; ), with striped bass (Morone saxatilis) being one such species that are commonly captured and consumed by reproducing osprey ().
The Atlantic menhaden is the focus of the largest fishery on the East Coast and the stock is managed by the Atlantic States Marine Fisheries Commission (ASMFC). The ASMFC considers Atlantic menhaden along the entire Atlantic Coast (Nova Scotia to Florida) to be composed of a single stock and classifies current harvest levels to be sustainable (; ). However, treatment of the fisheries as a single stock masks spatial and temporal patterns of abundance. Ospreys provision young with menhaden that typically range from 10 to 25 cm (19.1 ± 0.37, N = 253; Watts et al., 2024, unpublished observations). This size range suggests that ospreys are primarily using menhaden in the 2 to 4-year age classes (). We currently have no fisheries-independent monitoring of adult menhaden within Chesapeake Bay. This critical information gap prevents a direct assessment of the relationship between osprey reproduction and menhaden abundance on a local scale.
The Atlantic menhaden has a complex natural history (). The species spawns in ocean waters along the continental shelf and is a filter feeder throughout its entire life cycle. Several factors have been suggested to influence the abundance of menhaden within the Chesapeake Bay including sources of mortality such as commercial harvest, predation and competition and factors influencing recruitment (; ; ). Menhaden have been harvested commercially for more than 200 years. Harvest policy has evolved in recent decades from no management to maximum-sustainable-yield to the development of models designed to evaluate the tradeoffs between commercial take and ecosystem services (). Using an ad-hoc approach, the ASMFC board reallocated menhaden to ecosystem services in 2017 and reduced the harvest cap in the Chesapeake Bay to 51, 000 mt. The impact of harvest on the abundance of adult menhaden within the Chesapeake Bay remains controversial primarily because we do not have the abundance data required to evaluate the relationship between harvest and menhaden abundance (Watts and Hines, 2025). Many predators including fish (; ), birds (; Watts et al., 2023) and marine mammals () depend on menhaden as prey.
Estimated consumption by fish predators alone is comparable to the current commercial harvest (). Because menhaden are pelagic planktivores that occur in large schools, the availability of plankton may exert a controlling influence on abundance via inter- and intra-specific competition. Although these have not been demonstrated they remain a likely outcome during times of plankton scarcity. Menhaden larvae depend on transport currents to reach nearshore nursery areas and are sensitive to temperature, salinity, water quality and other local environmental factors (). Within Chesapeake Bay, early season rain and freshwater discharge from major tributaries appears to have a bottom-up influence on phytoplankton blooms, zooplankton availability, and the concentration of juvenile menhaden (). Spatial and temporal variation in zooplankton abundance has an influence on adult menhaden (). The Chesapeake Bay is predicted to experience a wide range of changes related to climate shifts over the next few decades, many of which are directly relevant to menhaden abundance (). Most of the factors posited to influence menhaden abundance are beyond management control. One of the primary factors that is subject to management control is commercial harvest.
4.3 Potential for density-dependent effects on osprey reproduction
Following the DDT lows, the Chesapeake Bay osprey population has grown exponentially increasing from 1, 450 to 3, 473 pairs between 1973 and 1995 (tdouble = 18.1 years) and from 3, 473 pairs to an estimated 10, 000 pairs between 1995 and 2024 (tdouble = 19.1 years) (, ; Watts and Paxton, 2007; Watts et al., 2024, unpublished data). Given this large increase in population size, it is possible that osprey pairs are now experiencing density-dependent processes that may impact reproductive rates. One common ecological constraint that may arise from increasing density is resource (nesting substrate and food) depletion. The number of substrates suitable for nesting is known to limit the number of pairs nesting in a given area and so may impose a ceiling on population size (e.g., ; ). However, substrate limitation alone would not influence reproductive rates. There is also little evidence that increasing osprey populations deplete prey. The metabolic demand of the osprey population in the Chesapeake Bay is not on a scale that would be expected to exert control over or deplete menhaden. Ospreys are a small consumer within the context of the broader Chesapeake Bay ecosystem where populations of predatory fish consume the equivalent of commercial take (; ). Even during the period when menhaden accounted for 75% of the osprey diet, the population would have consumed only 0.004% of the commercial landings ().
A second density-dependent process that may impact reproductive success is behavioral interference. Ospreys are a social species that often nest in dense aggregations referred to as “colonies” () where pairs benefit from nesting near other pairs by gaining increased synchrony and protection from nest predators (), getting information about prey location (), and reducing foraging time through local enhancement (). Despite the benefits of nesting in dense aggregations, it is possible that population growth may lead to behavioral interference. working in Corsica (1974-2004) found that despite an increase in provisioning rate with breeding density, productivity and fledging success exhibited negative density-dependent effects. They found no influence on clutch size or condition of young at fledging. The authors attributed the reduced productivity to an increase in territorial intrusions and behavioral interference by nonbreeding (floater) ospreys. Interference may have changed the time budgets of breeding adults though the increase in provisioning rate does not suggest this mechanism. Investigations of reproductive rates within other recovering populations in Oregon (Witt, 1996), France () and Germany () found no density-dependent effects. Population recovery throughout the Chesapeake Bay has been spatially variable (), with growth rates in low salinity reaches ranging from two to ten-fold higher compared to high salinity reaches. Due to the high growth rate within reference sites, nesting density is comparable to high salinity sites suggesting that density is unlikely to be the primary driver of observed differences. Nesting densities are high within all study areas and territorial intrusions are regular events. During trapping activities through the years, it has been reasonably common to trap both adults and have a third adult come into the nest to incubate. Given that nesting densities have been high throughout the study areas for a long period of time, behavioral interference does not appear to explain recent changes in reproductive rates.
