Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: Comparing natural wetlands to areas with shellfish aquaculture

Scott Jennings; David Lumpkin; Nils Warnock; T Emiko Condeso; John P Kelly

doi:10.1371/journal.pone.0261963

. 2021 Dec 31;16(12):e0261963. doi: 10.1371/journal.pone.0261963

Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: Comparing natural wetlands to areas with shellfish aquaculture

Scott Jennings ^1,^*, David Lumpkin ¹, Nils Warnock ¹, T Emiko Condeso ¹, John P Kelly ¹

Editor: Hans-Uwe Dahms²

PMCID: PMC8719746 PMID: 34972178

Abstract

Movement by animals to obtain resources and avoid predation often depends on natural cycles, and human alteration of the landscape may disrupt or enhance the utility of different habitats or resources to animals through the phases of these cycles. We studied habitat selection by GPS/accelerometer-tagged great egrets (Ardea alba) foraging in areas with shellfish aquaculture infrastructure and adjacent natural wetlands, while accounting for tide-based changes in water depth. We used integrated step selection analysis to test the prediction that egrets would express stronger selection for natural wetlands (eelgrass, tidal marsh, and other tidal wetlands) than for shellfish aquaculture areas. We also evaluated differences in foraging behavior among shellfish aquaculture areas and natural wetlands by comparing speed travelled (estimated from distance between GPS locations) and energy expended (Overall Dynamic Body Acceleration) while foraging. We found evidence for stronger overall habitat selection for eelgrass than for shellfish aquaculture areas, with results conditional on water depth: egrets used shellfish aquaculture areas, but only within a much narrower range of water depths than they used eelgrass and other natural wetlands. We found only slight differences in our metrics of foraging behavior among shellfish aquaculture areas and natural wetlands. Our results suggest that although great egrets appear to perceive or experience shellfish aquaculture areas as suitable foraging habitat during some conditions, those areas provide less foraging opportunity throughout tidal cycles than natural wetlands. Thus, expanding the footprint of shellfish aquaculture into additional intertidal areas may reduce foraging opportunities for great egrets across the range of tidal cycles. Over longer time scales, the ways in which natural wetlands and shellfish aquaculture areas adapt to rising sea levels (either through passive processes or active management) may change the ratios of these wetland types and consequently change the overall value of Tomales Bay to foraging great egrets.

Introduction

Motile organisms have the capacity to change the environmental conditions they experience through movement [1]. Often, these movements are tuned to natural cycles (e.g. seasonal, circadian, tidal) that drive resource availability or abundance (e.g. [2–4]). Human alteration of natural landscapes can alter animal movement behaviors and the ecological roles animals play. For example, human-altered landscapes were associated with reduced distance travelled across a range of mammalian taxa [5] and reduced connectivity within a population of birds [6]. In production landscapes where industry or agriculture can alter the natural availability of resources, animals may spend more time and energy travelling to find food and to avoid disturbance, and this may be especially true for larger, more mobile animals [7]. The impact of human landscape alteration on habitat and resource availability at certain times of seasonal migration has been well studied [8], but the interaction of human impacts with movement cycles spanning shorter time scales is less well understood.

Great egrets (Ardea alba) are mobile, generalist wetland predators found in temperate and tropical latitudes throughout the world [9]. They exist in coastal and inland areas and forage in a wide variety of freshwater, estuarine, and marine wetlands and some upland habitats. Great egrets are opportunistic foragers and will select foraging habitat where environmental or human-caused factors enhance prey abundance and availability [10]. They employ a range of foraging behaviors, most commonly slow-walking or sit-and-wait [9], and foraging behavior may be related to habitat type [11]. Formal energetic investigation has revealed that walking and striking at prey is inexpensive for great egrets relative to the energy gained from prey. As a result, energy gain is positively related to the number of strikes at prey and the number of steps taken [12]. However, it remains unclear under what circumstances different feeding activities are more profitable for great egrets and whether any differences may be influenced by habitat type.

Great egrets and other wading birds (Ardeidae) are well known to forage at finfish aquaculture facilities, and extensive research has been conducted and effort expended to reduce the economic impact of these foraging behaviors [13, 14]. Less attention has been given to the relationships between wading birds and shellfish aquaculture. Shellfish aquaculture can provide habitat for nekton [15], which may in turn serve as valuable prey for wading birds. However, if the structure of aquaculture equipment provides sufficient cover, even abundant prey may nevertheless be unavailable to wading birds [16]. Additionally, increased human activity associated with maintaining aquaculture infrastructure may dissuade wading birds from foraging there. Commercial shellfish harvest on Tomales Bay, Marin Co., CA, has increased more than four-fold since 1990 [17]. However, the degree to which these activities alter the value of intertidal areas as wildlife habitat remains understudied, limiting the empirical evidence available to agencies responsible for regulating this industry.

Extensive eelgrass (Zostera spp.) beds exist on Tomales Bay, often directly adjacent to shellfish aquaculture infrastructure. In coastal systems, eelgrass can provide important foraging areas for great egrets [18]. These habitats also provide important ecosystem services, including habitat for economically and culturally important species [19, 20], long-term sequestration of blue carbon [21], and buffering of nutrient pulses from terrestrial to marine systems [22]. Eelgrass habitats are threatened by a range of human activities and receive considerable attention by conservation and regulatory entities (e.g. [23]). A better understanding of the ways in which shellfish aquaculture facilities and infrastructure might operate as surrogates for natural habitat for foraging wading birds is important for determining reasonable limits on the loss or alteration of natural habitat. Additionally, because top predators, including wading birds, may exert top-down regulation of processes in natural systems like eelgrass [24, 25], it is important to understand the degree to which aquaculture operations adjacent to natural areas may alter those regulating effects.

We used GPS/Accelerometer dataloggers to quantify great egret habitat selection and foraging behavior in four types of tidal wetlands on Tomales Bay: eelgrass, tidal marsh, other (mostly unvegetated mudflat) natural tidal, and shellfish aquaculture areas. Our objectives were two-fold. First, we tested an a priori prediction about relative selection of tidal habitats on Tomales Bay. We hypothesized that great egrets perceive or experience eelgrass and other natural wetlands as higher quality foraging habitat than shellfish aquaculture areas. Based on this hypothesis we expected foraging great egrets to select natural wetlands, and especially eelgrass beds, more strongly than areas occupied by shellfish aquaculture infrastructure. Second, we quantitatively described egret foraging movement and behavior in these areas. The literature contains conflicting information about how behaviors (as measured with GPS/Accelerometry) might translate to true foraging success or other fitness metrics, so we did not develop and test formal predictions. Rather, our aim for this component of our study was descriptive and intended to generate hypotheses about behavioral differences by great egrets in shellfish aquaculture areas relative to natural tidal wetlands. To address this objective, we compared two quantitative measures of foraging behavior between wetland types: 1) foraging speed; and 2) energy expenditure.

Methods

Study area

Tomales Bay (38.161921° N, -122.905464° W) is a linear estuary in Marin Co., CA, USA. It covers an area totaling approximately 31.9 km² [26] and is permanently open to the Pacific Ocean at its northern end (Fig 1). The bay is mostly shallow (mean depth <6.5 m below Mean Lower Low Water [MLLW]), and of the total area approximately 14.7 km² is shallow enough for great egret foraging at some point in the tidal range (see below). We defined these 14.7 km² of tidal and subtidal wetlands as our study area. Of these 14.7 km², approximately 5.3 km² (35.8%) is covered by eelgrass, 0.3 km² (2.3%) by shellfish aquaculture infrastructure, 3.0 km² (20.9%) by tidal marsh, and the remaining 6.0 km² (41.1%) by mostly unvegetated intertidal and subtidal flats (see below for wetland classification). The most extensive shellfish aquaculture infrastructure in areas shallow enough for egrets to forage is in the northern part of the bay, near Walker Creek delta and Toms Point (Fig 1).

Trapping/Tagging

Great egrets were trapped and tagged at three locations within Tomales Bay: Toms Point, Walker Creek, and Cypress Grove (Fig 1). Trapping and banding methods conformed to the Ornithological Councils guidelines to the use of wild birds in research [27]. Great egrets were captured and tagged under the United States Department of Interior federal bird banding permit #24179 and the California Department of Fish and Wildlife’s Scientific Collecting Permit #SC-1383.

We lured egrets using decoys and bait and trapped them in padded leghold traps (Victor #3 Soft Catch Coil Spring Trap) following established methods for wading birds [28]. Traps were modified to close with less force and not close tighter than the diameter of egret legs. We removed birds from traps immediately upon capture. We attached GPS tags (Bird Solar 48g or Bird UMTS 25g, e-obs GmbH, Gruenwald, Germany) using a backpack harness of Teflon ribbon (Bally Ribbon Mills, PA, USA). We programmed the tags to collect a GPS location and a 6 sec burst (10 Hz per axis sampling rate) of acceleration data at 5 min intervals.

Data preparation

Habitat mapping

Our analysis required three types of environmental data: wetland type, elevation, and predicted tide height. We combined data from the following three sources to assign wetland type to GPS locations.

Wetlands. We used the California Aquatic Resources Inventory dataset (CARI), a publicly available habitat classification [26] to assign habitat values to each GPS location. The original CARI dataset segregated habitat into finer categories than necessary for our analysis. We retained the “tidal marsh” classification but combined unvegetated intertidal and subtidal areas into “other tidal.” We excluded areas classified as freshwater wetlands and non-wetland areas from this analysis.

Eelgrass. We also acquired a GIS layer representing the areal extent of eelgrass in Tomales Bay [29]. These data were collected in 2017 by a combination of vessel-borne side-scan sonar at high tides and low altitude unmanned aerial vehicle photography at low tide. We buffered the resulting eelgrass layer by 5m to close gaps between shapes representing small adjacent eelgrass patches. This meant more of these small patches were included in the eelgrass layer when converted to the 10 m² raster (see below), but it also meant that we classified more non-eelgrass areas as eelgrass than vice versa.

Shellfish aquaculture infrastructure. We digitized all visually apparent structures associated with shellfish aquaculture on the bay using 15.25 cm resolution ortho images [30] viewed in a GIS at scales between 1:300 and 1:3000. Shellfish aquaculture on Tomales Bay primarily uses tipping baskets, which are plastic mesh envelopes approximately 40 by 80 cm, either strung together in lines and allowed to rest on the bottom or suspended off the bottom by risers. We digitized visible aquaculture infrastructure using a combination of polygon and linear features: long lines of baskets (generally appearing <1m wide) separated by >2m were represented by lines then enclosed in a 1m buffer (resulting in a polygon). Where there was <2m between visible gear we enclosed the visible gear in shared polygons. Until just prior to our study, clams were cultured in cylindrical bags that become buried in the sediment; we also used polygons to enclose areas where substrate scars remaining from this activity were visible in the aerial images. All polygons were then merged, and we buffered the resulting shape by 1m to close small interior gaps.

Elevation. Although our study area extended above sea level, where height is termed “topographic elevation”, and below sea level, where height is termed “bathymetric depth”, we use “elevation” to represent both. We used two data sources for elevation. Our primary elevation source was a LiDAR-derived digital elevation map (DEM) [31]. However, this map did not provide accurate elevation below Mean Lower Low Water (MLLW). Thus, for areas below that elevation we used a Tomales Bay bathymetry layer [32].

Tide height. There is no tide gauge on Tomales Bay. We generated predicted tide height by offsetting hourly predictions from the San Francisco tide station (NOAA station #9414290) by the published time and height values for the mid-bay subordinate station at Blake’s Landing (station #9415396). From these hourly predictions we then interpolated tide estimates for the timestamp associated with each GPS location collected by the tags.