5 Conclusions
Ospreys are well-known sentinels for aquatic ecosystems both as indicators of environmental contaminants and as metrics for fish stocks. The Chesapeake Bay supports a large osprey population that has been responsive to previous changes in ecosystem conditions (Watts and Paxton, 2007; , ). During the 2024 breeding season, osprey productivity within high salinity reaches was well below the level required for population maintenance. Although several factors are operating within this ecosystem that may affect productivity, we suggest that Atlantic menhaden availability is the primary driver of low reproduction. This conclusion is consistent with both a 40-year retrospective (Watts et al., 2024) and response to an experimental manipulation of menhaden provisioning (). As suggested by food stress and associated reproductive failure in ospreys are reversible if fish stocks are managed to fulfill their role within the ecosystem.
A large degree of uncertainty remains about the status of adult menhaden within the Chesapeake Bay. Although the ASMFC has indicated that the stock along the Atlantic Coast has not been overfished (), the status and trends of adult menhaden within the Chesapeake Bay have not been evaluated. To more fully resolve this uncertainty, assessments of menhaden abundance could enhance understanding of their role within the ecosystem and support management of consumers that depend on them.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Ethics statement
The animal study was approved by Institutional Animal Care and Use Committee of the US Geological Survey, Eastern Ecological Science Center. The study was conducted in accordance with the local legislation and institutional requirements.
Author contributions
BW: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing. CH: Formal analysis, Investigation, Writing – review & editing, Visualization. MB: Writing – review & editing. RL: Investigation, Writing – review & editing. BP: Investigation, Writing – review & editing. LD: Investigation, Supervision, Writing – review & editing, Visualization. KS: Investigation, Writing – review & editing. GK: Investigation, Writing – review & editing. DD: Investigation, Methodology, Writing – review & editing. BR: Conceptualization, Methodology, Formal analysis, Funding acquisition, Supervision, Validation, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. Funding was provided by the Center for Conservation Biology at William & Mary, the U.S. Geological Survey, the Virginia Aquarium and Marine Science Center and the Virginia Osprey Foundation. The Center for Conservation Biology supported surveys on the Patuxent River, Fleets Bay, Eastern Shore, Piankatank River, Mobjack Bay, York River, Poquoson River and the Elizabeth River. USGS supported surveys on the Choptank River. The Virginia Aquarium supported surveys on the Lynnhaven River. Financial and logistical support for this project was provided by the Center for Conservation Biology, the U.S. Geological Survey Chesapeake Bay studies, the Virginia Aquarium and Maryland National-Capital Park and Planning Commission Patuxent River Park.
Acknowledgments
We thank M. Pitts, M. Watts, L. Sloan, L. Arthur, S. Graves, C. Shaw, R. Kellam, A. Pellegrini, J. Thompson, M. Fender, C. Equels, and M. Academia for assistance in the field and F. Rattner for obtaining information on ospreys admitted to rehabilitation facilities. Dr. Charles Henny graciously provided constructive comments on a draft of this manuscript. We thank the many landowners who allowed us to work with their private platforms and osprey pairs. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The handling editor RL declared a shared affiliation with the author BW at the time of review.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
AcademiaM. H. (2022). Food supplementation increases performance of ospreys in the lower Chesapeake Bay (Williamsburg, Virginia, USA: College of William and Mary). M.A. Thesis.
2
AcademiaM. H.WattsB. D. (2023). Food supplementation increases reproductive performance of ospreys in the lower Chesapeake Bay. Fron. Mar. Sci.10, 1172787. doi: 10.3389/fmars.2023.1172787
3
AhrenholzD. W.. (1991). Population biology and life history of the North American menhadens, Brevoortia spp. Mar Fish Rev53, 3–19. Available online at: https://spo.nmfs.noaa.gov/sites/default/files/pdf-content/mfr531.pdf.
4
AhrenholzD. W.NelsonW. R.EpperlyS. P. (1987). Population and fishery characteristics of Atlantic menhaden. Brevoortia tyrannus. Fish. Bull.85, 569–600.
5
AirolaD. A. (1987). Nest site competition between ospreys and Canada geese at Lake Almanor, California. J. Rapt Res.21, 121–124. Available online at: https://digitalcommons.usf.edu/jrr/vol21/iss3/9
6
AmesP. L. (1966). DDT residues in the eggs of the osprey in the northeastern United States and their relation to nesting success. J. Appl. Ecol.3, 87–97. doi: 10.2307/2401447
7
AnsteadK. A.DrewK.ChagarisD.CieriM.SchuellerA. M.McNameeJ. E.et al. (2021). The path to an ecosystem approach for forage fish management: A case study of Atlantic menhaden. Front. Mar. Sci.8, 607657. doi: 10.3389/fmars.2021.607657
8
BowmanR.PowellG. V. N.HovisJ. A.KlineN. C.WilmersT. (1989). Variations in reproductive success between subpopulations of the osprey (Pandion halietus) in south Florida. Bull. Mar. Sci.44, 245–250.
9
BretagnolleV.MougeotF.ThibaultJ. C. (2008). Density dependence in a recovering osprey population: demographic and behavioral processes. J. Anim. Ecol.77, 998–1007. doi: 10.1111/j.1365-2656.2008.01418.x
10
BuchheisterA.MillerT. J.HoudeE. D. (2017). Evaluating ecosystem-based reference points for Atlantic Menhaden. Mar. Coast. Fish.9, 457–473. doi: 10.1080/19425120.2017.1360420
11
ByrdM. A. (1985). “ Osprey population studies,” in Virginia nongame and endangered wildlife investigationsVirginia Commission of Game and Endangered Wildlife Investigations ( Virginia Commission of Game and Inland Fisheries, Richmond, Virginia).
12
CanalD. V.MorandiniB.MartinT.LanggemachR.MurielM.de Lucaset al. (2018). Productivity is related to nest site protection and nesting substrate in a German osprey population. J. Ornith.159, 265–273. doi: 10.1007/s10336-017-1498-8
13
ChagarisD.DrewK.SchuellerA.CieriM.BritoJ.BuchheisterA. (2020). Ecological reference points for Atlantic menhaden established using and ecosystem model of intermediate complexity. Front. Mar. Sci.7, 1043. doi: 10.3389/fmars.2020.606417
14
Chesapeake Bay Program (2023). Chesapeake bay mean surface salinity 1985 – 2018. Available online at: https://gis.chesapeakebay.net/ags/rest/services/Modeling/salinity_surface/MapServer (Accessed May 5, 2026).
15
ClarkK. E.StansleyW.NilesL. J. (2001). Changes in contaminant levels in New Jersey osprey eggs and prey 1989–1998. Arch. Environ. Contam. Toxicol.40, 277–284. doi: 10.1007/s002440010173
16
ColdC. W. (1993). Adult male osprey killed at nest by great horned owl. Passenger Pigeon55, 269–270.
17
CruzJ.WindelsS. K.ThogmartinW. E.CrimminsS. M.GrimL. H.LarsonJ. H.et al. (2019). Top-down effects of repatriating bald eagles hinder jointly recovering competitors. J. Anim. Ecol88, 1054–1065. doi: 10.1111/1365-2656.12990
18
DolbeerR. A.SeubertJ. L.BegierM. J. (2014). Population trend of resident and migratory Canada geese in relation to strikes with civil aircraft. Hum.-Wildl. Interact.8, 88–99. doi: 10.26077/ea1k-ch43
19
ElliottJ. E.MachmerM. M.HennyC. J.WilsonL. K.NorstromR. 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. Tox.35, 620–631. doi: 10.1007/s002449900424
20
ElliottJ. E.WilsonL. K.HennyC. J.TrudeauS. F.LeightonF. A.KennedyS. W.et al. (2001). Assessment of biological effects of chlorinated hydrocarbons in osprey chicks. Environ. Tox. Chem.20, 866–879. doi: 10.1897/1551-5028(2001)020<0866:aobeoc>2.0.co;2
21
EllisA. E.MeadD. G.AllisonA. B.StallknechtD. E.HowerthE. W. (2007). Pathology and epidemiology of natural West Nile viral infection of raptors in Georgia. J. Wildl. Dis.43, 214–223. doi: 10.7589/0090-3558-43.2.214
22
ElliserC. R.EdisonC.MacIverK.RustL. B. (2022). Kleptoparasitic interactions by bald eagles (Haliaeetus leucocephalus) during marine mammal foraging events. Behaviour159, 1387–1399. doi: 10.1163/1568539x-bja10177
23
ErikssonM. O. G. (1986). Fish delivery, production of young, and nest density of osprey (Pandion haliaetus) in southwest Sweden. Can. J. Zool.64, 1961–1965. doi: 10.5040/9781666996258.ch-009
24
FieldsL. E.PagelJ. E. (2016). Osprey occupancy of Mono Lake – unique habitat in eastern California. J. Rapt. Res.50, 97–102. doi: 10.3356/rapt-50-01-97-102.1
25
FlemmingS. P.BancroftR. P. (1990). Bald eagle attacks osprey nestling. J. Rapt Res.24, 26–27.
26
FlemmingS. P.SmithP. J. C.SeymourN. J.BancroftR. P. (1992). Ospreys use local enhancement and flock foraging to locate prey. Auk109, 649–654. Available online at: https://digitalcommons.usf.edu/auk/vol109/iss3/29
27
ForysE. A.HindsleyP. R.BryanS. (2021). Predictors of osprey nest success in a highly urbanized environment. J. Rapt. Res.55, 485–495. doi: 10.3356/jrr-20-97
28
FriedlandK. D.LynchP. D.GoblerC. J. (2011). Time series mesoscale response of Atlantic menhaden Trevoortia tyrannus to variation in plankton abundance. J. Coast. Res.27, 1148–1158. doi: 10.2112/jcoastres-d-10-00171.1
29
Gamarra-ToledoV.PlazaP. I.GutierrezR.LuyoP.HernaniL.AnguloF.et al. (2023). Avian flu threatens neotropical birds. Science379, 246. doi: 10.1126/science.adg2271
30
GlassK. A. (2007). Broad spatial trends in osprey provisioning, reproductive success, and population growth within the lower Chesapeake Bay (Williamsburg, VA, USA: College of William and Mary).