Combining layers. Finally, the CARI, eelgrass and shellfish layers were converted to rasters with 10m cell size, then combined such that CARI areas were reclassified as either eelgrass or shellfish where the later datasets existed; where the eelgrass and shellfish rasters contained values of NA, the CARI values were retained. Thus, our combined layer classified tidal wetlands into four categories: eelgrass, shellfish, tidal marsh (any vegetated tidal area not contained in the eelgrass layer), and other tidal (almost entirely unvegetated inter- and subtidal mud flats). In all analyses, other tidal was coded as the reference level for the wetland type variable.

We calculated water depth as the predicted tide height at the timestamp associated with each GPS fix minus the elevation at that GPS fix. The 0 values for the datum for the DEM (NAVD88 GEOID 12B) and that for the bathymetric and tide predictions (MLLW) differed by < 1 cm, a difference we deemed small enough to not warrant converting to make both equal. Our calculated water depth was specific to the location and time that each GPS fix was collected. The same location would experience different water depths through the tide cycle, and an individual egret could experience different water depths at a particular time by moving up or down in elevation. Furthermore, our water depth variable contained negative values, which represented the height above the water level at each specific GPS location and timestamp. Thus, a given location at 0.5 m elevation would have a water depth of 0.3 m at a tide height of 0.8 m, a water depth of 0 m at a tide height of 0.5 m, and a water depth of -0.5 m at a tide height of 0 m.

GPS/Accelerometer data

We first filtered GPS location data to just those points with tag-estimated accuracy < 10m (best accuracy was ~3m). Because our focus was on foraging habitat selection, we further filtered GPS data to include only locations collected during daylight hours (when great egrets forage) and when the bird’s speed (as estimated by the GPS unit) was ≤ 5 m s^-1 to exclude locations collected while flying (minimum reported great egret mean flight speed is 9.9 m s-1) [9]. The bathymetry of Tomales Bay is generally gradual except where tidal channels cut through shallow flats. Because our water depth variable represented the average depth over a 10 m² area, raster cells along the edges of these channels were occasionally assigned depths deeper than great egrets can forage in [9] but still provided some accessible foraging depth. We excluded points with assigned depths greater than 1 m because we felt that these points, although biologically plausible, had too strong an influence on the models we fitted. Because there were relatively few shellfish aquaculture raster cells with depths < -0.5 m (i.e., 0.5 m above the tidal level; S1 Fig), we also restricted our analysis of habitat selection to depths ≥ -0.5 m.

Dynamic acceleration is the acceleration an object experiences due to movement, rather than gravity (static acceleration). In biologging studies that employ accelerometers, Overall Dynamic Body Acceleration (ODBA) is the sum of dynamic acceleration across the X, Y and Z axes and is a useful index of energy expenditure across a range of vertebrate taxa [33–35]. We calculated ODBA by taking the difference between static and dynamic acceleration for each of the three accelerometer axes, with the static acceleration being estimated as the average value across the entire 6 sec burst. We summed the absolute values of these differences for the entire burst for each axis and, finally, summed the acceleration for all three axes [35, 36].

Analysis

In studies of habitat selection, how “available” habitats are defined for individual animals can have substantial influence on the inference that can be made [37, 38]. Integrated Step Selection Analysis (iSSA) allows simultaneous inference about both habitat selection processes and movement processes [39]. In iSSA, analysis is based on “steps” representing sequential locations separated by equal time intervals. At the starting point of each step, a particular area of habitat is available for the animal to select from, based on the movement capacity of that animal. The location of the animal at the end point of each step is paired with a set of “available” points that are randomly generated from the distributions of lengths and turn angles of all steps by that animal. Thus, each animal’s observed movement characteristics (step length and turn angle) determines which habitat is considered available from the starting location of each step, and each starting location has its own unique domain of available habitat. This reduces some bias in arbitrary, investigator defined availability domains, and it allows treatment of each observed step (and associated control steps) as the level of measurement while reducing the effect of lack of independence on standard errors [39, 40].

GPS sampling error can contribute substantially to estimated distance travelled by tagged animals [41], particularly as the step length travelled during the sampling interval approaches the magnitude of that measurement error [42]. Although we had 5-min-interval GPS data, preliminary data summarization indicated that a sampling interval of 10 minutes yielded step lengths longer than our GPS tag sampling error and avoided an excessive number of steps with no movement, but was a sufficiently short time span to evaluate fine scale habitat selection. We generated 10 random steps (yielding a balance between estimation error and computational burden [39]) for each observed step and extracted habitat characteristics at the start and end of each observed and available step. For all parts of our analysis, it is important to keep in mind that inference was dependent on both the sampling interval (time between locations) and the spatial resolution at which we considered habitat attributes. For example, increasing the sampling interval (and thus expected distance travelled on each step) would expand the radius of the availability domain, whereas a coarser resolution habitat raster may have limited the diversity of wetland types available at a given sampling interval. Additionally, the water depths we calculate are not necessarily the precise water depths that egrets were selecting at each GPS timestamp. Rather, these depths represented the average depth across the 10 m² raster cell within which the GPS point was located. We chose to use 10 m habitat raster resolution to match the accuracy of our GPS tags. We did not explore the sensitivity of our results to variation in sampling interval or habitat raster resolution.

We were interested in examining third order resource selection (selection of feeding sites), and this selection was conditional on tagged egrets first selecting to live in the San Francisco area (first order selection) and then selecting to forage at Tomales Bay (second order selection) [43]. We fitted seven conditional logistic regression models to the data for each individual egret, to test our hypotheses about habitat selection. The response of these models was whether a location was “used” or “available.” Because iSSA is focused on movement steps, one can choose to evaluate wetland type at the start or the end of the step. We used wetland type at the end of the step for inference about habitat selection [44]. We fitted iSSA models to each individual egret separately rather than using bird id as a random effect [39, 44]. To test our prediction of lower selection for shellfish aquaculture areas than for natural wetlands, we fitted a “full” model with the interaction between the 4-level wetland type variable and a quadratic effect for time-specific water depth (i.e., to allow selection of each wetland type to vary differently as water depth changed). We did not consider additional possible predictor variables (e.g., time of year, sex of bird, etc.) because of the small number of birds in our study. To evaluate evidence for the effects of wetland type and water depth on selection, we compared AICc values (Akaiki Information Criterion corrected for small sample size) [45] between this full model and the six nested models that represent the possible combinations between these variables, including the linear and quadratic effects of water depth (candidate models shown in S1 Table). We calculated relative selection strength from the coefficients in the best-supported (lowest AICc value) of these models.

We used a combination of the conditional logistic regression models described above and linear mixed effect models to evaluate evidence for differences in foraging behavior between wetland types. To evaluate wetland-based variation in step lengths we added the interactions between wetland type at the start of the step and both step length and the natural logarithm of step length [44, 46] to the best supported habitat selection model. We judged this model’s ability to explain variability in our data by comparing its AICc value to the best-supported model from the habitat selection analysis. We then used the estimated coefficients for step lengths to modify the naïve step length estimates (from the raw data) to remove the effect of habitat selection from our estimate of movement differences between wetland types [44]. We used linear mixed models to evaluate whether ODBA varied between habitats, and we used ODBA values collected at the original 5 min interval to maximize available data. This model included a random effect for bird id, and therefore model estimates were taken as our estimate across all tagged birds. We fitted a model with wetland type as the predictor variable and compared it to an intercept-only model with the Likelihood Ratio Test to evaluate evidence for wetland type-based differences in ODBA. We fitted these models with Maximum Likelihood rather than Restricted Maximum Likelihood because we were interested in evaluating the importance of the fixed effects [47]. As with the habitat selection part of our analysis, we did not consider other possible variables that might affect foraging behavior (e.g., sex, time of year) because of the small number of birds in our sample.

Where our best supported models included interaction terms, we base interpretation of results on plotted effects rather than coefficients for individual predictor variables. We used ArcGIS Pro v2.4.0 [48] and R version 4.0.2 [49] for spatial data processing, and R for all analysis. We used the function predict.tidem from the R package oce [50] to interpolate tide level values at each GPS timestamp. We used the R package amt [44] to process GPS data and analyze habitat selection and step length. We used the package lme4 [51] for linear mixed models to analyze ODBA. Code files are archived here: https://doi.org/10.5281/zenodo.5571072. Tagging data are archived here: https://www.movebank.org/cms/webapp?gwt_fragment=page=studies,path=study247850178.

Results

We obtained a mean = 177 ± SE 31 days of Tomales Bay foraging data per tagged great egret, collected between 10 June 2017 and 31 July 2020 (Table 1). Although we had data for 10 egrets using Tomales Bay, we excluded 3 from formal modelling. These birds almost completely avoided using shellfish aquaculture areas (Table 1), which led to convergence issues when maximizing likelihood for the conditional logistic regression models. Some tagged birds spent time outside the study area during the study period. For all tagged birds for which we ceased receiving data, we were unable to determine whether the cause was death, tag loss or tag failure.

Table 1. Summary of tagged great egrets.

		days tracked on Tomales Bay		number of steps by end location
Bird ID	capture site	total days	date range	eelgrass	shellfish	tidal marsh	other tidal	total steps
GREG_10	CG	360	Sep 21, 2018; Jul 31, 2020	4074	100	1116	7420	12710
GREG_8	TP	343	Jul 23, 2018; Jun 30, 2020	3874	386	414	5417	10091
GREG_2	TP	188	Jun 10, 2017; Jun 22, 2018	1463	1038	2029	4006	8536
GREG_1	TP	182	Jun 10, 2017; Aug 31, 2018	2545	672	666	3167	7050
GREG_3	TP	113	Jun 11, 2017; Jun 22, 2018	1672	882	1754	2016	6324
GREG_6	WC	123	Jun 12, 2018; Aug 17, 2019	1740	674	641	2708	5763
GREG_11*	CG	173	Mar 26, 2019; Jul 9, 2020	2078	3	754	2417	5252
GREG_7*	CG	144	Jul 5, 2018; Dec 12, 2018	978	6	593	1822	3399
GREG_5	TP	73	Jun 8, 2018; Feb 7, 2019	1288	199	111	1588	3186
GREG_9*	CG	77	Sep 19, 2018; Dec 4, 2018	275	2	807	1944	3028

Open in a new tab

Number of days tracked on Tomales Bay, CA and number of steps ending in each wetland type for GPS tagged great egrets. Capture sites are as follows: CG = Cypress Grove; TP = Toms Point; WC = Walker Creek. Models failed to converge for birds with few observed steps in shellfish aquaculture areas (indicated by *), so these egrets were excluded from statistical analysis. Total number of days is less than the number of days comprising the date range for egrets that left the study area for migration or other movements.

For our first objective, testing the relative selection of natural wetlands vs. shellfish aquaculture areas, there was consistency among birds in the best-supported models. For all birds, the model with the interaction between wetland type and the quadratic effect of water depth was the best supported and the one with the main effects only of habitat and quadratic water depth was the second best supported. The difference in AICc values (Δ AICc) for these second-ranked models were large (69–229), indicating strong evidence that depth-based resource selection took a substantially different form among wetland types. Thus, we based inference on the best models only (AICc values for all models presented in S1 Table). Recall that the water depth variable represented the average depth over a 10 m² area where each GPS point was located, rather than the precise depth the bird was standing in at the GPS timestamp.

Plotted estimates of log-Relative selection strength from the best-supported models show a degree of individual variation in habitat selection, but also some consistent patterns shared among all or most birds (Fig 2). The most-consistent pattern, observed across all egrets, was a strong quadratic effect (concave downward) of water depth on selection of shellfish aquaculture areas. Between depths of approximately 0 m to 0.6 m, great egrets selected shellfish aquaculture areas about as strongly as eelgrass areas. For two birds (GREG_3 and GREG_6) selection was stronger for shellfish aquaculture areas than eelgrass (with non-overlapping 95% CI) in a very narrow depth band around 0.25 m to 0.5 m. However, for most birds selection for shellfish aquaculture areas was lower than for eelgrass (with non-overlapping 95% CI) when water was deeper than about 0.5 m and in areas that were above the tide line (Fig 2). This result was not simply a factor of shellfish aquaculture infrastructure existing in a narrower, intermediate elevation than natural wetlands, although this may have contributed somewhat at the most negative depths (S1 Fig); the depth range of higher selection for shellfish aquaculture areas was much narrower than the overall depth range we observed for shellfish aquaculture areas (see Discussion for further detail). We also observed a quadratic effect of water depth on selection for eelgrass and tidal marsh, but this effect was generally much weaker than we observed for shellfish aquaculture areas and the quadratic shape varied more among egrets (Fig 2). We found some evidence for differences in selection strength among the three natural wetland types. Eelgrass was selected more strongly than other tidal (mostly unvegetated flats) for most egrets across most depths. For some egrets selection was strongest for tidal marsh when those areas were deeply flooded by higher tides (e.g. GREG_6; Fig 2).