31
GlassK. A.WattsB. D. (2009). Osprey diet composition and quality in high and low salinity areas of lower Chesapeake Bay. J. Rapt. Res.43, 27–36. doi: 10.3356/jrr-08-22.1
32
GoldenN. H.RattnerB. A. (2003). Ranking terrestrial vertebrate species for utility in biomonitoring and vulnerability to environmental contaminants. Rev. Environ. Contam. Toxicol.176, 67–136. doi: 10.1007/978-1-4899-7283-5_2
33
GreeneE. (1987). Individuals in an osprey colony discriminate between high and low quality information. Nature329, 239–241. doi: 10.1038/329239a0
34
GroveR. A.HennyC. J.KaiserJ. L. (2009). Osprey: worldwide sentinel species for Assessing and monitoring environmental contamination in rivers, lakes, reservoirs and estuaries. J. Toxicol. Environ. Health B. Crit. Rev.12, 25–44. doi: 10.1080/10937400802545078
35
GuntherA.KroneO.GlobigA.PohlmannA.KingJ.FastC.et al. (2024). Avian raptors are indicator species and victims of high pathogenicity avian influenza virus HPAIV H5N1 (clade 2.3.4.4b) in Germany. Sci. Rep.14, 28799. doi: 10.1038/s41598-024-79930-x
36
HaganJ. I.SalWaltersR. J. (1990). Foraging behavior, reproductive success, and colonial nesting in ospreys. Auk107, 506–521. doi: 10.1093/auk/107.3.506
37
HaganJ. M. (1986). Temporal patterns in pre-fledging survival and brood reduction in an osprey colony. Condor88, 200–205. doi: 10.2307/1368916
38
HallV.CardonaC.MendozaK.TorchettiM.LantzK.BuenaI.et al. (2024). Surveillance for highly pathogenic avian influenza A (H5N1) in a raptor rehabilitation center – 2022. PloS One19, e0299330. doi: 10.1371/journal.pone.0299330
39
HarmataP. J.RestaniM.HarmataA. R. (2007). Settlement patterns, foraging behavior, and reproductive success of Ospreys along a heterogeneous riverine corridor. Can. J. Zoo.85, 56–62. doi: 10.1139/z06-190
40
HartmanK. J.BrandtS. P. (1995). Predatory demand and impact of striped bass, bluefish, and weakfish in the Chesapeake Bay: applications of bioenergetic models. Can. J. Fish. Aqua. Sci.52, 1667–1687. doi: 10.1139/f95-760
41
HennyC. J. (1983). “ Distribution and abundance of ospreys in the United States,” in Biology and management of bald eagles and ospreys. Ed. BirdD. M. ( Harpell Press, Ste. Anne de Bellevue Quebec, CA), 175–186.
42
HennyC. J.CollinsJ. A.DeibertW. J. (1978). Osprey distribution, reproduction and status in western North America II: The Oregon population. Murrelet59, 14–25. doi: 10.2307/3534722
43
HennyC. J.GroveR. A.KaiserJ. L. (2008). Osprey distribution, reproductive success and contaminant burdens (OC pesticides, PCBs, PCDDs, PCDFs and mercury) along Lower Columbia River 1997/1998 vs. 2004. Arch. Environ. Contam. Toxicol.54, 525–534. doi: 10.1007/s00244-007-9041-1
44
HennyC. J.GroveR. A.KaiserJ. L.JohnsonB. L. (2010). North American osprey populations and contaminants: Historic and contemporary perspectives. J. Toxicol. Environ. Health Part B13, 579–603. doi: 10.1080/10937404.2010.538658
45
HennyC. J.KaiserJ. L. (2022). Rotenone use and subsequent prey loss lowers osprey fledging rates via brood reduction. J. Rapt. Res.56, 37–54. doi: 10.3356/jrr-21-14
46
HennyC. J.OvertonW. S.WrightH. M. (1970). Determining parameters for populations by using structured models. J. Wildl. Manage.34, 690–703. doi: 10.2307/3799133
47
HennyC. J.SmithM. M.StottsV. D. (1974). The 1973 distribution and abundance of breeding Ospreys in the Chesapeake Bay. Chesapeake Science.15, 125–133. doi: 10.2307/1351027
48
HennyC.WightH. M. (1969). An endangered osprey population: estimates of mortality and production. Auk86, 188–198. doi: 10.2307/4083495
49
HensonS. M.DesharnaisR. A.FunasakiE. T.GalushaJ. G.WatsonJ. W.HaywardJ. L. (2019). Predator-prey dynamics of bald eagles and glaucous-winged gulls at Protection Island, Washington, USA. Ecol. Evol.9, 3850–3867. doi: 10.1002/ece3.5011
50
HindmanL. J.HarveyW. F.CostanzoG. R.ConverseK. A.SteinG. (1997). Avian cholera in ospreys: First occurrence and possible mode of transmission. J. Field. Ornithol.68, 503–508. Available online at: https://digitalcommons.usf.edu/jfo/vol68/iss4/1
51
HoustonG. S.ScottF. (1992). The effect of man-made platforms on osprey reproduction at Loon Lake, Saskatchewan. J. Rapt. Res.26, 152–158. Available online at: https://digitalcommons.usf.edu/jrr/vol26/iss3/7
52
JohnsonD. R.MelquistW. E.FlemingT. L. (2008). Rainstorm effects on osprey brood survival. J. Rapt. Res.42, 51–53. doi: 10.3356/jrr-07-14.1
53
KennedyR. S. (1977). The status of the osprey in tidewater Virginia, 1970-1971. In: OgdenJ. C. (ed.) Trans Proc North Am Osprey Res Conf. (Washington D.C.: U.S. Natl Park Serv), 121–133.
54
KilpatrickA. M.LaDeauS. L.MarraP. P. (2007). Ecology of West Nile virus transmission and its impact on birds in the Western Hemisphere. Auk.124, 1121–1136. doi: 10.1093/auk/124.4.1121
55
KimmelD. G.MillerW. D.HardingL. W.HoudeE. D.RomanM. R. (2009). Estuarine ecosystem response captured using synoptic climatology. Estuaries Coast.32, 403–409. doi: 10.1007/s12237-009-9147-y
56
KinkeadT. P. (1985). Age structure and dispersal of Chesapeake Bay ospreys (Williamsburg, VA. USA: College of William and Mary). M.A. Thesis.
57
KnightR. L.SkagenS. K. (1988). Agonistic asymmetries and the foraging ecology of bald eagles. Ecology69, 1188–1194. doi: 10.2307/1941273
58
KushlanJ. A.BassO. L. (1983). “ Decreases in the southern Florida osprey population, a possible result of food stress,” in Biology and management of bald eagles and ospreys. Eds. BirdD. M.SeymourN. R.GerrardJ. M. ( Harpell Press, Ste. Anne de Bellevue, Quebec), 187–200.
59
LanktonJ. S. (2022). Wildlife mortality events in counties surrounding the Chesapeake Bay recorded in the Wildlife Health Information Sharing Partnership Event Reporting System (WHISPers) from 2000-2020 ( U.S. Geological Survey data release). doi: 10.5066/P9ULAM7G
60
LazarusR. S.RattnerB. A.BrooksB. W.DuB.McGowanP. C.BlazerV. S.et al. (2015b). Exposure and food web transfer of pharmaceuticals in ospreys (Pandion haliaetus): Predictive model and empirical data. Integr. Environ. Assess. Manag11, 118–129. doi: 10.1002/ieam.1570
61
LazarusR. S.RattnerB. A.McGowanP. C.HaleR. C.Karouna-RenierN. K.EricksonR. A.et al. (2016). Chesapeake Bay fish-osprey (Pandion haliaetus) food chain: evaluation of contaminant exposure and genetic damage. Environ. Toxicol. Chem.35, 1560–1575. doi: 10.1002/etc.3386
62
LazarusR. S.RattnerB. A.McGowanP. C.HaleR. C.SchultzS. L.Karouna-RenierN. K.et al. (2015a). Decadal re-evaluation of contaminant exposure and productivity of ospreys (Pandion haliaetus) nesting in Chesapeake Bay regions of concern. Environ. pollut.205, 278–290. doi: 10.1016/j.envpol.2015.05.026
63
MachmerM. M.YdenbergR. C. (1998). The relative roles of hunger and size asymmetry in sibling aggression between nestling ospreys, Pandion haliaetus. Can. J. Zool.76, 181–186. doi: 10.1139/z97-183
64
MartinT. E. (1987). Food as a limit on breeding birds: a life-history perspective. Annu. Rev. Ecol. Syst.18, 453–487. doi: 10.1146/annurev.es.18.110187.002321
65
McGowanK. J. (2020). “ Fish crow (Corvus ossifragus),” in Birds of the world. Eds. PooleA. F.GillF. B. ( Cornell Lab of Ornithology, Ithaca, NY, USA). doi: 10.2173/bow.fiscro.01
66
McLeanP. K. (1986). The feeding ecology of the Chesapeake Bay Ospreys and the growth and behavior of their young (Williamsburg, VA USA: The College of William and Mary). M.A. Thesis.
67
McLeanP. K.ByrdM. A. (1991). The diet of Chesapeake Bay ospreys and their impact on the local fishery. J. Rapt. Res.25, 109–112. Available online at: https://digitalcommons.usf.edu/jrr/vol25/iss4/2
68
MeadeP. S.BandawaneP.BushfieldK.HoxieI.AzconaK. R.BurgosD.et al. (2024). Detection of clade 2.3.4.4b highly pathogenic HFN1 Influenza virus in New York City. J. Virol.98, e00626-24. doi: 10.1128/jvi.00626-24
69
NajjarR. G.PykeC. R.AdamsM. B.BreitburgD.HershnerC.KempM.et al. (2010). Potential climate-change impacts on the Chesapeake Bay. Est. Coast. Shelf Sci.86, 1–20. doi: 10.1016/j.ecss.2009.09.026
70
NemethN. M.KratzG. E.BatesR.ScherpelzJ. A.BowenR. A.KomarfN. (2009). Clinical evaluation and outcomes of naturally acquired west nile virus infection in raptors. J. Zoo Wildl. Med.40, 51–63. doi: 10.1638/2007-0109.1
71
NewtonI. (1998). Population limitation in birds (London, UK: Academic Press).