Fig 2 — Other tidal areas (the reference level) were mostly unvegetated inter- and subtidal mud flats. Water depth represents the average depth in the 10 m² grid cell around where the bird was located. Negative values of water depth indicate the bird was located above predicted water level.

For our second objective, describing foraging movement and behavior, we found differences among wetland types, but they mostly involved tidal marsh being different than the other wetlands. In our investigation of step length, the model with the step length*habitat interaction received more support than the one without those variables for all birds (Δ AICc for habitat selection models = 129–601), providing some evidence for consistent differences in step length among all tagged egrets (S2 Table). The plotted probability density of step length distributions, once adjusted for movement related to habitat selection, showed that step length in all wetland types were heavily skewed toward shorter step lengths (Fig 3). However, this skewing was much stronger (i.e., greater probability density for shorter steps) for tidal marsh than other wetlands for all birds.

Fig 3 — Probability density was calculated for step lengths 0–400 m, but only step lengths where there was visible difference in plotted values among habitats are shown.

In our test for different ODBA among wetland types, the model with wetland type provided a substantially better fit to the data than did the intercept only model (χ² (3) = 1249, p < 0.001). The estimated coefficient for tidal marsh was strongly negative and with 95% CI that substantially excluded zero (β = -292.4, 95% CI -315.6 to -269.2). The coefficients for eelgrass and shellfish aquaculture areas were both positive and with 95% CI not overlapping zero (eelgrass β = 154.5, 95% CI 136.7 to 172.2; shellfish β = 104.3, 95% CI 68.5 to 140.2). Thus, there was good evidence that ODBA was different among eelgrass, shellfish aquaculture areas and other tidal, and convincing evidence that ODBA in tidal marsh was different than the other 3 wetland types. ODBA was greatest for eelgrass and shellfish aquaculture areas (about 50 ms^-1s^-1 greater for the former), intermediate for other tidal wetlands (the reference level) and was substantially lower in tidal marsh than the other three wetland types (Fig 4).

Fig 4 — Estimates and 95% profile likelihood confidence intervals are from a linear mixed effect model with bird ID included as a random effect.

Discussion

In this first investigation of great egret foraging behavior and habitat selection in areas where shellfish aquaculture operations exist, we found support for our prediction of greater selection for eelgrass than for shellfish aquaculture areas. Three of 10 egrets effectively avoided shellfish aquaculture areas. Among the egrets that did forage in shellfish aquaculture areas, habitat selection was contingent on tide-based water depth. Specifically, great egrets were more likely to select eelgrass than shellfish aquaculture areas across most water depths, suggesting lower use or avoidance of aquaculture areas at some times through the range of the tidal cycle. We also found that step lengths while foraging were similar among all wetlands except tidal marsh, where they were shorter. Energy expenditure was greatest in eelgrass and shellfish aquaculture areas, intermediate other (mostly unvegetated) tidal areas, and substantially lower in tidal marsh than the former three wetland types. We believe these results for tidal marsh partially reflect short-term roosting there between foraging bouts.

Because we acquired data for only 10 birds, and used data from only seven for formal modelling, our results effectively describe the foraging responses to shellfish aquaculture by the subject individuals but may not accurately predict the behaviors of other individuals in other locations or times. We did not find consistent support among all egrets for our prediction that eelgrass would be selected more strongly than other natural wetlands. Because of our small sample, we have most confidence in the observed pattern that was consistent among all egrets (the difference between shellfish aquaculture and natural wetlands as a group), and we don’t discuss further the inconsistent patterns of selection among natural wetlands.

Three out of ten egrets avoided shellfish aquaculture areas almost entirely and were excluded from formal modelling, a pattern that may suggest these three birds did not perceive the shellfish aquaculture areas as valuable foraging habitat. Similarly, shorebirds (Suborders Charadrii and Solopaci) at Tomales Bay generally avoid shellfish aquaculture areas, although Willet (Tringa semipalmata) appears an exception [52]. The three egrets that avoided shellfish aquaculture areas were captured at Cypress Grove, approximately 5 km from the Walker Creek delta where most of the shellfish aquaculture placed in egret-accessible depths is located on Tomales Bay (Fig 1). Most other birds were captured at Toms Point or Walker Creek, within 0.3–1.5 km of these main shellfish aquaculture areas. We cannot rule out that this apparent avoidance of shellfish aquaculture areas was due to some unrelated bay-wide segregation of foraging areas, and not directly due to avoidance of shellfish aquaculture. However, great egrets in the San Francisco Bay area regularly fly up to 10 km from colonies to forage [53], so reaching the shellfish aquaculture areas was well within the flight capabilities and foraging flight distances of egrets captured at all 3 trapping locations. Indeed, two of these Cypress Grove-captured birds (GREG_7 and GREG_11) repeatedly used areas around Walker Creek delta, but nevertheless still mostly avoided shellfish aquaculture areas. The fourth Cypress Grove-captured bird also regularly foraged around Walker Creek and did not avoid shellfish aquaculture areas.

We found that time- and location-specific changes in water depth (due to tidal action) were an important component in how great egret habitat selection varied between shellfish aquaculture areas and natural wetlands. Relative selection for shellfish aquaculture areas showed a strong quadratic response across the range of water depths we investigated, whereas selection for natural wetlands was generally constant across that range. Selection of shellfish aquaculture areas reached the strength of selection for natural wetlands only in a narrow depth range, but relative selection of shellfish aquaculture areas did not exceed that for natural wetlands at any water depth. Great egrets appear quite capable of ascertaining the relative benefits and costs of foraging in different areas that result from variance in prey density, prey capturability, and competition [16, 54, 55]. Generally, the density of foraging great egrets is greatest when water depths are between 20–40 cm [9]. Based on our results, shellfish aquaculture areas appear to provide foraging opportunities only in this narrow, preferred range of water depths, whereas natural habitats provide more diverse foraging opportunities across a broader range of water depths.

Although the four wetland types we considered span slightly different depth ranges at Tomales Bay, our results for habitat selection are not simply an artifact of aquaculture infrastructure occurring in a narrower, intermediate elevation range than the other wetland types (and thus experiencing a narrower range of water depths). We restricted our analysis to depths with sufficient representation among all wetland types, and the quadratic pattern of selection we observed was apparent across the range of depths that shellfish aquaculture areas experienced. Given the opportunity to forage in adjacent shellfish aquaculture areas or natural wetlands, the egrets in our study selected natural wetlands more strongly when water depths were greater than about 0.5 m and in areas that were exposed above the tide line. Because depths > 0.5 m are approaching the deepest water egrets regularly forage in [9], it is also important to recall our scale of inference when interpreting our results. Our water depth variable represented the average depth in the 10 m² area around where each GPS location was recorded, not the specific depth of water the egret was standing in. Where our models indicated greater selection for natural wetlands than shellfish aquaculture areas in water deeper than 0.5 m, we interpret these results to indicate that the egrets were more likely to probe the limits of their foraging depth (i.e., search for and use small areas of shallower water in deeper areas) in natural wetlands than in shellfish aquaculture areas. It may also be that the natural wetlands contained more small-scale heterogeneity in depths than did shellfish aquaculture areas.

Water depth, and especially temporal change in depth, is an important component of great egret foraging in other places it has been investigated. For instance, in the Everglades, great egrets select for areas where patterns of flooding and water drawdown operate at multiple spatiotemporal scales (e.g. daily and weekly tide patterns and seasonal climatic patterns) and act to concentrate prey [18, 56–58]. In tidal systems in southern Florida, time-integrated habitat availability (due to tidal cycles) was the resource attribute with the strongest effect on probability of use by wading birds across all habitats investigated [54]. The intertidal and shallow subtidal areas of Tomales Bay are characterized by subtle heterogeneity in the substrate surface, and egrets foraging above the tide line often seem to be focusing on small tidal puddles (pers. obs.). It is likely that shallow depressions in the intertidal areas serve as hydrologic refugia for egret prey during lower tides, and that egrets respond to the higher prey densities there. The quadratic pattern of selection for shellfish aquaculture areas may be driven by the interaction of prey density and prey availability. Although prey density may be greater in areas with thicker vegetation, a combination of intermediate vegetation thickness and shallower water depth may yield better prey capture by wading birds [16]. On Tomales Bay, these conditions may have been mimicked by shellfish aquaculture equipment as the tides rose and fell through it. The resulting shallow depths and moderate cover may have concentrated aquatic prey or made their escape responses slower or otherwise reduced, or some combination of both.

Tagged great egrets expended less energy and had shorter step lengths in tidal marsh than the other wetland types we considered. Egrets often roost in Tomales Bay tidal marsh areas between periods of active foraging (pers. obs.), as well as forage there, so these results may partially reflect this behavior. We found only slight differences in foraging behavior between natural wetlands and shellfish aquaculture areas. The cost of flying for great egrets has been estimated to be much greater than that of foraging [12]. If egrets are flushed more frequently in shellfish aquaculture areas during harvest or other maintenance activities, compared to flush rates in eelgrass, then the difference between energy acquisition and energy expenditure may differ between these wetland types despite the similar values of ODBA. Additionally, differences in prey type between shellfish aquaculture areas and natural wetlands (which we did not examine) may also lead to differences in energy acquisition despite similar energy expenditure. Based on our results we believe that a valuable next investigation would be to test the hypothesis that prey capture and energy acquisition are equal among shellfish aquaculture areas and natural wetlands.

Management implications

It appears that the current arrangement of eelgrass beds, tidal marsh, shellfish aquaculture areas, and mud flat provide a diversity of foraging opportunities to egrets across the tidal range. Although we did not directly quantify energy gain in each wetland type, our results suggest that shellfish aquaculture areas were perceived or experienced by these tagged great egrets as providing lower foraging quality than eelgrass or tidal marshes; however foraging studies explicitly addressing energetics are needed to evaluate this. Since eelgrass is federally designated in the U.S. as Essential Fish Habitat and a Habitat of Particular Concern, and managing agencies have adopted a “no net loss” policy [23], it is unlikely that any potential expansion of shellfish aquaculture in Tomales Bay will directly reduce availability of eelgrass to foraging great egrets. However, further conversion of unvegetated tidal areas in Tomales Bay for shellfish aquaculture may nevertheless reduce the amount of time that those areas provide suitable foraging opportunities for great egrets across the entire range of tidal depths. If this causes egrets to spend more time foraging in eelgrass, then this may change the nature of top down predation pressure in eelgrass systems [24].

Sea level rise associated with climate change may alter the overall foraging opportunities for great egrets on Tomales Bay, as both eelgrass and shellfish aquaculture are forced to migrate upslope. If eelgrass cannot migrate upslope to match the pace of sea level rise, the overall availability, and use, of eelgrass by egrets may decline. Thus, management actions that encourage upslope expansion of eelgrass beds to match sea level rise seem likely to benefit great egrets. In addition, such conservation efforts, which are likely to be critical in sustaining the substantial, broader conservation values of dwindling eelgrass beds [23], could conflict dramatically with any landward movement of intertidal shellfish growing areas needed to sustain the viability of shellfish aquaculture.

Supporting information

S1 Fig. Elevations for each wetland type.