72
NormanD. M.BreaultA. M.MoulI. E. (1989). Bald eagle incursions and predation at great blue heron colonies. Colonial Waterbirds.12, 215–217. doi: 10.2307/1521343
73
OdsjöT.RoosA.JohnelsA. G. (2004). The tail feathers of osprey nestlings (Pandion Haliaetus L.) as indicators of change in mercury load in the environment of southern Sweden, (1969–1998): A case study with a note on the simultaneous intake of selenium. AMBIO33, 133–137. Available online at: https://www.jstor.org/stable/4315468
74
OdsjöT.SondellJ. (1976). Reproductive success in ospreys Pandion haliaetus in southern and central Sweden 1971-1973. Ornis Scandinavica.7, 71–84.
75
OehlertG. W. (1992). A note on the delta method. Am. Stat.46, 27–29. doi: 10.1080/00031305.1992.10475842
76
OgdenJ. C. (1975). Effects of bald eagle territoriality on nesting ospreys. Wilson Bull.87, 496–505. Available online at: https://digitalcommons.usf.edu/wilson_bulletin/vol87/iss4/5
77
ParrishJ. K.MarvierM.PaineR. T. (2001). Direct and indirect effects: interactions between bald eagles and common murres. Ecol. Appl.11, 1858–1869. doi: 10.2307/3061101
78
PooleA. (1982). Brood reduction in temperate and sub-tropical ospreys. Oecologia53, 111–119. doi: 10.1007/bf00377144
79
PooleA. (1985). Courtship feeding and osprey reproduction. Auk102, 479–492. doi: 10.1093/auk/102.3.479
80
PooleA. F. (1989). Ospreys: a natural and unnatural history (New York: Cambridge University Press).
81
PooleA. F.BierregaardR. O.MartellM. S. (2002). “ Osprey (Pandion haliaetus),” in The birds of North America, no. 683. Eds. PooleA.GillF. ( The Birds of North America, Inc, Philadelphia).
82
Potomac River Fisheries Commission (2024). Commercial finfish landing report (Colonial Beach, VA: Potomac River Fisheries Commission). Available online at: https://www.prfc.us/pdfs/2025/2024finfish4-25-25.pdf (Accessed July 1, 2025).
83
PrevostY. (1979). Osprey-bald eagle interactions at a common foraging site. Auk96, 413–414. Available online at: https://digitalcommons.usf.edu/auk/vol96/iss2/24
84
RameyA. M.HillN. J.DelibertoT. J.GibbsS. E. J.HopkinsM. C.LangA. S.et al. (2022). Highly pathogenic avian influenza is an emerging disease threat to wild birds in North America. J. Wildl. Manage86 (2), e22171. doi: 10.1002/jwmg.22171
85
RattnerB. A.DayD. D. (2024). Osprey reproduction in vicinity of the choptank river (Reston, VA: U.S. Geological Survey Data Release). doi: 10.5066/P14MRUZC
86
RattnerB. A.DrovetskiS. V.LanktonJ. S.McGowanP. C.WazniakC. E.EgertonT. A. (2022). Review of harmful algal bloom effects on birds and implications for avian wildlife in the Chesapeake Bay region. Harmful Algae120, 102319. doi: 10.1016/j.hal.2022.102319
87
RattnerB. A.EisenreichK. M.GoldenN. H.McKernanM. A.HothemR. L.CusterT. W. (2005). Retrospective ecotoxicological data and current information needs for terrestrial vertebrates residing in coastal habitat of the United States. Arch. Environ. Contam. Toxicol.49, 257–265. doi: 10.1007/s00244-004-0193-y
88
RattnerB. A.GoldenN. H.ToschikP. C.McGowanP. C.CusterT. W. (2008). Concentrations of metals in blood and feathers of nestling ospreys (Pandion haliaetus) in Chesapeake and Delaware Bays. Arch. Environ. Contam. Toxicol.54, 114–122. doi: 10.1007/s00244-008-9143-4
89
RattnerB. A.McGowanP. C.GoldenN. H.HatfieldJ. S.ToschikP. C.LukeiR. F.et al. (2004). Contaminant exposure and reproductive success of ospreys (Pandion haliaetus) nesting in Chesapeake Bay regions of concern. Arch. Environ. Contam. Toxicol.47, 126–140. doi: 10.1007/s00244-003-3160-0
90
R Core Team (2024). R: A language and environment for statistical computing (Vienna, Austria: R Foundation for Statistical Computing). Available online at: https://www.R-project.org/ (Accessed July 1, 2025).
91
ReeseJ. G. (1969). A Maryland osprey population 75 years ago and today. Maryland Birdlife.25, 116–119. Available online at: https://digitalcommons.usf.edu/maryland_birdlife/vol25/iss4/13
92
ReeseJ. G. (1970). Reproduction in a chesapeake bay osprey population. Auk87, 747–759. doi: 10.2307/4083709
93
ReeseJ. G. (1975). Osprey nest success in Eastern Bay, Maryland. Ches. Sci.16, 56–61. doi: 10.2307/1351084
94
ReeseJ. G. (1991). “ Osprey (Pandion haliaetus),” in Habitat requirements for chesapeake bay living resources. Eds. FunderburkS. L.JordanS. J.MihurskyJ. A.RileyD. ( Chesapeake Research Consortium, Inc, Solomons, Maryland), 20.1–20.11.