Density of used and available points for each wetland type across the range of depths considered, for investigating habitat selection by GPS-tagged great egrets at Tomales Bay, CA, 2017–2020. Lines represent density of available and used points at each depth for each wetland type. Negative depth values indicate locations above the predicted water level.

(TIF)

Click here for additional data file.^{(809.3KB, tif)}

S1 Table. Habitat selection model selection.

Model selection results for evaluating differences in foraging habitat selection among wetland habitat types, while accounting for water depth, by GPS-tagged great egrets at Tomales Bay, CA, 2017–2020. K is the number of parameters, Δ AICc is the difference in AICc value between the top model and the current model, and AICc Wt. is the AICc model weight.

(DOCX)

Click here for additional data file.^{(22.7KB, docx)}

S2 Table. Step length model selection.

Model selection results for evaluating differences in foraging step length among wetland habitat types, while accounting for water depth, by GPS-tagged great egrets at Tomales Bay, CA, 2017–2020. K is the number of parameters, Δ AICc is the difference in AICc value between the top model and the current model, and AICc Wt. is the AICc model weight.

(DOCX)

Click here for additional data file.^{(18.7KB, docx)}

Acknowledgments

We thank Gary Fleener and John Finger of Hog Island Oyster Company for providing information about shellfish aquaculture practices, Cassidy Teufel of the California Coastal Commission for information on regulations regarding shellfish aquaculture placement and eelgrass, Richard James (coastodian.org) for photo use, and John Fieberg, Tal Avgar, Brian Smith and Johannes Signer for advice on analysis methods. We thank Drs. John Brzorad and Alan Maccarone for training in egret capture and tagging, and Ginny Fifield, Mark McCaustland, Richard James, Sandra Hunt-von Arb, Libby Porzig and Barbara Wechsberg for assistance with field work.

Data Availability

Tagging data are archived here: https://www.movebank.org/cms/webapp?gwt_fragment=page=studies,path=study247850178.

Funding Statement

Funding came from generous individual donors to Audubon Canyon Ranch and through the continued support of its Board of Directors. None of our funders had any influence on the content of the manuscript.

References

1.Van Moorter B, Rolandsen CM, Basille M, Gaillard J-MM. Movement is the glue connecting home ranges and habitat selection. J Anim Ecol. 2016;85: 21–31. doi: 10.1111/1365-2656.12394 [DOI] [PubMed] [Google Scholar]
2.Calle L, Gawlik DE, Xie Z, Green L, Lapointe B, Strong A, et al. Effects of tidal periodicities and diurnal foraging constraints on the density of foraging wading birds. Auk. 2016;133: 378–396. doi: 10.1642/auk-15-234.1 [DOI] [Google Scholar]
3.Norevik G, Åkesson S, Andersson A, Bäckman J, Hedenström A. The lunar cycle drives migration of a nocturnal bird. Chittka L, editor. PLOS Biol. 2019;17: e3000456. doi: 10.1371/journal.pbio.3000456 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Cruz SM, Hooten M, Huyvaert KP, Proaño CB, Anderson DJ, Afanasyev V, et al. At–Sea Behavior Varies with Lunar Phase in a Nocturnal Pelagic Seabird, the Swallow-Tailed Gull. Ropert-Coudert Y, editor. PLoS One. 2013;8: e56889. doi: 10.1371/journal.pone.0056889 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Tucker MA, Böhning-Gaese K, Fagan WF, Fryxell JM, Van Moorter B, Alberts SC, et al. Moving in the Anthropocene: Global reductions in terrestrial mammalian movements. Science (80-). 2018;359: 466–469. doi: 10.1126/science.aam9712 [DOI] [PubMed] [Google Scholar]
6.Teitelbaum CS, Hepinstall-Cymerman J, Kidd-Weaver A, Hernandez SM, Altizer S, Hall RJ. Urban specialization reduces habitat connectivity by a highly mobile wading bird. Mov Ecol. 2020;8: 49. doi: 10.1186/s40462-020-00233-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Doherty TS, Driscoll DA. Coupling movement and landscape ecology for animal conservation in production landscapes. Proc R Soc B Biol Sci. 2018;285. doi: 10.1098/rspb.2017.2272 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Allen AM, Singh NJ. Linking movement ecology with wildlife management and conservation. Front Ecol Evol. 2016;3: 1–13. doi: 10.3389/fevo.2015.00155 [DOI] [Google Scholar]
9.McCrimmon DAJ, Ogden JC, Bancroft GT, Martínez-Vilalta A, Motis A, Kirwan GM, et al. Great Egret (Ardea alba), version 1.0. In: Billerman SM, editor. Birds of the World. Ithaca, New York, USA: Cornell Lab of Ornitholog; 2020. [Google Scholar]
10.Maccarone AD, Brzorad JN. Foraging behavior and energetics of Great Egrets and Snowy Egrets at interior rivers and weirs. J F Ornithol. 2007;78: 411–419. doi: 10.1111/j.1557-9263.2007.00133.x [DOI] [Google Scholar]
11.Maccarone AD, Brzorad JN. Foraging Patterns of Breeding Egrets at Coastal and Interior Locations. Waterbirds. 2002;25. doi: 10.1675/1524-4695(2002)025[0001:FPOBEA]2.0.CO;2 [DOI] [Google Scholar]
12.Brzorad JN, Maccarone AD, Conley KJ. Foraging energetics of Great Egrets and Snowy Egrets. J F Ornithol. 2004;75: 266–280. doi: 10.1648/0273-8570-75.3.266 [DOI] [Google Scholar]
13.Burr PC, Avery JL, Street GM, Strickland BK, Dorr BS. Historic and contemporary use of catfish aquaculture by piscivorous birds in the Mississippi Delta. Condor. 2020;122. doi: 10.1093/condor/duaa009 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Gorenzel WP, Conte FS, Salmon TP. Bird damage at aquaculture facilities. Handb Prev Control Wildl Damage. 1994;57. Available: https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1055&context=icwdmhandbook [Google Scholar]
15.Dumbauld BR, Ruesink JL, Rumrill SS. The ecological role of bivalve shellfish aquaculture in the estuarine environment: A review with application to oyster and clam culture in West Coast (USA) estuaries. Aquaculture. 2009;290: 196–223. doi: 10.1016/j.aquaculture.2009.02.033 [DOI] [Google Scholar]
16.Lantz SM, Gawlik DE, Cook MI. The effects of water depth and emergent vegetation on foraging success and habitat selection of wading birds in the Everglades. Waterbirds. 2011;34: 439–447. doi: 10.1675/063.034.0406 [DOI] [Google Scholar]
17.California Department of Fish and Wildlife. Tomales Bay shellfish production, unpublished data.
18.Custer T, Osborn R. Feeding habitat use by colonially-breeding herons, egrets, and ibises in North Carolina. Auk. 1978;95: 733–743. doi: 10.1093/auk/95.4.733 [DOI] [Google Scholar]
19.Hughes BB, Levey MD, Brown JA, Fountain MC, Carlisle AB, Litvin SY, et al. Nursery functions of U.S. West Coast estuaries: the state of knowledge for juveniles of invertebrate and fish species. The Nature Conservancy, Arlington, VA; 2014. p. 168. Available: http://research.pbsci.ucsc.edu/eeb/bbhughes/wp-content/uploads/2014/12/Hughes_etal_2014_NurseryFunction_TNC.pdf [Google Scholar]
20.Plummer ML, Harvey CJ, Anderson LE, Guerry AD, Ruckelshaus MH. The role of eelgrass in marine community interactions and ecosystem services: results from ecosystem-scale food web models. Ecosystems. 2013;16: 237–251. doi: 10.1007/s10021-012-9609-0 [DOI] [Google Scholar]
21.Röhr ME, Holmer M, Baum JK, Björk M, Boyer K, Chin D, et al. Blue carbon storage capacity of temperate eelgrass (Zostera marina) meadows. Global Biogeochem Cycles. 2018;32: 1457–1475. doi: 10.1029/2018GB005941 [DOI] [Google Scholar]
22.Zarnoch CB, Hoellein TJ, Furman BT, Peterson BJ. Eelgrass meadows, Zostera marina (L.), facilitate the ecosystem service of nitrogen removal during simulated nutrient pulses in Shinnecock Bay, New York, USA. Mar Pollut Bull. 2017;124: 376–387. doi: 10.1016/j.marpolbul.2017.07.061 [DOI] [PubMed] [Google Scholar]
23.National Oceanic and Atmospheric Association. West Coast Region California Eelgrass Mitigation Policy and Implementing Guidelines. 2014. Available: https://www.fisheries.noaa.gov/resource/document/california-eelgrass-mitigation-policy-and-implementing-guidelines
24.Huang AC, Essak M, O’Connor MI. Top-down control by great blue herons Ardea herodias regulates seagrass-associated epifauna. Oikos. 2015;124: 1492–1501. doi: 10.1111/oik.01988 [DOI] [Google Scholar]
25.Hughes BB, Eby R, Dyke E Van, Tinker MT, Marks CI, Johnson KS, et al. Recovery of a top predator mediates negative eutrophic effects on seagrass. PNAS. 2013;110: 15313–15318. doi: 10.1073/pnas.1302805110 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.San Francisco Estuary Institute. California Aquatic Resource Inventory (CARI) version 0.3. 2017. Available: https://www.sfei.org/data/california-aquatic-resource-inventory-cari-version-03-gis-data#sthash.K4v5RIFd.dpbs
27.Fair JM, Paul E, Jones J, Clark AB, Davie C, Kaiser G. Guidelines to the use of wild birds in research. Ornithological Council; 2010. Available: http://www.nmnh.si.edu/BIRDNET [Google Scholar]
28.Brzorad JN, Maccarone AD. An inexpensive technique for capturing gregarious wading birds on their foraging grounds. Waterbirds. 2014;37: 335–339. doi: 10.1675/063.037.0312 [DOI] [Google Scholar]
29.San Francisco Estuary Institute. Eelgrass Survey GIS Data version 2.0. 2017. Available: https://www.sfei.org/data/eelgrass-survey-gis-data#sthash.eBKfqYBx.dpbs
30.Marin County. Marin County ortho images. 2018 [cited 6 Nov 2020]. Available: https://gis.marinpublic.com/arcgis/rest/services/Orthos/2018Orthos/ImageServer
31.Marin County. Marin Map LIDAR DEM version 10.71. 2018. Available: https://gis.marinpublic.com/arcgis/rest/services/LIDAR
32.National Centers for Environmental Information N. NOAA NOS Estuarine Bathymetry—Tomales Bay. 2018 [cited 6 Nov 2020]. Available: https://data.noaa.gov/metaview/page?xml=NOAA/NESDIS/NGDC/MGG/DEM/iso/xml/tomales_bay_P110_2018.xml&view=rubricv2/recordHTML&header=none
33.Qasem L, Cardew A, Wilson A, Griffiths I, Halsey LG, Shepard ELC, et al. Tri-axial dynamic acceleration as a proxy for animal energy expenditure; should we be summing values or calculating the vector? Ropert-Coudert Y, editor. PLoS One. 2012;7: e31187. doi: 10.1371/journal.pone.0031187 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Wilson RP, White CR, Quintana F, Halsey LG, Liebsch N, Martin GR, et al. Moving towards acceleration for estimates of activity-specific metabolic rate in free-living animals: the case of the cormorant. J Anim Ecol. 2006;75: 1081–1090. doi: 10.1111/j.1365-2656.2006.01127.x [DOI] [PubMed] [Google Scholar]
35.Gleiss AC, Wilson RP, Shepard ELCC. Making overall dynamic body acceleration work: On the theory of acceleration as a proxy for energy expenditure. Methods Ecol Evol. 2011;2: 23–33. doi: 10.1111/j.2041-210X.2010.00057.x [DOI] [Google Scholar]
36.Green JA, Halsey LG, Wilson RP, Frappell PB. Estimating energy expenditure of animals using the accelerometry technique: activity, inactivity and comparison with the heart-rate technique. J Exp Biol. 2009;212: 471–82. doi: 10.1242/jeb.026377 [DOI] [PubMed] [Google Scholar]
37.Manly BFJ, McDonald LL, Thomas DL, McDonald TL, Erickson WP. Resource selection by animals: statistical design and analysis for field studies. Kluwer Academic Publishers; 2002. [Google Scholar]
38.Paton RS, Mattiopoulos J. Defining the scale of habitat availability for models of habitat selection. Ecology. 2016;97: 1113–1122. doi: 10.1890/14-2241.1 [DOI] [PubMed] [Google Scholar]
39.Avgar T, Potts JR, Lewis MA, Boyce MS. Integrated step selection analysis: bridging the gap between resource selection and animal movement. Börger L, editor. Methods Ecol Evol. 2016;7: 619–630. doi: 10.1111/2041-210X.12528 [DOI] [Google Scholar]
40.Muff S, Signer J, Fieberg J. Accounting for individual-specific variation in habitat-selection studies: Efficient estimation of mixed-effects models using Bayesian or frequentist computation. J Anim Ecol. 2020;89: 80–92. doi: 10.1111/1365-2656.13087 [DOI] [PubMed] [Google Scholar]
41.Ranacher P, Brunauer R, Trutschnig W, Der S Van, Reich S, Ranacher P, et al. Why GPS makes distances bigger than they are. Int J Geogr Inf Sci. 2016;30: 316–333. doi: 10.1080/13658816.2015.1086924 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Noonan MJ, Fleming CH, Akre TS, Drescher-Lehman J, Gurarie E, Harrison AL, et al. Scale-insensitive estimation of speed and distance traveled from animal tracking data. Mov Ecol. 2019;7: 1–15. doi: 10.1186/s40462-019-0147-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Johnson DH. The comparison of usage and availability measurements for evaluating resource preference. Ecology. 1980;61: 65–71. doi: 10.2307/1937156 [DOI] [Google Scholar]
44.Signer J, Fieberg J, Avgar T. Animal movement tools (amt): R package for managing tracking data and conducting habitat selection analyses. Ecol Evol. 2019;9: 880–890. doi: 10.1002/ece3.4823 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Burnham KP, Anderson DR. Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. New York: Springer-Verlag; 2002. [Google Scholar]
46.Fieberg J, Signer J, Smith B, Avgar T. A “how-to” guide for interpreting parameters in resource-and step-selection analyses. bioRxiv. 2020; 2020.11.12.379834. Available: 10.1101/2020.11.12.379834 [DOI] [Google Scholar]
47.Zuur AF, Ieno EN, Walker N, Saveliev AA, Smith GM. Mixed effects models and extensions in ecology with R. 2009. [Google Scholar]
48.ESRI. ArcGIS Pro. Redlands, CA: Environmental Systems Research Institute; 2021. [Google Scholar]
49.R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2019. Available: https://www.r-project.org/ [Google Scholar]
50.Kelley D, Richard C. oce: Analysis of Oceanographic Data. 2020. Available: https://cran.r-project.org/package=oce [Google Scholar]
51.Bates D, Maechler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67: 1–48. doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]
52.Kelly JP, Evens JG, Stallcup RW, Wimpfheimer D. Effects of aquaculture on habitat use by wintering shorebirds. Calif Fish Game. 1996;82: 160–174. Available: https://egret.org/sites/default/files/scientific_contributions/kelly_etal_1996_aquaculture.pdf [Google Scholar]
53.Kelly JP, Stralberg D, Etienne K, Mccaustland M. Landscape influence on the quality of heron and egret colony sites. Wetlands. 2008;28: 257–275. doi: 10.1672/07-152.1 [DOI] [Google Scholar]
54.Calle L, Green L, Strong A, Gawlik DE. Time-integrated habitat availability is a resource attribute that informs patterns of use in intertidal areas. Ecol Monogr. 2018;88: 600–620. doi: 10.1002/ecm.1305 [DOI] [Google Scholar]
55.Stolen ED, Collazo JA, Percival HF. Group-foraging effects on capture rate in wading birds. Condor. 2012;114: 744–754. doi: 10.1525/cond.2012.110159 [DOI] [Google Scholar]
56.Calle L, Gawlik DE, Xie Z, Green L, Lapointe B, Strong A. Effects of tidal periodicities and diurnal foraging constraints on the density of foraging wading birds. Auk. 2016;133: 378–396. doi: 10.1642/auk-15-234.1 [DOI] [Google Scholar]
57.Beerens JM, Noonburg EG, Gawlik DE. Linking dynamic habitat selection with wading bird foraging distributions across resource gradients. PLoS One. 2015;10: e0128182. doi: 10.1371/journal.pone.0128182 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Botson BA, Gawlik DE, Trexler JC. Mechanisms that generate resource pulses in a fluctuating wetland. PLoS One. 2016;11: e0158864. doi: 10.1371/journal.pone.0158864 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0261963.r001