95
ReeseJ. G. (1996). “ Osprey,” in Atlas of the breeding birds of Maryland and the District of Columbia. Eds. RobbinsC. S.BlomE. A. T. ( University of Pittsburgh Press, Pittsburgh, PA), 92–93.
96
RuffinoL.SaloP.KoivistoE.BanksP. B.KorpimakiE. (2014). Reproductive responses of birds to experimental food supplementation: a meta-analysis. Front. Zool.11, 1–13. doi: 10.1186/preaccept-1035746911122745
97
SauerJ. R.NivenD. K.HinesJ. E.ZiolkowskiD. J.PardieckK. L.FallonJ. E.et al. (2019). The north american breeding bird survey, results and analysis 1966 - 2019 (Laurel, MD: USGS Patuxent Wildlife Research Center). Version 2.07.2019.
98
SaurolaP. (1997). The osprey (Pandion haliaetus) and modern forestry: a review of population trends and their causes in Europe. J. Rapt. Res.31, 129–137. doi: 10.3356/JRR-13-OSPR-13-03
99
ScharfF. S.MandersonJ. P.FabrizioM. C.PessuttiJ. P.RosendaleJ. E.ChantandR. J.et al. (2004). Seasonal and interannual patterns of distribution and diet of bluefish within a Middle Atlantic Bight estuary in relation to abiotic and biotic factors. Estuaries27, 426–436. doi: 10.1007/bf02803535
100
SchmidF. C. (1977). “ A 1934 vs. 1967 comparison of osprey nesting populations,” in OgdenJ. C., ed. Transactions and Proceedings No. 2 of the North American Osprey Research Conference (Williamsburg, Virginia: College of William and Mary), 135–137.
101
SchuellerA. M.WilliamsE. H.CheshireR. T. (2014). A proposed, tested, and applied adjustment to account for bias in growth parameter estimates due to selectivity. Fish. Res.158, 26–39. doi: 10.1016/j.fishres.2013.10.023
102
SlabeV. A.AndersonJ. T.CooperJ.BrownB.OrtizP.BuchweitzJ.et al. (2019). Lead in piscivorous raptors during breeding season in the Chesapeake Bay region of Maryland and Virginia, USA. Environ. Toxicol.38, 862–871. doi: 10.1002/etc.4376
103
SmithL. A.LinkJ. S.CadrinS. X.PalkaD. L. (2015). Consumption by marine mammals on the Northeast US continental shelf. Ecol. Appl.25, 373–389. doi: 10.1890/13-1656.1
104
Southeast Data Assessment and Review [SEDAR] (2020). SEDAR 69 – Atlantic menhaden benchmark stock assessment report (North Charleston, SC: SEDAR 691).
105
Southeast Data Assessment and Review [SEDAR] (2022). Atlantic states marine fisheries commission – atlantic menhaden stock assessment update (North Charleston, SC: Atlantic States Marine Fisheries Commission).
106
Southeastern Cooperative Wildlife Disease Study (2024). Report for accession number W24-00516 (Athens, GA: University of Georgia).
107
SoylandA. (2010). Effects of the introduced lake trout (Salvelinus namaycush) on the osprey (Pandion haliaetus) population in Yellowstone National Park (Sørhellinga (Norway: Institute of Nature Management, Norwegian University of Life Science).
108
SpitzerP. R. (1980). Dynamics of a discrete coastal breeding population of ospreys in the northeastern USA 1969-1979 (Ithaca, New York, USA: Cornell University).
109
SpitzerP. R.RisebroughR. W.WalkerW.HernandezR.PooleA.PulestonD.et al. (1978). Productivity of ospreys in Connecticut–Long Island increases as DDE residues decline. Science202, 333–335. doi: 10.1126/science.99818
110
SteenhofK.NewtonI. (2007). “ Assessing nesting success and productivity,” in Raptor research and management techniques. Eds. BirdD. M.BildsteinK. L. (Surrey, BC, Canada: Hancock House), 181–191.
111
SteidlR. J. (1990). Reproductive ecology of ospreys and peregrine falcons nesting in New Jersey (Amherst, Massachusetts, USA: University of Massachusetts).
112
StinsonC. H. (1976). The evolutionary and ecological significance of the clutch size of the osprey (Williamsburg, Virginia, USA: William and Mary). M.A. Thesis.
113
StinsonC. H.ByrdM. A. (1976). A comparison of past and present Osprey breeding populations in Coastal Virginia. Bird-Banding47, 58–262. doi: 10.2307/4512239
114
StinsonC. H.LauthnerJ.RayR. T. (1987). The effect of weather conditions on the behavior of ospreys in northwestern Washington. Can. J. Zoo65, 2116–2118. doi: 10.1139/z87-325
115
ToschikP. C.RattnerB. A.McGowanP. C.ChristmanM. C.CarterD. B.HaleR. C.et al. (2005). Effects of contaminant exposure on reproductive success of ospreys (Pandion haliaetus) nesting in Delaware River and Bay, USA. Environ. Toxicol. Chem.24, 617–628. doi: 10.1897/04-141r.1
116
TyrrellW. B. (1936). The ospreys of smith’s point, Virginia. Auk53, 261–268.
117
UphoffJ. H. (2003). Predator–prey analysis of striped bass and Atlantic menhaden in upper Chesapeake Bay. Fish. Manage. Ecol.10, 313–322. doi: 10.1046/j.1365-2400.2003.00366.x
118
U.S. Department of Agriculture (2025). Detections of highly pathogenic avian influenza database. Available online at: https://www.aphis.usda.gov/livestock-poultry-disease/avian/avian-influenza/hpai-detections/wild-birds (Accessed August 10, 2025).