Decision Letter 0

Vanesa Magar

21 Jul 2021

PONE-D-21-19522

Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: comparing natural wetlands to areas with shellfish aquaculture

PLOS ONE

Dear Dr. Jennings,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Sep 04 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Vanesa Magar, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

3. In your Methods section, please provide additional location information of the study site, including geographic coordinates for the data set if available.

4. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

5. Thank you for stating the following in the Acknowledgments Section of your manuscript:

"Funding came from generous individual donors to Audubon Canyon Ranch and through the continued support of its Board of Directors. None of our funders had any influence on the content of the manuscript."

We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

6. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript.

7. We note that Figure(s) 1 in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure(s) 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b. If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Line 22 insert comma after marsh

Line 79 need space between [16] and comma

Line 116 labeled with

Line 117 eelgrass shown in dark gray and , shellfish….

Line 137 Does Publicly have to be capitalized. Hyphen is not necessary

Line 138 end sentence after ...our analysis.

Line 139 Capital W to start line

Line 151 remove a after in GIS….

Line 182 I did not find that DEM was defined prior to this sentence, will your reader know what this is?

Line 185 end sentence after collected

Line 186 start line with capital T

Line 198 add space between ) and [5]

Line 206 delete dur to

Line 234 add space between ) and [33]

Line 245 What were an example of the observations of foraging behavior that prompted you to select this scale, an explanation would help the reader.

Line 248 what is a small number, state how many models, be specific

Line 252-254 you chose to evaluate wetland type at the start or end of the step. Use this same order in the next sentence

Line 256 than for natural wetlands

Line 261 … we deemed a quadratic response to be the most biologically likely… based on what? How did you determine this approach was the most biologically likely?

Line 261-262 The authors decided to use quadratic functions as described in the methods section without further justification using the literature or some preliminary analysis. The relationship with other three wetland types appeared to be linear based on Fig. 2.

Line 264 add space between ) and [37]

Line 301 Table 1 do you have an explanation as to the egrets that quit (?) sending data in 2018 and 2019? Did they leave the area due to migration? If so, did they revisit the area? Did you have any transmitter failure? Did you have mortality?

Lines 320-332 To better support and clarify the statements, a box plot of tidal water depth for four wetland types (maybe as a supplementary figure) would be useful.

Line 331-332 … natural wetlands than for shellfish aquaculture areas across most water depths… define most, number, percentage? Hard to determine how important this is.

Line 372 delete they

Line 393 aquaculture areas were well within …

Line 459 insert comma after marsh

Overview:

This is an important study because there has been little attention to the relationships between shellfish aquaculture and wading-bird habitat selection during the past. An interesting part of this study is the quantification of foraging activities with movement speed and activity level (with accelerometer data). Their results of the analyses look fine from their reports of AICc values and Figures. However, the quadratic curve is only obvious with shellfish aquaculture in Figure 2. As the tidal water depth increased between 0 and 1, relative selection strength converged among the four types of wetlands (Fig. 2). The authors decided to use quadratic functions as described in the method section without further justification using the literature or some preliminary analysis (L.261-262). The relationship with other three wetland types appeared to be linear based on Fig. 2. To better support and clarify the statements between lines 320-332, a box plot of tidal water depth for the four wetland types (maybe as a supplementary figure) would be useful.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Dec 31;16(12):e0261963. doi: 10.1371/journal.pone.0261963.r002

Author response to Decision Letter 0

4 Nov 2021

Dear Dr. Magar,

Thank you very much for the helpful comments on our manuscript titled “Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: comparing natural wetlands to areas with shellfish aquaculture”. We appreciate the feedback from you and the reviewer and feel that collectively it made our paper stronger. Attached please find a marked up and unmarked versions of our paper, all figure and supporting information files, and a text version of an email correspondence regarding rights to use map material.

We have addressed all comments and suggestions by you and the reviewer (see red below). In nearly every instance we made the recommended change but see below for justification where we did not make the recommended change. The most substantial change we made is the inclusion of 3 additional models in our candidate set for evaluating habitat selection. The original best-supported model is still the best supported and remains the only model we base inference on for that part of the paper. However, we appreciate the reviewer’s comments and feel that the larger candidate set provides a fuller evaluation of the patterns we observed.

Thank you very much for your time and for considering our manuscript for publication.

Scott Jennings

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

We have reviewed our submission against the style requirements. To the best of our knowledge our submission meets all requirements.

We have reviewed our reference list. We note that in the original submission we cited unpublished data from the California Department of Fish and Wildlife as (CDFW, unpublished data). We have replaced this text with an inline citation number following instructions for citations, and in our reference list for the appropriate inline number we have added “California Department of Fish and Wildlife. Tomales Bay shellfish production, unpublished data.”

3. In your Methods section, please provide additional location information of the study site, including geographic coordinates for the data set if available. We have added coordinates

4. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section. We received no grants to conduct this work.

5. Thank you for stating the following in the Acknowledgments Section of your manuscript:

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

Please keep our funding statement as it currently reads.

We have contacted ESRI regarding copyright status of material included in our figure 1. According to the email correspondence copied to the end of this document as well as the following webpage (https://doc.arcgis.com/en/arcgis-online/reference/static-maps.htm), we feel that we have sufficient permission to use these materials. We have added attribution for these materials to our figure caption.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figure(s) 1 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

________________________________________

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

________________________________________

3. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: Yes

________________________________________

4. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

________________________________________

5. Review Comments to the Author

Reviewer #1: Line 22 insert comma after marsh added

Line 79 need space between [16] and comma It is unclear whether the reviewer is referring to the comma inside the brackets or outside. If outside, we believe there shouldn’t be a space before the comma. If inside, the Plos One style template linked above indicates there should be no spaces separating multiple citations inside brackets.

Line 116 labeled with we have changed the wording of this sentence to be more clear.

Line 117 eelgrass shown in dark gray and , shellfish…. We disagree that a comma should be added here. We have eliminated the semi-colon to make this two sentences for ease of reading.

Line 137 Does Publicly have to be capitalized. Hyphen is not necessary hyphen removed; we do not believe Publicly should be capitalized.

Line 138 end sentence after ...our analysis. Done

Line 139 Capital W to start line Done

Line 151 remove a after in GIS…. We disagree that this “a” should be removed. We are referring here to a single object.

Line 182 I did not find that DEM was defined prior to this sentence, will your reader know what this is? We have added “(DEM)” at line 165 where this term is first used.

Line 185 end sentence after collected Done

Line 186 start line with capital T Done

Line 198 add space between ) and [5] Done

Line 206 delete dur to Done

Line 228 “third order selection” needs to be better defined if the concept of Johnson’s hierarchical habitat selection is used. This term has not been described either in the introduction or methods prior to this point. We have added some text in parentheses to this sentence clarifying the distinction between first, second and third order resource selection. We feel that the concept of hierarchical resource selection is well enough understood and accepted in ecological research (the Johnson 1980 paper we cite has >3950 citations according to Google scholar) that further explanation is not warranted here. If the editor disagrees with this conclusion, we can add more text, but in the interest of maintaining overall readability of this paper we feel the current text is sufficient. Note also that we have moved this sentence to the start of the following paragraph where we feel it fits better.