119
ViveretteC. B.GarmanG. C.McIninchS.MarkhamA. C.WattsB. D.MackoS. A. (2007). Finfish-waterbird trophic interactions in tidal freshwater tributaries of the Chesapeake Bay. Waterbirds30, 50–62. doi: 10.1675/1524-4695(2007)030[0050:ftiitf]2.0.co;2
120
WahlR.BarbraudC. (2014). The demography of a newly established osprey Pandion haliaetus population in France. Ibis.156, 84–96. doi: 10.1111/ibi.12114
121
WalterJ. F.OvertonA. S.FerryK. H.MathersM. E. (2003). Atlantic coast feeding habits of striped bass: a synthesis supporting a coast-wide understanding of trophic biology. Fish. Manage. Ecol.10, 349–360. doi: 10.1046/j.1365-2400.2003.00373.x
122
WattsB. D.ByrdM. A. (2022). Policy and the social burden of bald eagle recovery. Conserv. Sci. Pract.4 (9), e12764. doi: 10.1111/csp2.12764
123
WattsB. D.ByrdM. A.WattsM. U. (2004). Status and distribution of breeding Ospreys in the Chesapeake Bay: 1995-96. J. Rapt. Res.38, 47–54. Available online at: https://digitalcommons.usf.edu/jrr/vol38/iss1/6
124
WattsB. D.HinesC. (2025). Response: Commentary: Demographic response of osprey within the lower Chesapeake Bay to fluctuations in menhaden stock. Front. Mar. Sci.12, 1565843. doi: 10.3389/fmars2025.1565843
125
WattsB. D.MarkhamA. C.ByrdM. A. (2006). Salinity and population parameters of Bald Eagles (Haliaeetus leucocephalus) in the lower Chesapeake Bay. Auk123, 393–404. doi: 10.1093/auk/123.2.393
126
WattsB. D.PaxtonB. J. (2007). Ospreys of the Chesapeake Bay: Population recovery, ecological requirements, and current threats. Waterbirds30, 39–49. doi: 10.1675/1524-4695(2007)030[0039:ootcbp]2.0.co;2
127
WattsB. D.PaxtonB. J.HinesC.AndersonS. K. (2023). Estimating potential menhaden consumption by double-crested cormorants along the coast of North Carolina. Fron. Mar. Sci.10, 1193429. doi: 10.3389/fmars.2023.1193429
128
WattsB. D.StinsonC. H.McLeanP. K.GlassK. A.AcademiaM. H.ByrdM. A. (2024). Demographic response of osprey within the lower Chesapeake Bay to fluctuations in menhaden stock. Front. Mar. Sci10, 1284462. doi: 10.3389/fmars.2023.1284462
129
WattsB. D.TherresG. D.ByrdM. A. (2007). Status, distribution and the future of Bald Eagles in the Chesapeake Bay. Waterbirds30, 25–38. doi: 10.1675/1524-4695(2007)030[0025:sdatfo]2.0.co;2
130
WiemeyerS. N. (1971). Reproductive success of Potomac River ospreys 1971. Ches. Sci.12, 278–280. doi: 10.2307/1350918
131
WiemeyerS. N.BunckC. M.KrynitskyA. J. (1988). Organochlorine pesticides, polychlorinated biphenyls, and mercury in osprey eggs—1970-79—and their relationships to eggshell thinning and productivity. Arch. Environ. Contam. Toxicol.17, 767–787. doi: 10.1007/bf01061982
132
WiemeyerS. N.SchmelingS. K.AndersonA. (1987). Environmental pollutant and necropsy data for ospreys from the eastern United States 1975-1982. J. Wildl. Dis.23, 279–291. doi: 10.7589/0090-3558-23.2.279
133
Wildlife Center of Virginia (2024). Unvalidated data. WILD-ONe database. Available online at: www.WILD-ONe.org (Accessed January 12, 2025).
134
WittJ. W. (1996). Long-term population monitoring of osprey along the Umpaqua River in Western Oregon. J. Rapt. Res.30, 62–69. Available online at: https://digitalcommons.usf.edu/jrr/vol30/iss2/2
Summary
Keywords
Atlantic menhaden, Chesapeake Bay, food stress, osprey, productivity
Citation
Watts BD, Hines C, Byrd MA, Lukei Jr. RF, Paxton BJ, Duval L, Spiewak KE, Kearns GD, Day DD and Rattner BA (2026) Widespread reproductive deficits in Chesapeake Bay ospreys. Front. Mar. Sci. 13:1685158. doi: 10.3389/fmars.2026.1685158
Received
13 August 2025
Revised
05 May 2026
Accepted
18 May 2026
Published
08 June 2026
Volume
13 - 2026
Edited by
Romuald Lipcius, College of William & Mary, United States
Reviewed by
Maxwell Wilder, University of Minnesota Duluth, United States
Joseph Smith, Independent Researcher, Cape May, United States
Updates
Copyright
© 2026 Watts, Hines, Byrd, Lukei, Paxton, Duval, Spiewak, Kearns, Day and Rattner.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Bryan D. Watts, bdwatt@wm.edu
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.