L.232-233 “evaluate fine scale habitat selection but avoid an excessive number of steps with no movement” – this data thinning may result in bias: “stay-put” during foraging may suggest selection, and more residence time. You may need movement steps for non-zero step selection functions; however, the more the locations you filtered out, the more the information of preference (staying there longer) you may lose. Caution needs to be exercised here. This is a fair point; however, it should be noted that any sampling frequency that researchers select may yield different results than other possible sampling frequencies. Our filtering was done to avoid GPS measurement error being represented in our step length values, though we appreciate that this needed to be better explained. As demonstrated by Ranacher (2016), GPS measurement error can lead to systematic overestimation of movement distances when sampling intervals are short relative to movement speeds. This can be particularly problematic as sampling interval decreases to a degree that the step length over that interval approaches the GPS tag measurement error (Noonan 2019). Our preliminary data summarization revealed that median step lengths at the 5 minute sampling interval were of similar magnitude as the tag-estimated measurement error (3-10m; see figure below), whereas at the 10 minute interval median step lengths were ≥3-4 times longer than tag error. Thus, our use of the 10 minute interval represents the shortest interval that would yield reliable estimates of movement distances, given the inherent measurement error of the GPS tags we used. We have expanded the description of our “preliminary data summarization” and subsequent rationale for choosing 10 minutes here.

Noonan MJ, Fleming CH, Akre TS, Drescher-Lehman J, Gurarie E, Harrison AL, et al. Scale-insensitive estimation of speed and distance traveled from animal tracking data. Mov Ecol. 2019;7: 1–15. doi:10.1186/s40462-019-0177-1

Ranacher P, Brunauer R, Trutschnig W, Der S Van, Reich S, Ranacher P, et al. Why GPS makes distances bigger than they are. Int J Geogr Inf Sci. 2016;30: 316–333. doi:10.1080/13658816.2015.1086924

Line 234 add space between ) and [33] Done

Line 245 What were an example of the observations of foraging behavior that prompted you to select this scale, an explanation would help the reader. We have deleted the text referring to foraging behavior and habitat structure from this sentence. We believe that basing our habitat raster resolution on the accuracy of our GPS tags is sufficient justification.

Line 248 what is a small number, state how many models, be specific Done

L. 252-253 The time-interval that you “chose” may affect the model outcomes because birds may choose to stay longer with shorter step length if wetland types are desirable; however, a part of this information has been thrown out by filtering the data. A sensitivity analysis as the authors mentioned (L. 246-247) is critical. See our response to the reviewer’s similar comment at line 232-233. We believe that the suggested sensitivity analysis is not required, but if the editor feels it is necessary we can conduct it.

Line 252-254 you chose to evaluate wetland type at the start or end of the step. Use this same order in the next sentence We have restructured this paragraph slightly to be more clear about which models were used for which part of our analysis. This suggestion is no longer relevant.

Line 256 than for natural wetlands added “for”

Line 257 “interactions between the 4-level wetland type variable and time-specific water depth”: have you checked whether the wetland types and water depth are correlated before they are included in the same model? From a strictly statistical perspective, one cannot calculate a Pearson correlation between continuous and categorical variables. However, the reviewer makes a good point here and on their comment for lines 320-332. Thus, we have added a plot as supplementary material showing the distribution of elevations for each wetland type. We feel that the ranges of elevations for each wetland type in our original analysis overlap sufficiently for our comparison to be valid. Nevertheless, because shellfish aquaculture wetland type is of primary interest in our analysis and because there were relatively few shellfish aquaculture raster cells with depths < -0.5 m (i.e., 0.5 m above the tidal level), we also restricted our analysis of habitat selection to depths ≥ -0.5 m to minimize any effect of this data sparseness on our results. This data filtering added some nuance to the comparison of selection among natural wetlands but does not change the observed patterns or ecological conclusions about shellfish aquaculture. We have edited the text here in Methods, as well as in Results and Discussion to reflect this change.

Line 261 … we deemed a quadratic response to be the most biologically likely… based on what? How did you determine this approach was the most biologically likely? These are fair questions. We mostly based this on our observations of egrets foraging in the study system through the tidal cycle, and what is generally known about Great Egret foraging water depth preferences. However, we appreciate that this rationale may not satisfy the editor or readers, and in the interest of conducting a more complete investigation we have added 3 models to the original candidate set of 4 models. These new models investigate linear effects of water depth: wetland type * depth; wetland type + depth; depth. As can be seen in our new S1 Table, these linear depth models were very poorly supported by the data, with Delta AICc values between 295-1766. Thus, while we now provide model selection results for the full 7 model candidate set, our results and inference are based on the same models as in the original submission. We have edited the text in this section of the manuscript to reflect this new candidate model set.

Line 264 add space between ) and [37] Done

Lines 320-332 To better support and clarify the statements, a box plot of tidal water depth for four wetland types (maybe as a supplementary figure) would be useful. This is a good suggestion, we have added a plot to show the distribution of elevations for each wetland type as a supplementary figure.

Line 331-332 … natural wetlands than for shellfish aquaculture areas across most water depths… define most, number, percentage? Hard to determine how important this is. We have deleted this sentence, as it repeats what was written earlier in the paragraph. We believe that the text earlier in this paragraph provides sufficient detail to understand our results.

Line 372 delete they Done

Line 393 aquaculture areas were well within … We disagree with this word change (“was” to “were”). The noun in this sentence is the collective ability of all tagged egrets to fly a certain distance, which is singular. Thus this sentence employs third person past tense, singular, in which case “was” is the correct form of “to be”.

Lines 398-399 “…showed a strong quadratic response across the range of water depths we investigated,…” Does this suggest your model only fit data on shellfish aquaculture well, failing on the other three wetland types? No. A quadratic predictor can fit the response variable just as well as a linear predictor (by fitting a very small value to the squared term); in such cases the quadratic predictor won’t “fail”, it will simply be penalized in AIC model selection for having an extra parameter. Because we included the interaction term between water depth and wetland type in our models, we allowed the form of the quadratic water depth response to be fit separately to each wetland type. This interaction term allowed the maximum likelihood procedure to find coefficient values that optimized the fitted values for each wetland types separately, rather than finding a single coefficient value that adequately fit all wetland types. The reviewer’s concern would be correct had we only fit the model with the additive wetland type and water depth terms (“wetland type + depth2” in S1 Table). Indeed, the very large Delta AICc values for this model among all birds (S1 Table) provide strong evidence that resource selection takes a substantially different form in each wetland type. We have added text to the third paragraph of the Results (lines 316-317) clarifying this. See also our reply to the reviewer’s comment for line 261.

Line 459 insert comma after marsh Done

Overview:

The actual description of their statistical analysis is good, but my main concerns are twofold. 1) sample size of seven birds is too small for a reliable inference even when analysis is by individual bird; and 2) data filtering may have affected the results and conclusions. We have added cautionary text to the first paragraph of our Discussion, noting the small sample size and the descriptive rather than predictive nature of our results. We have also further justified our data filtering to describe how we used the finest temporal scale that was possible given the inherent error of our GPS tags.

Email correspondence with ESRI regarding map copyrights:

From: Emiko Condeso

Sent: Thursday, August 12, 2021 9:05 PM

To: Scott Jennings

Subject: FW: Esri Case #02864788 - Proper attribution for a map made with data from the Living

Atlas

Hi Scott,

See below for the answer from ESRI regarding publication permission for their data layers. I actually

think this link might more explicitly show that ESRI permits this use:

https://doc.arcgis.com/en/arcgis-online/reference/static-maps.htmI

The attribution needed for the ESRI terrain layer is:

Source: Airbus, USGS, NGA, NASA, CGIAR, NLS, OS, NMA, Geodatastyrelsen, GSA, GSI

and the GIS User Community

Attribution for the state boundary layer is:

Sources: Esri, TomTom, U.S. Department of Commerce, U.S. Census Bureau

Also included in the map that would need attribution are:

Eelgrass: SFEI (I emailed SFEI to check for proper attribution on maps, in case they want us to cite the

local Tomales mappers rather than them, but no answer), county boundary layer: California Department

of Forestry and Fire Protection, shellfish gear: ACR.

Cheers,

Emi

T. Emiko Condeso

Ecologist/GIS Specialist, Cypress Grove Research Center

My pandemic work week is M-F, 3-5pm.

Due to my family/work balance you might receive messages

from me outside normal working hours. Please do not feel

pressured to reply outside your usual work schedule.

Audubon Canyon Ranch

P.O. Box 808, Marshall, CA 94940

P: 415-663-8203 ext. 401 | C: 707-364-3274

emiko.condeso@egret.org | WEB: egret.org

Connecting nature, people and science in a rapidly changing world

From: Esri Customer Care <customercare@esri.com>

Sent: Monday, August 9, 2021 2:10 PM

To: Emiko Condeso <emiko.condeso@egret.org>

Subject: Esri Case #02864788 - Proper attribution for a map made with data from the Living Atlas

Hello Emiko,

Thank you for replying to my email. To answer your question, yes they are (provided that proper credits

as listed in the Item Details page are given).

I hope this helps! Other than that, the link below should provide further clarification:

* Terms of use for items: https://doc.arcgis.com/en/arcgis-online/reference/access-use-

constraints.htm

Please feel free to let me know if you have any additional questions. Looking forward to hearing from

you soon!

Regards,

Dalilah

Esri Support Services

https://my.esri.com/#/support/cases/02864788

https://support.esri.com/en/

THE SCIENCE OF WHERE™

ref:_00D70JXts._5005x1dXQYm:ref

--------------- Original Message ---------------

From: Emiko Condeso [emiko.condeso@egret.org]

Sent: 07/08/2021 10:56

To:

Subject: RE: Esri Case #02864788 - Proper attribution for a map made with data from the Living Atlas

Original Recipients:

To: Esri Customer Care <customercare@esri.com>,

Cc: []

Thank you for this information. Are you telling me that the data are free to publish under a Creative

Commons Attribution License, as long as I use the appropriate attribution? I need written permission

for the publisher.

Thank you,

Emiko

T. Emiko Condeso

Ecologist/GIS Specialist, Cypress Grove

Research Center

My pandemic work week is M-F, 3-5pm.

Due to my family/work balance you

might receive messages from me outside

normal working hours. Please do not feel

pressured to reply outside your usual

work schedule.

image001.jpg

[egret.org]

Audubon Canyon Ranch

P.O. Box 808, Marshall, CA 94940

P: 415-663-8203 ext. 401 | C: 707-364-

3274

emiko.condeso@egret.org | WEB:

egret.org [egret.org]

image002.png

[facebook.com]

image003.png

[instagram.com]

image004.png

[twitter.com]

image005.png

[vimeo.com]

Connecting nature, people and science in a rapidly changing world

image006.png

From: Esri Customer Care <customercare@esri.com>

Sent: Friday, August 6, 2021 8:22 AM

To: Emiko Condeso <emiko.condeso@egret.org>

Subject: Esri Case #02864788 - Proper attribution for a map made with data from the Living Atlas

Hello Emiko,

Thank you for being patient while we are looking into this query. For your information, the Credits

(Attribution) section in the data's Item Details page is where users can find the appropriate attribution

for credits and appreciation section if they are using the data in their journals. Users should always

credit the sources found in the Credits field of the item page for an ArcGIS Online map. Example below is

the Credits (attribution) section for the multi-directional hillshade data:

ima

ge0

07.p

You may navigate to this link to view the Item Details page of the data and scroll down on the page until

you find a Credits (Attribution) section. Since the section already listed several organizations that should

be credited, you may use the list in the section to give credits to the organizations.

I hope this helps! Please let me know if you have any additional questions. Looking forward to hearing

from you soon.

Thank you and keep safe.

Regards,

Dalilah

Esri Support Services

https://my.esri.com/#/support/cases/02864788

https://support.esri.com/en/

THE SCIENCE OF WHERE™

ref:_00D70JXts._5005x1dXQYm:ref

--------------- Original Message ---------------

From: Emiko Condeso [emiko.condeso@egret.org]

Sent: 06/08/2021 08:56

To:

Subject: RE: Esri Case #02864788 - Proper attribution for a map made with data from the Living Atlas

Original Recipients:

To: Esri Customer Care <customercare@esri.com>,

Cc: []

Thank you!

T. Emiko Condeso

Ecologist/GIS Specialist, Cypress Grove Research Center

My pandemic work week is M-F, 3-5pm.

Due to my family/work balance you might receive

messages from me outside normal working hours. Please do

not feel pressured to reply outside your usual work

schedule.

ima

ge0

01.j

[egret.org]

Audubon Canyon Ranch

P.O. Box 808, Marshall, CA 94940

P: 415-663-8203 ext. 401 | C: 707-364-3274

emiko.condeso@egret.org | WEB: egret.org

[egret.org]

ima

ge0

02.p

[facebook.com]

ima

ge0

03.p

[instagram.com]

ima

ge0

04.p

[twitter.com]

ima

ge0

05.p

[vimeo.com]

Connecting nature, people and science in a rapidly changing world

ima

ge0

06.p

From: Esri Customer Care <customercare@esri.com>

Sent: Thursday, August 5, 2021 4:42 PM

To: Emiko Condeso <emiko.condeso@egret.org>

Subject: Esri Case #02864788 - Proper attribution for a map made with data from the Living Atlas

Hello Emiko,

Warm greetings and good day to you.

This is Dalilah from ESRI Support Services, writing in reference to the case #02864788. I have taken

ownership of the case and will be working with you to reach a resolution.

As per notification received, I understand that you have some queries about proper attribution for a

map made with data from the Living Atlas, mainly whether these data are available to freely publish

with proper attribution. I also understand that the map will be published by a journal under Creative

Commons Attribution License, which requires written permission by the copywrite holder for

publication.

I would appreciate if you could allow me some time to perform adequate research and to consult this

internally with my team before getting back to you with more information. I appreciate your patience

while this process takes place.

In the meantime, have a good rest of your day and keep safe!

Regards,

Dalilah

Esri Support Services

https://my.esri.com/#/support/cases/02864788

https://support.esri.com/en/

THE SCIENCE OF WHERE™

ref:_00D70JXts._5005x1dXQYm:ref

Learn more about the Esri Support [play.google.com] [itunes.apple.com]

Learn more about the Esri Support

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(61.4KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0261963.r003

Decision Letter 1

Hans-Uwe Dahms

15 Dec 2021

Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: comparing natural wetlands to areas with shellfish aquaculture

PONE-D-21-19522R1

Dear Dr. JENNINGS,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Hans-Uwe Dahms, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

ACCEPTED

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #2: No

**********

6. Review Comments to the Author

Reviewer #2: Main comments:

1. The reason why the shellfish aquaculture was only selected at a narrow range of water depths probably because it is not the prime foraging habitat for great egrets. During low tide period, great egrets have mass natural habitats to forage; therefore, the relative selection strength for shellfish aquaculture decreased rapidly. In addition, after shellfish aquaculture infrastructures are exposed above water level, it would be difficult for great egrets to forage around. However, when tide comes up, the shellfish aquaculture is the last area to be inundated and thus becomes a potential foraging site for great egrets. During high tide, shellfish aquaculture was used more frequently probably because of its location, with higher elevation. If the proposition is right, the most important factor affecting the foraging of the great egrets would be the availability of natural wetlands. They would always prefer natural habitats unless high tide pushes them away.

2. Since the GPS-datalogger cannot provide information on foraging success, it is no way to tell a great egret is foraging or resting in a particular habitat. Especially, “sit and wait” is one of the foraging tactics they apply frequently. As a result, movement speed might not directly reflect foraging success.

3. Is it OK to turn the graph around along the X axis in Figure 2? Making negative values for data below water level would be easily to understand.

4. Based on Fig. 3 and 4, it appears that great egrets used more short movements and spent less energy (ODBA) in tidal marsh than in other habitat types. I do not know if it implies that great egrets forage more frequently in tidal marsh.

5. In Figure 4, it is better to designate the difference among wetland types. For example, mark different letters (a, b, or c) to show significant difference based on the Tukey’s test or non-parametric analysis.

6. In discussion, some results could be explained more explicitly. For example, how tidal movement affects the foraging behavior of great egrets and why shellfish aquaculture was selected at a narrow range of water depth? In addition, the disturbance of human activity in shellfish aquaculture was not discussed at all. It could be an important factor preventing great egrets to use shellfish aquaculture.

Others:

L25-26: “We found evidence for stronger overall habitat selection for natural wetlands than for shellfish aquaculture areas,…”

When we refer to habitat selection, the availability of each habitat type must be taken into consideration. A use and availability analysis is usually required before any conclusion on habitat selection is made; otherwise the result is about habitat use, not habitat selection. Although the authors applied iSSA (integrated step selection analysis), the results might not agree with a use and availability analysis.

L206: due to not dur to.

L296: ”We obtained a mean = 177 ± SE 31 days of Tomales Bay foraging data per tagged great egret,…”

Actually, we do not know if great egrets are really foraging based on GPS data even though data have been filtered before data analysis.

In Figures 2 and 3, GREG_10 was located between GREG_1 and GREG_2. It is better to arrange them in order. The authors probably can rename the first 6 by GREG_01, GREG_02, etc, and the software will place the GREG_10 into the last place.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

PLoS One. doi: 10.1371/journal.pone.0261963.r004

Acceptance letter

Hans-Uwe Dahms

21 Dec 2021

PONE-D-21-19522R1

Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: comparing natural wetlands to areas with shellfish aquaculture

Dear Dr. Jennings:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Hans-Uwe Dahms

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Fig. Elevations for each wetland type.

(TIF)

Click here for additional data file.^{(809.3KB, tif)}

S1 Table. Habitat selection model selection.

(DOCX)

Click here for additional data file.^{(22.7KB, docx)}

S2 Table. Step length model selection.

(DOCX)

Click here for additional data file.^{(18.7KB, docx)}

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file.^{(61.4KB, docx)}

Data Availability Statement

Tagging data are archived here: https://www.movebank.org/cms/webapp?gwt_fragment=page=studies,path=study247850178.

[pone.0261963.ref001] 1.Van Moorter B, Rolandsen CM, Basille M, Gaillard J-MM. Movement is the glue connecting home ranges and habitat selection. J Anim Ecol. 2016;85: 21–31. doi: 10.1111/1365-2656.12394 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref002] 2.Calle L, Gawlik DE, Xie Z, Green L, Lapointe B, Strong A, et al. Effects of tidal periodicities and diurnal foraging constraints on the density of foraging wading birds. Auk. 2016;133: 378–396. doi: 10.1642/auk-15-234.1 [DOI] [Google Scholar]

[pone.0261963.ref003] 3.Norevik G, Åkesson S, Andersson A, Bäckman J, Hedenström A. The lunar cycle drives migration of a nocturnal bird. Chittka L, editor. PLOS Biol. 2019;17: e3000456. doi: 10.1371/journal.pbio.3000456 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref004] 4.Cruz SM, Hooten M, Huyvaert KP, Proaño CB, Anderson DJ, Afanasyev V, et al. At–Sea Behavior Varies with Lunar Phase in a Nocturnal Pelagic Seabird, the Swallow-Tailed Gull. Ropert-Coudert Y, editor. PLoS One. 2013;8: e56889. doi: 10.1371/journal.pone.0056889 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref005] 5.Tucker MA, Böhning-Gaese K, Fagan WF, Fryxell JM, Van Moorter B, Alberts SC, et al. Moving in the Anthropocene: Global reductions in terrestrial mammalian movements. Science (80-). 2018;359: 466–469. doi: 10.1126/science.aam9712 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref006] 6.Teitelbaum CS, Hepinstall-Cymerman J, Kidd-Weaver A, Hernandez SM, Altizer S, Hall RJ. Urban specialization reduces habitat connectivity by a highly mobile wading bird. Mov Ecol. 2020;8: 49. doi: 10.1186/s40462-020-00233-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref007] 7.Doherty TS, Driscoll DA. Coupling movement and landscape ecology for animal conservation in production landscapes. Proc R Soc B Biol Sci. 2018;285. doi: 10.1098/rspb.2017.2272 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref008] 8.Allen AM, Singh NJ. Linking movement ecology with wildlife management and conservation. Front Ecol Evol. 2016;3: 1–13. doi: 10.3389/fevo.2015.00155 [DOI] [Google Scholar]

[pone.0261963.ref009] 9.McCrimmon DAJ, Ogden JC, Bancroft GT, Martínez-Vilalta A, Motis A, Kirwan GM, et al. Great Egret (Ardea alba), version 1.0. In: Billerman SM, editor. Birds of the World. Ithaca, New York, USA: Cornell Lab of Ornitholog; 2020. [Google Scholar]

[pone.0261963.ref010] 10.Maccarone AD, Brzorad JN. Foraging behavior and energetics of Great Egrets and Snowy Egrets at interior rivers and weirs. J F Ornithol. 2007;78: 411–419. doi: 10.1111/j.1557-9263.2007.00133.x [DOI] [Google Scholar]

[pone.0261963.ref011] 11.Maccarone AD, Brzorad JN. Foraging Patterns of Breeding Egrets at Coastal and Interior Locations. Waterbirds. 2002;25. doi: 10.1675/1524-4695(2002)025[0001:FPOBEA]2.0.CO;2 [DOI] [Google Scholar]

[pone.0261963.ref012] 12.Brzorad JN, Maccarone AD, Conley KJ. Foraging energetics of Great Egrets and Snowy Egrets. J F Ornithol. 2004;75: 266–280. doi: 10.1648/0273-8570-75.3.266 [DOI] [Google Scholar]

[pone.0261963.ref013] 13.Burr PC, Avery JL, Street GM, Strickland BK, Dorr BS. Historic and contemporary use of catfish aquaculture by piscivorous birds in the Mississippi Delta. Condor. 2020;122. doi: 10.1093/condor/duaa009 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref014] 14.Gorenzel WP, Conte FS, Salmon TP. Bird damage at aquaculture facilities. Handb Prev Control Wildl Damage. 1994;57. Available: https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1055&context=icwdmhandbook [Google Scholar]

[pone.0261963.ref015] 15.Dumbauld BR, Ruesink JL, Rumrill SS. The ecological role of bivalve shellfish aquaculture in the estuarine environment: A review with application to oyster and clam culture in West Coast (USA) estuaries. Aquaculture. 2009;290: 196–223. doi: 10.1016/j.aquaculture.2009.02.033 [DOI] [Google Scholar]

[pone.0261963.ref016] 16.Lantz SM, Gawlik DE, Cook MI. The effects of water depth and emergent vegetation on foraging success and habitat selection of wading birds in the Everglades. Waterbirds. 2011;34: 439–447. doi: 10.1675/063.034.0406 [DOI] [Google Scholar]

[pone.0261963.ref017] 17.California Department of Fish and Wildlife. Tomales Bay shellfish production, unpublished data.

[pone.0261963.ref018] 18.Custer T, Osborn R. Feeding habitat use by colonially-breeding herons, egrets, and ibises in North Carolina. Auk. 1978;95: 733–743. doi: 10.1093/auk/95.4.733 [DOI] [Google Scholar]

[pone.0261963.ref019] 19.Hughes BB, Levey MD, Brown JA, Fountain MC, Carlisle AB, Litvin SY, et al. Nursery functions of U.S. West Coast estuaries: the state of knowledge for juveniles of invertebrate and fish species. The Nature Conservancy, Arlington, VA; 2014. p. 168. Available: http://research.pbsci.ucsc.edu/eeb/bbhughes/wp-content/uploads/2014/12/Hughes_etal_2014_NurseryFunction_TNC.pdf [Google Scholar]

[pone.0261963.ref020] 20.Plummer ML, Harvey CJ, Anderson LE, Guerry AD, Ruckelshaus MH. The role of eelgrass in marine community interactions and ecosystem services: results from ecosystem-scale food web models. Ecosystems. 2013;16: 237–251. doi: 10.1007/s10021-012-9609-0 [DOI] [Google Scholar]

[pone.0261963.ref021] 21.Röhr ME, Holmer M, Baum JK, Björk M, Boyer K, Chin D, et al. Blue carbon storage capacity of temperate eelgrass (Zostera marina) meadows. Global Biogeochem Cycles. 2018;32: 1457–1475. doi: 10.1029/2018GB005941 [DOI] [Google Scholar]

[pone.0261963.ref022] 22.Zarnoch CB, Hoellein TJ, Furman BT, Peterson BJ. Eelgrass meadows, Zostera marina (L.), facilitate the ecosystem service of nitrogen removal during simulated nutrient pulses in Shinnecock Bay, New York, USA. Mar Pollut Bull. 2017;124: 376–387. doi: 10.1016/j.marpolbul.2017.07.061 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref023] 23.National Oceanic and Atmospheric Association. West Coast Region California Eelgrass Mitigation Policy and Implementing Guidelines. 2014. Available: https://www.fisheries.noaa.gov/resource/document/california-eelgrass-mitigation-policy-and-implementing-guidelines

[pone.0261963.ref024] 24.Huang AC, Essak M, O’Connor MI. Top-down control by great blue herons Ardea herodias regulates seagrass-associated epifauna. Oikos. 2015;124: 1492–1501. doi: 10.1111/oik.01988 [DOI] [Google Scholar]

[pone.0261963.ref025] 25.Hughes BB, Eby R, Dyke E Van, Tinker MT, Marks CI, Johnson KS, et al. Recovery of a top predator mediates negative eutrophic effects on seagrass. PNAS. 2013;110: 15313–15318. doi: 10.1073/pnas.1302805110 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref026] 26.San Francisco Estuary Institute. California Aquatic Resource Inventory (CARI) version 0.3. 2017. Available: https://www.sfei.org/data/california-aquatic-resource-inventory-cari-version-03-gis-data#sthash.K4v5RIFd.dpbs

[pone.0261963.ref027] 27.Fair JM, Paul E, Jones J, Clark AB, Davie C, Kaiser G. Guidelines to the use of wild birds in research. Ornithological Council; 2010. Available: http://www.nmnh.si.edu/BIRDNET [Google Scholar]

[pone.0261963.ref028] 28.Brzorad JN, Maccarone AD. An inexpensive technique for capturing gregarious wading birds on their foraging grounds. Waterbirds. 2014;37: 335–339. doi: 10.1675/063.037.0312 [DOI] [Google Scholar]

[pone.0261963.ref029] 29.San Francisco Estuary Institute. Eelgrass Survey GIS Data version 2.0. 2017. Available: https://www.sfei.org/data/eelgrass-survey-gis-data#sthash.eBKfqYBx.dpbs

[pone.0261963.ref030] 30.Marin County. Marin County ortho images. 2018 [cited 6 Nov 2020]. Available: https://gis.marinpublic.com/arcgis/rest/services/Orthos/2018Orthos/ImageServer

[pone.0261963.ref031] 31.Marin County. Marin Map LIDAR DEM version 10.71. 2018. Available: https://gis.marinpublic.com/arcgis/rest/services/LIDAR

[pone.0261963.ref032] 32.National Centers for Environmental Information N. NOAA NOS Estuarine Bathymetry—Tomales Bay. 2018 [cited 6 Nov 2020]. Available: https://data.noaa.gov/metaview/page?xml=NOAA/NESDIS/NGDC/MGG/DEM/iso/xml/tomales_bay_P110_2018.xml&view=rubricv2/recordHTML&header=none

[pone.0261963.ref033] 33.Qasem L, Cardew A, Wilson A, Griffiths I, Halsey LG, Shepard ELC, et al. Tri-axial dynamic acceleration as a proxy for animal energy expenditure; should we be summing values or calculating the vector? Ropert-Coudert Y, editor. PLoS One. 2012;7: e31187. doi: 10.1371/journal.pone.0031187 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref034] 34.Wilson RP, White CR, Quintana F, Halsey LG, Liebsch N, Martin GR, et al. Moving towards acceleration for estimates of activity-specific metabolic rate in free-living animals: the case of the cormorant. J Anim Ecol. 2006;75: 1081–1090. doi: 10.1111/j.1365-2656.2006.01127.x [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref035] 35.Gleiss AC, Wilson RP, Shepard ELCC. Making overall dynamic body acceleration work: On the theory of acceleration as a proxy for energy expenditure. Methods Ecol Evol. 2011;2: 23–33. doi: 10.1111/j.2041-210X.2010.00057.x [DOI] [Google Scholar]

[pone.0261963.ref036] 36.Green JA, Halsey LG, Wilson RP, Frappell PB. Estimating energy expenditure of animals using the accelerometry technique: activity, inactivity and comparison with the heart-rate technique. J Exp Biol. 2009;212: 471–82. doi: 10.1242/jeb.026377 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref037] 37.Manly BFJ, McDonald LL, Thomas DL, McDonald TL, Erickson WP. Resource selection by animals: statistical design and analysis for field studies. Kluwer Academic Publishers; 2002. [Google Scholar]

[pone.0261963.ref038] 38.Paton RS, Mattiopoulos J. Defining the scale of habitat availability for models of habitat selection. Ecology. 2016;97: 1113–1122. doi: 10.1890/14-2241.1 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref039] 39.Avgar T, Potts JR, Lewis MA, Boyce MS. Integrated step selection analysis: bridging the gap between resource selection and animal movement. Börger L, editor. Methods Ecol Evol. 2016;7: 619–630. doi: 10.1111/2041-210X.12528 [DOI] [Google Scholar]

[pone.0261963.ref040] 40.Muff S, Signer J, Fieberg J. Accounting for individual-specific variation in habitat-selection studies: Efficient estimation of mixed-effects models using Bayesian or frequentist computation. J Anim Ecol. 2020;89: 80–92. doi: 10.1111/1365-2656.13087 [DOI] [PubMed] [Google Scholar]

[pone.0261963.ref041] 41.Ranacher P, Brunauer R, Trutschnig W, Der S Van, Reich S, Ranacher P, et al. Why GPS makes distances bigger than they are. Int J Geogr Inf Sci. 2016;30: 316–333. doi: 10.1080/13658816.2015.1086924 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref042] 42.Noonan MJ, Fleming CH, Akre TS, Drescher-Lehman J, Gurarie E, Harrison AL, et al. Scale-insensitive estimation of speed and distance traveled from animal tracking data. Mov Ecol. 2019;7: 1–15. doi: 10.1186/s40462-019-0147-7 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref043] 43.Johnson DH. The comparison of usage and availability measurements for evaluating resource preference. Ecology. 1980;61: 65–71. doi: 10.2307/1937156 [DOI] [Google Scholar]

[pone.0261963.ref044] 44.Signer J, Fieberg J, Avgar T. Animal movement tools (amt): R package for managing tracking data and conducting habitat selection analyses. Ecol Evol. 2019;9: 880–890. doi: 10.1002/ece3.4823 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref045] 45.Burnham KP, Anderson DR. Model selection and multimodel inference: a practical information-theoretic approach, 2nd edn. New York: Springer-Verlag; 2002. [Google Scholar]

[pone.0261963.ref046] 46.Fieberg J, Signer J, Smith B, Avgar T. A “how-to” guide for interpreting parameters in resource-and step-selection analyses. bioRxiv. 2020; 2020.11.12.379834. Available: 10.1101/2020.11.12.379834 [DOI] [Google Scholar]

[pone.0261963.ref047] 47.Zuur AF, Ieno EN, Walker N, Saveliev AA, Smith GM. Mixed effects models and extensions in ecology with R. 2009. [Google Scholar]

[pone.0261963.ref048] 48.ESRI. ArcGIS Pro. Redlands, CA: Environmental Systems Research Institute; 2021. [Google Scholar]

[pone.0261963.ref049] 49.R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2019. Available: https://www.r-project.org/ [Google Scholar]

[pone.0261963.ref050] 50.Kelley D, Richard C. oce: Analysis of Oceanographic Data. 2020. Available: https://cran.r-project.org/package=oce [Google Scholar]

[pone.0261963.ref051] 51.Bates D, Maechler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. J Stat Softw. 2015;67: 1–48. doi: 10.18637/jss.v067.i01 [DOI] [Google Scholar]

[pone.0261963.ref052] 52.Kelly JP, Evens JG, Stallcup RW, Wimpfheimer D. Effects of aquaculture on habitat use by wintering shorebirds. Calif Fish Game. 1996;82: 160–174. Available: https://egret.org/sites/default/files/scientific_contributions/kelly_etal_1996_aquaculture.pdf [Google Scholar]

[pone.0261963.ref053] 53.Kelly JP, Stralberg D, Etienne K, Mccaustland M. Landscape influence on the quality of heron and egret colony sites. Wetlands. 2008;28: 257–275. doi: 10.1672/07-152.1 [DOI] [Google Scholar]

[pone.0261963.ref054] 54.Calle L, Green L, Strong A, Gawlik DE. Time-integrated habitat availability is a resource attribute that informs patterns of use in intertidal areas. Ecol Monogr. 2018;88: 600–620. doi: 10.1002/ecm.1305 [DOI] [Google Scholar]

[pone.0261963.ref055] 55.Stolen ED, Collazo JA, Percival HF. Group-foraging effects on capture rate in wading birds. Condor. 2012;114: 744–754. doi: 10.1525/cond.2012.110159 [DOI] [Google Scholar]

[pone.0261963.ref056] 56.Calle L, Gawlik DE, Xie Z, Green L, Lapointe B, Strong A. Effects of tidal periodicities and diurnal foraging constraints on the density of foraging wading birds. Auk. 2016;133: 378–396. doi: 10.1642/auk-15-234.1 [DOI] [Google Scholar]

[pone.0261963.ref057] 57.Beerens JM, Noonburg EG, Gawlik DE. Linking dynamic habitat selection with wading bird foraging distributions across resource gradients. PLoS One. 2015;10: e0128182. doi: 10.1371/journal.pone.0128182 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0261963.ref058] 58.Botson BA, Gawlik DE, Trexler JC. Mechanisms that generate resource pulses in a fluctuating wetland. PLoS One. 2016;11: e0158864. doi: 10.1371/journal.pone.0158864 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Great egret (Ardea alba) habitat selection and foraging behavior in a temperate estuary: Comparing natural wetlands to areas with shellfish aquaculture

Scott Jennings

David Lumpkin

Nils Warnock

T Emiko Condeso

John P Kelly

Roles

Abstract

Introduction

Methods

Study area

Fig 1. Study site map.

Trapping/Tagging

Data preparation

Habitat mapping

GPS/Accelerometer data

Analysis

Results

Table 1. Summary of tagged great egrets.

Fig 2. log-Relative selection strength and 95% confidence intervals of different tidal wetland types by seven GPS tagged great egrets at Tomales Bay, CA, 2017–2020, while accounting for tide-based changes in water depth.

Fig 3. Distribution of step lengths (straight distance between consecutive evenly spaced timestamps) in different tidal wetland types by foraging great egrets at Tomales Bay, CA, 2017–2020.

Fig 4. Overall dynamic body acceleration (ODBA) by seven GPS tagged great egrets in different tidal wetland types at Tomales Bay, CA, 2017–2019.

Discussion

Management implications

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Vanesa Magar

Roles

Author response to Decision Letter 0

Decision Letter 1

Hans-Uwe Dahms

Roles

Acceptance letter

Hans-Uwe Dahms

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases