Nothing Special   »   [go: up one dir, main page]
Next Article in Journal
Assessment of the Impact of Coastal Wetland Saltmarsh Vegetation Types on Aboveground Biomass Inversion
Previous Article in Journal
TDMSANet: A Tri-Dimensional Multi-Head Self-Attention Network for Improved Crop Classification from Multitemporal Fine-Resolution Remotely Sensed Images
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 16
Issue 24
10.3390/rs16244764
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Piping Plover Habitat Changes and Nesting Responses Following Post-Tropical Cyclone Fiona on Prince Edward Island, Canada

by
Ryan Guild
1,2 and
Xiuquan Wang
1,2,*
1
Canadian Centre for Climate Change and Adaptation, University of Prince Edward Island, St. Peters Bay, PE C0A 2A0, Canada
2
School of Climate Change and Adaptation, University of Prince Edward Island, Charlottetown, PE C1A 4P3, Canada
*
Author to whom correspondence should be addressed.
Remote Sens. 2024, 16(24), 4764; https://doi.org/10.3390/rs16244764
Submission received: 1 October 2024 / Revised: 13 December 2024 / Accepted: 16 December 2024 / Published: 20 December 2024
Browse Figures
Figure 1
<p>Map of all nesting sites on PEI with breeding activity since 2011.</p> "> Figure 2
<p>Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical barrier island habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circle on the island map indicates nesting site with no data (ND).</p> "> Figure 3
<p>Classified landcover pre-storm (A) and post-storm (B) and total change in open sand area (C) over critical sandspit/bar habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND). Additional classified sandspit/bar nesting sites are displayed in <a href="#app1-remotesensing-16-04764" class="html-app">Figure S3</a>.</p> "> Figure 4
<p>Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical mainland beach habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND).</p> "> Figure 5
<p>Nest locations and outcomes during three breeding seasons preceding (top three panels) and the initial season following (fourth panel) PTC Fiona over Conway Sandhills, PEI. Classified change in dry and wet sand area after one-year post-storm depicted in bottom panel, with colours representing change classes (in <a href="#remotesensing-16-04764-f002" class="html-fig">Figure 2</a>, <a href="#remotesensing-16-04764-f003" class="html-fig">Figure 3</a> and <a href="#remotesensing-16-04764-f004" class="html-fig">Figure 4</a>).</p> "> Figure 6
<p>Fledging rate across common nesting sites with at least three years of nesting attempts between 2020 and 2023. Horizontal line at 1.65 fledglings/pair indicates ECCC productivity target for the Eastern Canadian recovery unit. Vertical line distinguishes between pre- and post-storm breeding seasons.</p> "> Figure 7
<p>Model-averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of binary hatch success (<b>left</b>) and data summaries of hatch outcomes across habitat metrics from 2020–2023 on PEI (<b>right</b>). Number in white indicates nest counts in each respective category; error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.</p> "> Figure 8
<p>Summaries of binary hatch success by year across habitat measures on PEI from 2020–2023. Number in white indicates the number of nests in each respective category. D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.</p> "> Figure 9
<p>Model averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of flooding and predation occurrences (<b>left</b>) and data summaries of nest outcomes across habitat metrics from 2020–2023 on PEI (<b>right</b>). Error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.</p> ">
Versions Notes

Abstract

:
Climate change is driving regime shifts across ecosystems, exposing species to novel challenges of extreme weather, altered disturbances, food web disruptions, and habitat loss. For disturbance-dependent species like the endangered piping plover (Charadrius melodus), these shifts present both opportunities and risks. While most piping plover populations show net growth following storm-driven habitat creation, similar gains have not been documented in the Eastern Canadian breeding unit. In September 2022, post-tropical cyclone Fiona caused record coastal changes in this region, prompting our study of population and nesting responses within the central subunit of Prince Edward Island (PEI). Using satellite imagery and machine learning tools, we mapped storm-induced change in open sand habitat on PEI and compared nest outcomes across habitat conditions from 2020 to 2023. Open sand areas increased by 9–12 months post-storm, primarily through landward beach expansion. However, the following breeding season showed no change in abundance, minimal use of new habitats, and mixed nest success. Across study years, backshore zones, pure sand habitats, and sandspits/sandbars had lower apparent nest success, while washover zones, sparsely vegetated areas, and wider beaches had higher success. Following PTC Fiona, nest success on terminal spits declined sharply, dropping from 45–55% of nests hatched in pre-storm years to just 5%, partly due to increased flooding. This suggests reduced suitability, possibly from storm-induced changes to beach elevation or slope. Further analyses incorporating geomorphological and ecological data are needed to determine whether the availability of suitable habitat is limiting population growth. These findings highlight the importance of conserving and replicating critical habitat features to support piping plover recovery in vulnerable areas.

1. Introduction

Coastal ecosystems are on the frontlines of climate change, where rising sea levels and altered weather patterns are reshaping life for many species. The disturbance regimes that have historically shaped these environments, and to which coastal species have adapted, are intensifying due to global warming, particularly through escalations in sea level rise, rainfall intensities, and regional storminess [1]. Poleward shifts in cyclones, for example, are subjecting northern hemisphere coasts to more frequent and severe storms that rapidly restructure habitats and affect the vital rates of coastal species. Consequences have included degradation to coastal vegetated zones (e.g., [2,3,4]), harmful hydrological changes (e.g., [5,6]), and population declines for species directly exposed to such events (e.g., [7,8,9]). However, disturbance events can also benefit species adapted to early-successional habitats, such as seabirds nesting on ephemeral islands and scoured cliffs and sea turtles and shorebirds that exploit storm-modified beaches.
The migratory piping plover (Charadrius melodus; hereafter “PIPL”), one of North America’s most endangered shorebirds, relies on open-sand habitats created and maintained by storms and floods. Intensive monitoring efforts in Canada and the U.S., including long-term population surveys, habitat mapping, and demographic modelling, have documented numerous examples of population “irruptions” following natural or human-driven habitat increases [10,11,12,13,14,15,16,17,18,19] and declines in their absence (e.g., [12,20]). These studies linked population increases to storm-generated features such as large devegetated sand flats, ephemeral pools, and washover fans that offer benefits of lower nest-site competition, better foraging conditions, and reduced detectability by predators (e.g., [11,13,15,21,22]). Many PIPL populations are thus considered habitat-limited [23] and rapidly colonize storm-created washovers and dune blowouts for nesting [12,18,19,24,25,26,27,28,29], where they often achieve higher productivity [27,28]. However, these benefits can be diminished or negated by anthropogenic shoreline modifications that disrupt natural overwash processes [28] or by factors like predation, inclement weather, and human activity that offset habitat and productivity gains [30,31]. Even when post-storm productivity increases, local population growth may not follow [28,32] due to challenges of unbalanced dispersal, small population effects, and overwintering threats. Moreover, not all severe storms offer benefits; some heavily erode beachfronts that reduce nesting areas and heighten flood risks [26], while others raise mortality rates when they overlap with population distributions [33,34]. These variable outcomes underscore the need for local and sub-regional studies to understand how PIPL populations respond to habitat changes. Such research informs conservation strategies to protect or replicate key habitat features that support this endangered species.
In the Eastern Canadian recovery unit, which spans five coastal provinces and supports about 10% of the Atlantic-breeding melodus subspecies, research has not observed the same population irruptions that follow habitat creation events elsewhere. Studies from New Brunswick [24,32] and Nova Scotia [35], the most populous nesting provinces, indicate that post-storm productivity increases have not led to population growth. One explanation is that the nesting habitat in the region was already undersaturated with breeders prior to habitat creation (as speculated in [36,37]). This would imply that other factors, such as suboptimal foraging habitats, the rigors of a long-distance migration, or overwintering threats, might be limiting population growth despite adequate productivity (as speculated in [23,34]). However, the insights from a subset of nesting provinces may not represent the entire recovery unit. Moreover, assumptions about “suitable” habitat in Eastern Canada have been challenged by disproportionate habitat use across otherwise “suitable” landscapes [38,39]. Thus, further research is needed to determine whether habitat availability remains a constraint on population growth in this region.
Prince Edward Island (PEI) plays a central role within this recovery unit, serving as both a source and sink for dispersing PIPL [36]. Over the past three decades, PEI has hosted 16–22% of the recovery unit’s breeding pairs [40], comprising mostly island-hatched birds, with some immigrants from other provinces and even U.S. breeding sites [36,41]. Despite this connectivity, the island’s population has not yet met the abundance target (60 pairs) and has mostly fallen short of the productivity target (1.65 fledglings per pair) required for population stability [40,42]. While PEI is thought to offer abundant suitable habitat, and descriptive accounts of nest sites on the island exist (e.g., [41,43,44,45]), this assumption has not been verified with geomorphological and ecological data. Observing population responses to habitat-altering events, such as severe storms, may help clarify habitat availability. In 2022, post-tropical cyclone (PTC) Fiona, the strongest storm to make landfall in over 50 years, created a unique opportunity to study these dynamics.
This study aimed to assess habitat availability and environmental influences on breeding activity in the focal population on PEI following PTC Fiona. Specifically, we classified time-series satellite imagery and analyzed nest productivity and related environmental factors before and after the storm. We first hypothesized that storm-induced habitat creation would not immediately increase abundance on PEI. This assumption draws on longstanding but unverified ideas that (1) breeding habitat is already sufficient and (2) non-breeding season threats, such as migratory challenges, winter storm mortality, and poor-quality overwintering habitat, constrain juvenile recruitment and thus limit population growth in the region [37,40]. Additionally, while a neighboring population was found to quickly occupy storm-created habitats ([24]; cf. 32,35), other research suggests that colonization may be delayed because second-year birds, the primary dispersers, need time to prospect new nesting habitats before returning to breed (e.g., [46]). This context supports our first hypothesis yet complicates predictions of post-storm nesting choices in the focal population. We also hypothesized that certain environmental features (e.g., washovers, wider beaches, cobble substrates, limited beach access) would enhance nest success by meeting habitat requirements and reducing predation or disturbance risks. Overall, this research contributes to improving understanding of the factors constraining PIPL populations in Eastern Canada.

2. Data and Methods

2.1. Study Area

Nesting habitats on PEI are located along the island’s northern and eastern coastlines [42], situated between dune systems and the high water line (HWL) of barrier islands, sandspits/bars, and mainland beaches. These habitats are generally characterized by wide, flat beaches with sparse foredune vegetation, mottled nesting substrates, and access to inland water or bays [39,43,45]. Around 41 distinct nesting sites on the island have been designated as “critical habitat” for PIPL, defined by the presence of key habitat features and at least one nesting pair since 1991 (Figure 1) [39,40]. The island’s shorelines are particularly susceptible to erosion due to low relief (<45 m AMSL), brittle sandstone bedrock, and coastal subsidence [47]. This vulnerability is most pronounced along its northern coast, where most PIPL nest, as sediment-starved beaches there face significant wave energy generated across several hundred kilometers of open water in the Gulf of St. Lawrence [48,49]. Intense wind and waves are often driven by mid-latitude cyclones (October–January) and hurricanes (September–October) originating in the Caribbean. These systems typically weaken before reaching the region but occasionally reintensify into powerful post-tropical systems [48,50], as seen with PTC Fiona. More than half a century of tidal data (e.g., [51]) show that storm surges exceeding 1.2 m on the island are rare, and none before Fiona have surpassed 1.5 m.

2.2. PTC Fiona

PTC Fiona originated in the Caribbean as a powerful Category 4 hurricane and made landfall in Nova Scotia on 23 Sep 2022, setting a Canadian record for lowest atmospheric pressure [52]. On PEI, Fiona brought sustained winds exceeding 100 km/h, gusts up to 170 km/h, and a storm surge of more than 2 m that led to waves exceeding 6 m along the northern coastline [53,54,55]. The result was significant dune overtopping and shoreline erosion in a region where the majority of PIPL nest. The combination of prolonged northerly gale-force winds, high tides, and an ice-free coastline made Fiona the strongest erosive storm in the region in at least 70 years [53]. Comparable events have only been recorded during a period of anomalous storm activity in the late 1800s and early 1900s [48], which levelled dune systems along PEI’s northern coastline and took approximately six decades to recover [50].

2.3. Image Classification

Visual inspection of time-series coastal imagery across PEI shows that, while no new nesting sites (i.e., distinct expansive beach areas) emerged after PTC Fiona, many existing critical habitats saw major changes in the open-sand areas that PIPL nest in and around. High-resolution PlanetScope Scene imagery (8-band analytical; 3 m/px) was used to track storm-related habitat changes at 25 nesting sites (hereafter “sites”) that hosted 92% of nesting attempts and 90% of breeding pairs between 2020 and 2023. Excluded sites included those with fewer than two nesting attempts (n = 6) or without cloud-free satellite images (n = 11). Imagery was acquired before (“pre-storm”; between 10 Jun and 13 Sep 2022), immediately after (“post-storm”; between 30 September and 5 December 2022), and 9–12 months after (“revisit”; between 12 June and 22 August 2023) the passage of PTC Fiona. All images were pre-calibrated to top of atmosphere radiance to allow for standardized time-series comparison. Bands included red, green, blue, near-infrared, coastal blue, green II, yellow, and red edge. A lack of updated LiDAR data to generate higher high water at large tide (HHWLT; or highest astronomical tide) contours prevented the standardized delineation of beach areas in the imagery. Thus, effort was made to procure images at high tide levels by cross-referencing tidal gauge data from five coastal stations equally spread across the island. Tide levels between images at a given site differed on average by 0.41 m (±0.15 m), where maximum daily tidal amplitudes ranged between 1.5 and 2.5 m depending on the station.
Images were classified in Google Earth Engine using the CART (Classification and Regression Trees) algorithm (cf. [56,57,58]) to map four landcover classes: dry sand (nesting), wet sand (foraging), vegetation, and water (Figure S1). These categories were chosen based on their relevance to PIPL habitat preferences and demonstrated effectiveness in mapping habitat changes in other regions (e.g., [17,59]). Finer habitat features such as sparse vegetation cover and substrate types (e.g., shells, cobble, wrack) could not be resolved at the given image resolution. The classification process involved several steps repeated for each image across sites. First, by referencing drone and ground-level photographs across sites to ensure accuracy, an equal number of training pixels for each land cover class were delineated using points and polygons. Training points were merged into a single feature collection and utilized to train the CART classifier, incorporating all eight image bands. Masks were then used to partition each image into habitat (dry and wet sand) and non-habitat (vegetation and water) categories.
To produce final change maps, we took precautions to avoid classifying transient changes in land cover (e.g., from sediment redistribution) that dissipated shortly after PTC Fiona. “Immediate” habitat gain/loss was defined as changes detected between pre-storm and post-storm imagery that persisted into the revisit imagery, while “subsequent” gain/loss was defined as changes detected between pre-storm and revisit imagery only. Subsequent transitions may have resulted from ongoing sediment redistribution following PTC Fiona or from other erosive forces (e.g., winter storms, ice scour, high winds) before the next breeding season. Changes were then quantified as proportions of habitat change (i.e., change pixels) for each site. To validate classification accuracy, testing pixels (30% of training pixel counts per class) were independently delineated for each image and confusion matrices were generated to calculate producer’s accuracy (PA), user’s accuracy (UA), and overall accuracy (OA). PA was measured by correctly classified testing pixels as a proportion of total testing pixels in each class, UA was measured by correctly classified testing pixels as a proportion of all testing pixels classified as that class, and OA was measured by the proportion of all correctly classified pixels across all classes. These metrics were averaged across sites and reported with standard deviations.

2.4. Nest Data Collection

Since 2011, trained observers have conducted standardized monitoring of reproductive attempts across potential nesting sites on PEI during the breeding season (March to August). As detailed in [42], teams systematically and regularly search the open sand beaches and dune areas of critical and historical nesting sites for PIPL and their nests. Once sighted, observers revisit the site frequently to monitor nest and brood fates until loss or fledging. Throughout this process, various characteristics are recorded, including nest coordinates, habitat type (i.e., open sand, cobble, or sparse vegetation), egg and chick counts, and, when discernible from secondary evidence, the cause of nest loss. Following standardized protocols, year-end measures of productivity (i.e., fledging rates) are calculated by dividing fledgling counts by the number of monitored breeding pairs.

2.5. Nest Data Analysis

In addition to the image classification, we examined habitat-related influences on nest success before and after PTC Fiona by comparing nest data (2020–2023) against habitat measures derived from monitoring observations and satellite imagery (Table 1). We primarily utilized very-high resolution Maxar and Airbus satellite imagery (on Google Earth Pro; 30 cm/px) during the nesting periods, supplemented by PlanetScope imagery (3 m/px) when breeding season images from the former were unavailable. With this imagery, measurement tools in Google Earth Pro and QGIS were used to obtain per-nest measures of beach width and distances to various landscape features (Table 1; Figure S2). Distances to shorelines were measured to the high tide line (i.e., intertidal zones), identified by wet sand and verified by comparing sequential image dates to rule out wave effects and tidal stages of any given image. We also classified each nest attempt based on the geophysical characteristics of its location, such as the landform type (i.e., geomorphology), beach zone, and whether the nesting habitat had been visibly altered within the past year or was accessible to humans during the breeding season (Table 1).
Our analysis included a sample of 146 nests, excluding those discovered post-hatching and those monitored too infrequently to determine fates with certainty. A lack of nest-level metadata (i.e., visitation dates and times) prevented the assessment of detection probabilities using nest survival models. We thus relied upon apparent nest success calculated from the nests that were detected and monitored. This approach may underrepresent nests that failed before being found, potentially introducing detection bias, but previous research indicates that increased survey effort did not significantly improve nest detectability across 14 nesting sites on PEI [60]. Moreover, the systematic and regular nature of monitoring efforts on PEI minimizes potential detection biases across different habitats. Nevertheless, we acknowledge that some nests, particularly those that failed early, may not have been detected and thus may be underrepresented in our sample.
Nest outcomes from 2020 to 2023 were analyzed in R version 4.4.1 [61] using “dplyr” [62], “tidyr” [63], and “ggplot2” [64]. Binary hatch success (overall and per year) and attributed nest fates were summarized across habitat variables and represented visually as “data summaries”. Additionally, with the glm function from the base “stats” package, logistic regression Generalized Linear Models (GLM) were used to analyze three response variables (RV) of interest: hatch success, flooded nests, and predated nests. Due to the small sample size, abandoned/buried nests (n = 2) were not analyzed independently as a RV. Predictor variables for each RV were chosen based on assumptions of potential influence (e.g., for the flooded nests RV: beach width, beach zone, distance to ocean ITZ, and terminal end habitats). For flooded and predated nests, GLMs excluded nests lost for unknown reasons (n = 40) to minimize noise, as these losses could have resulted from either flooding or predation. All full-model GLMs adhered to best practices, i.e., standardizing continuous variables and avoiding collinearity (Pearson’s r < 0.7, VIF < 4), maintaining a minimum sample-to-parameter ratio of at least 10:1, and testing for a random effect of Year, which was deemed unnecessary in all models and thus excluded. Using the “MuMIn” package version 1.48.4 [65], an all-possible-subsets model selection approach ranked models by AIC weights, with top-ranked models (Δ AIC < 2) included in model-averaging to estimate conditional average coefficients. The latter was performed to address over-reliance on a single model when several performed similarly well. Predictive performance was assessed using k-fold cross validation (k = 10), with accuracy (proportion of correct predictions) calculated for each fold, averaged, and reported with standard deviations. Discriminative ability was evaluated using area under the curve (AUC) metrics with the “pROC” package [66] (Figure S1). Finally, 95% confidence intervals of coefficient estimates for each covariate in the model-averaged sets were calculated as 1.96 times their standard errors and visualized as error bars in model outputs.

3. Results

3.1. Habitat Change Classification

The accuracy assessment for land cover classification across PEI nesting sites showed consistently high PA, UA, and OA (all ≥ 0.92), with water and dry sand performing the best (Table 2). Vegetation and wet sand also showed strong accuracy despite more frequent confusion between these two classes. Overall, classification accuracy was high with minimal variation across sites (Table 2).
Most nesting sites showed increases in open sand areas following PTC Fiona (Figure 2, Figure 3 and Figure 4), driven primarily by landward expansion through dune blowouts, overwash, and vegetation scouring. Some sites also saw localized seaward expansion via progradation (e.g., Greenwich; Eglington Cove). The most extensive landward gains occurred on the Conway Sandhills (remote northwestern barrier island) where numerous dune blowouts and washover fans formed and remained unvegetated a year after the storm (Figure 2). At its narrowest section, full dune breaches created new channels to the bay, replacing an older washthrough nesting area with open water and narrow terminal spits. Nearby, the Cascumpec Sandhills, less than 100 m away, showed less significant morphological changes and vegetation scouring. At both sites, gains and losses were evenly split between the immediate and subsequent detection periods (Figure 2). At the smaller Wood Islands, gains predominantly occurred immediately post-storm due to vegetation scouring, while losses mostly occurred later from oceanside shoreline retreat. Overall, barrier island sites experienced landward movement of the oceanside shoreline, partially offsetting gains in open sand areas (Figure 2).
Compared to barrier island sites, sandspits, sandbars, and mainland beaches saw greater net gains in open sand, ranging from approximately +3% to +28% on sandspits/bars (Figure 3) to +15% to +30% on mainland beaches (Figure 4). Most of these gains occurred immediately post-storm, except at a few sites (e.g., Greenwich, Eglington Cove, Spry Cove) where more gains were detected in the subsequent period (Figure 3 and Figure 4). In addition to novel beach areas, older blowouts and incipient dunes between previous washthroughs were also reset to an early successional stage at several sites. Oceanside shorelines on mainland beaches and sandspits/bars showed mixed responses, with localized areas of both retreat and progradation. At Greenwich Tip, the loss of ~180 m of terminal beach was offset by only 90 m of landward dune scouring (Figure 3), but this loss created expansive tidal flats at low tide that were used by foraging shorebirds in subsequent breeding seasons (pers. obs., RG).

3.2. Post-Storm Nesting

In the first post-storm breeding season following PTC Fiona, 88% of nests were initiated in pre-existing open sand habitat. While some post-storm nests were initiated near areas of vegetation scouring or dune blowouts, only 6 out of 48 nesting attempts were found directly in newly created habitat. Of these, one mid-season nest at Shaw’s Beach successfully fledged a single chick, while the other five failed (four for unknown reasons, one due to flooding). A record eight flood-related nest failures were recorded at Conway Sandhills in 2023, all in washover sheets and terminal end habitats and lost during perigean spring tides (Figure 5). Before 2023, flooding had caused only two recorded nest failures at this site since 2011. Two additional flood-related losses were recorded in 2023 at Greenwich Tip (base of a scoured foredune) and Cavendish Sandspit (terminal backshore).
Annual island-wide abundance, productivity, and nesting characteristics from 2020 to 2023 are summarized in Table 3. In the first post-storm breeding season, overall abundance remained unchanged, but average productivity exceeded the ECCC target of 1.65 fledglings per pair for the first year since 2011. This increase occurred despite lower hatch success, smaller clutch sizes, and an overall decline in productivity at common nesting sites compared to the previous year (Table 2; Figure 6). Above-average brood sizes and higher brood success rates in 2023 contributed to the average rise in productivity (Table 2). Nearly half (n = 26/54; 48%) of the fledglings in 2023 were from the Conway Sandhills, consistent with pre-storm years (50–56%). All post-storm fledglings from this site came from two pre-existing washover sheets, most assumed to be renests following flooding. Conway Sandhills remain the only site consistently exceeding the productivity target from 2020 to 2023 (Figure 6).

3.3. Hatch Success Across Habitat Conditions

Our nest data analysis (146 samples from 2020 to 2023) summarized hatch outcomes in relation to ground-measured and satellite-derived nesting characteristics (Table 1). On average, hatched nests (those that produced at least one hatchling) were farther from the ocean ITZ (132.6 ± 9.6 m) and on wider beaches (220.8 ± 14 m) than those that failed before hatching (96 ± 7.6 m and 151.7 ± 10.2 m, respectively) (Figure 7), a pattern that remained consistent across all observation years (Figure 8). Original nests had higher hatch success (46% nests hatched) than confirmed renests (16%) (Figure 7 and Figure 8). Washovers consistently had the highest success rates of all beach zones (76%), followed by bayside (53%), foredune (45%), and backshore zones (28%). Barrier island (62%) and mainland sites (61%) had greater success on average than sandspit/sandbar sites (31%) (Figure 7). Backshore zones, barrier islands, sandspits/bars, and especially terminal end habitats saw lower hatch success in 2023 compared to pre-storm years (Figure 8). Sparse vegetation (57%) supported higher success than cobble (43%) or pure sand substrates (24%) (Figure 7). Finally, sites where public access was either restricted or physically impeded experienced higher success on average than open-access sites (Figure 7 and Figure 8).
The model-averaged GLMs identified wider beaches (n = 146, p = 0.0078) and washover zones (n = 146, p = 0.0011) as significant positive predictors of binary hatch success (Figure 7, Table S1). A one-SD increase in beach width (~107 m, from the mean of 182 m) doubled the odds of hatching at least one chick (odds ratio: 2.23; 95% CI: 1.24–3.99), raising a (hypothetical) 50% baseline hatch probability to ~69% (55–80% CI). Similarly, nests in washover zones had more than seven-times the odds of hatching at least one chick than those in other zones (odds ratio: 7.66; 95% CI: 2.29–25.57), increasing a 50% baseline hatch probability to around 88% (70–96% CI). Sandspit/sandbar sites were a significant negative predictor (n = 146, p = 0.0049) (Figure 7, Table S1), with an odds ratio of 0.14 (CI: 0.04–0.55), reducing a 50% baseline hatch probability to roughly 12% (4–36% CI).
Among nests with known outcomes, predated nests were closer to the ocean ITZ (90.8 ± 13.2 m) than those with other fates (129.9 ± 8.2 m), and predation was more frequent in confirmed renests (n = 4/9 nests) than original nests (n = 4/45) (Figure 9). Model-averaged GLMs identified distance to the ocean ITZ as a significant negative predictor of predation risk (n = 106, p = 0.047) (Figure 9; Table S1). Increasing nest distances by one SD from the ocean ITZ (~75 m, from the mean of 123 m) reduced predation odds by nearly half (odds ratio: 0.51; 95% CI: 0.26–0.98), lowering a 50% baseline predation probability to ~34% (21–50% CI). Flood-related losses were more common in original nests (n = 12/45) than confirmed renests (n = 1/9); in terminal end habitats (n = 15/42) versus other areas (n = 6/64); and in habitats that had changed since the previous year (n = 10/29) compared to unchanged habitats (n = 11/77). Terminal habitats were a significant positive predictor of flood risk (n = 106, p = 0.0022) (Figure 9; Table S1), with nests in these areas experiencing more than five-times the odds of flooding than in other habitats (odds ratio: 5.38; 95% CI: 1.85–15.6), increasing a 50% baseline flood probability to ~84% (65–94% CI).

4. Discussion

4.1. Habitat Change from PTC Fiona

Despite net increases in open sand areas across PEI nesting sites following PTC Fiona, returning PIPL showed minimal use of the newly created habitats, and overall abundance remained similar to the previous year. This limited short-term colonization aligns with other findings in the Eastern Canadian recovery unit [32,35], where low nesting densities appear to reinforce high site fidelity and reduce incentives for dispersal [36,38]. However, our results contrast with observations from New Brunswick, where PIPL abandoned older, degraded nesting sites for novel washover habitats immediately post-storm [24]. There, the immediate colonization of storm-generated habitats suggests that factors like delayed prey establishment or habitat prospecting by juveniles [46,67] may be less relevant when previous nesting sites are poor quality.
On PEI, repeated use of pre-existing nesting habitats over novel storm-created ones likely reflects strong site fidelity, suggesting that established sites were either of sufficient quality to deter dispersal or were reused despite potential declines in suitability (cf. [68]). The latter explanation aligns with our observations of lower island-wide hatch success in the first post-storm year, partly due to increased flood-related losses. Although beach widening after PTC Fiona could have enhanced hatch success by reducing predator efficiency and allowing birds to nest farther from the HWL [69,70], these benefits may have been offset by other factors like adverse weather or storm-related changes to beach profiles or nesting substrates (cf. [30]). Nevertheless, fledging rates in 2023 were the highest since monitoring began in 2011, supported by more nesting attempts, larger brood sizes, and higher brood success. Storm-driven increases in surface prey abundance (e.g., storm-deposited wrack) and improved access to tidal flats [14,32] may have contributed to elevated brood success, as the area of these critical foraging habitats expanded somewhat after PTC Fiona [55]. These observations highlight a short-term preference for familiar nesting sites despite new habitat availability and suggest longer-term monitoring is needed, supported by more precise data (e.g., annual LiDAR, invertebrate sampling) to evaluate habitat suitability.
Most fledglings produced on PEI during the study period were from the northwestern uninhabited barrier islands, which hosted only a third of island-wide breeding pairs annually. These islands consistently had higher nest success, even for presumed renests following 2023 flooding, which are typically less successful elsewhere. This success likely reflects lower human presence, fewer predators (e.g., [71,72]), and more expansive washovers compared to mainland-connected sites, which provide ideal nesting conditions. The most expansive and productive washover habitats, termed washover sheets [73], can form rapidly after storms or gradually through the merging of adjacent terminal spits, as evident in long-term satellite imagery over these sites. However, our classification results show that severe storms also cause full dune breaching that can replace these critical habitat features with bay channels and narrow terminal spits that are less suitable for nesting. While expansive washover sheets persisted after Fiona and supported many fledglings, their long-term viability is uncertain when faced with increasingly severe storms [53,74]. Factors like geomorphology, littoral sediment budgets, storm characteristics, and relative sea-level rise (RSLR) all influence the evolution of these habitats [75]. Frequent overwash and rapid RSLR could lower beach profiles, increase flood risks, and lead to barrier island narrowing or drowning, particularly where longshore sediment transport depletes nearshore deposits [75], as is the case for these shorelines [49]. Any decline in the suitability or extent of these critical habitat features could compromise the productivity of this population, highlighting the need for morphodynamic studies to assess future changes.

4.2. Habitat Influences on Nest Success

From 2020 to 2023, most broods (96%) produced fledglings, with average brood success exceeding 2.4 fledglings per brood, emphasizing hatch success as critical for population recovery [42]. In general, hatched nests were located on significantly wider beaches (consistent with [41,69,70,76]) and slightly farther from the ocean ITZ than failed nests, conditions linked to reduced risks of both flooding and predation [12,44,77]. Washovers also had higher hatch success than other zones, likely due to their expansive areas allowing nests to be farther from ITZs and benefitting from superior prey availability [28,59,78,79] and predator sightlines [24]. Our models identified proximity to the ocean ITZ as a key predictor of predation risk, with nests closer to shorelines being more vulnerable, consistent with knowledge that predators forage along wrack lines after high tides [44].
Predation rates were higher for renests than original nests, potentially due to increased predator and human activity later in the season [11,77]. These risks may also contribute to the lower hatch success observed among open-access beaches relative to restricted-access sites. The success of the latter is mainly attributable to above-average success rates at uninhabited barrier island sites (53% nests hatched) where access is physically impeded, rather than at mainland PEINP sites (34% nests hatched) where access is officially restricted. We speculate that despite lower disturbance from humans, closed PEINP sites may experience average or elevated predator activity due to their proximity to human-use areas and their value as predator refuges. In contrast, closed barrier island sites likely have fewer terrestrial predators, which, combined with low human presence and ideal habitat features, leads to higher success rates. Interestingly, nests’ distances to beach access points did not appear to affect success rates, suggesting these measures may not truly reflect disturbance levels.
Nests were significantly more likely to flood on terminal end habitats, both pre- and post-storm, indicating that these areas were inherently flood prone (cf. [80]). Increased flooding after PTC Fiona may reflect storm-induced beach profile lowering (cf. [81,82]) and/or RSLR-amplified spring tides [83]. Post-storm flooding occurred across all zones, including baysides and washovers, where flooding was rare before 2023. These trends suggest storm-induced changes heightened flood risks, warranting further research into how Fiona altered beach profiles across PEI nesting sites.

4.3. Limitations and Future Work

Our study has several limitations worth noting. First, the absence of HHWLT contours in image classification may introduce minor inaccuracies in net change calculations due to slight tidal variations. We minimized this risk by selecting images taken at similar high tidal stages and measuring beach areas accordingly. Importantly, these minor inaccuracies do not affect the main focus of our classification analysis on landward-created habitats and their post-storm utilization. Second, our image classification approach does not account for beach slope/elevation, percent vegetation cover, substrate type, or disturbance levels, which potentially overestimates habitat availability (cf. [38,39]). While our goal was to map storm-driven changes in potential nesting habitat and productivity, our study does not offer a comprehensive habitat assessment that is nonetheless warranted to test longstanding assumptions of habitat suitability in the region. Third, using data from a single post-storm breeding season limits conclusions about population responses given the potential for delayed responses to storm-induced habitat changes. Thus, longer-term monitoring is needed to fully understand the effects of severe storms on this population.
Finally, potential detection bias may have influenced our nest data analyses. Since we relied on apparent nest success without adjusting for detection probabilities, nests in certain habitats, particularly those where early nest failure and lower detectability coincide, might be underrepresented. However, systematic monitoring and prior research suggesting minimal detectability improvements with increased survey effort on PEI [60] reduce the likelihood of substantial bias, and we consider its impact on our overall conclusions to be minimal. To enhance insights and the availability of data for studying this population, future research should prioritize (1) re-assessing nest detectability across more sites, (2) regularly acquiring high-resolution topographical data and aerial imagery to track habitat changes, (3) studying human beach visitation to evaluate disturbance impacts, and (4) investigating habitat enrichment efforts (e.g., cobble placement, beach nourishment of terminal habitats) and its effects on nest site selection and productivity.

5. Conclusions

Our study highlights landward expansions of potential PIPL nesting habitat on PEI via vegetation scouring, dune blowouts, and overwash following PTC Fiona. However, limited use of these new habitats, stable population numbers, and lower hatch success in the first post-storm breeding season suggest that the short-term benefits of storm-induced habitat creation were not fully realized. High site fidelity and unmeasured factors like topography, microhabitat features, and disturbance levels may have influenced this response. Nest-level analyses showed higher hatch success in washovers, sparse vegetation, and wide beaches, while sandspits, sandbars, pure sand habitats, and backshore zones had lower success. Terminal end habitats were particularly flood-prone, with post-storm suitability appearing further compromised by increased flood-related losses. Northwestern uninhabited barrier islands remain critical sources of productivity for this population, but rising sea levels and increased storminess could reduce their suitability, forcing PIPL into more disturbed and lower-quality habitats. To better understand these dynamics, future research should focus on incorporating data on topography, microhabitat features, disturbance pressures, and prey dynamics into morphodynamic modeling of habitat changes to inform conservation strategies for this endangered population.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/rs16244764/s1. Figure S1: Example of pixel-based landcover classification on Planet (3-m) satellite imagery used in this study. Figure S2: Example of habitat metrics derived for each nest sample (n = 146) from 2020-2023 using high-resolution satellite imagery. Figure S3: Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over all critical sandspit/bar habitats for piping plovers on PEI. Black circles on island map indicates nesting sites with no data (ND).

Author Contributions

Conceptualization, R.G. and X.W.; methodology, R.G. and X.W.; software, R.G.; validation, R.G.; formal analysis, R.G. and X.W.; investigation, R.G. and X.W.; resources, R.G. and X.W.; data curation, R.G.; writing—original draft preparation, R.G.; writing—review and editing, R.G. and X.W.; visualization, R.G.; supervision, X.W.; project administration, X.W.; funding acquisition, R.G. and X.W. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Mitacs, Island Nature Trust, Planet Labs, the Natural Science and Engineering Research Council of Canada, the New Frontiers in Research Fund, and the Government of Prince Edward Island.

Data Availability Statement

The nesting data used by this study are available from Environment and Climate Change Canada. Restrictions apply to the availability of these data, which were used under license for this study. The codes used to perform image classification on Google Earth Engine are available at https://github.com/ryan-guild/PEI_change_classification.git (accessed on 13 December 2024).

Acknowledgments

We thank the staff at Island Nature Trust and Prince Edward Island National Park for their dedicated survey efforts and thoughtful discussions that helped our research. We also thank four anonymous reviewers for their helpful comments on our manuscript draft.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Seneviratne, S.I.; Zhang, X.; Adnan, M.; Badi, W.; Dereczynski, C.; Di Luca, A.; Ghosh, S.; Iskandar, I.; Kossin, J.; Lewis, S.; et al. Weather and climate extreme events in a changing climate. In Climate Change 2021: The Physical Science Basis; Contribution of WGI to the AR6 of the IPCC; Masson-Delmotte, V., Zhai, P., Pirani, A., Eds.; Cambridge University Press: Cambridge, UK, 2021; pp. 1513–1766. [Google Scholar]
  2. Waycott, M.; Duarte, C.M.; Carruthers, T.J.B.; Orth, R.J.; Dennison, W.C.; Olyarnik, S.; Calladine, A.; Fourqurean, J.W.; Heck, K.L.; Hughes, A.R.; et al. Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proc. Natl. Acad. Sci. USA 2009, 106, 12377–12381. [Google Scholar] [CrossRef]
  3. Orth, R.J.; Carruthers, T.J.B.; Dennison, W.C.; Duarte, C.M.; Fourqurean, J.W.; Heck, K.L.; Hughes, A.R.; Kendrick, G.A.; Kenworthy, W.J.; Olyarnik, S.; et al. A global crisis for seagrass ecosystems. BioScience 2006, 56, 987. [Google Scholar] [CrossRef]
  4. Lagomasino, D.; Fatoyinbo, T.; Castañeda-Moya, E.; Cook, B.D.; Montesano, P.M.; Neigh, C.S.R.; Corp, L.A.; Ott, L.E.; Chavez, S.; Morton, D.C. Storm surge and ponding explain mangrove dieback in southwest Florida following Hurricane Irma. Nat. Commun. 2021, 12, 4003. [Google Scholar] [CrossRef]
  5. Anthony, A.; Atwood, J.; August, P.; Byron, C.; Cobb, S.; Foster, C.; Fry, C.; Gold, A.; Hagos, K.; Heffner, L.; et al. Coastal lagoons and climate change: Ecological and social ramifications in U.S. Atlantic and Gulf Coast ecosystems. Ecol. Soc. 2009, 14, 8. [Google Scholar] [CrossRef]
  6. Day, J.W.; Agboola, J.; Chen, Z.; D’Elia, C.; Forbes, D.L.; Giosan, L.; Kemp, P.; Kuenzer, C.; Lane, R.R.; Ramachandran, R.; et al. Approaches to defining deltaic sustainability in the 21st century. Estuar. Coast. Shelf Sci. 2016, 183, 275–291. [Google Scholar] [CrossRef]
  7. Yannic, G.; Aebischer, A.; Sabard, B.; Gilg, O. Complete breeding failures in ivory gull following unusual rainy storms in North Greenland. Polar Res. 2014, 33, 22749. [Google Scholar] [CrossRef]
  8. Büẞer, C.; Kahles, A.; Quillfeldt, P. Breeding success and chick provisioning in Wilson’s storm-petrels (Oceanites oceanicus) over seven years: Frequent failures due to food shortage and entombment. Polar Biol. 2004, 27, 613–622. [Google Scholar] [CrossRef]
  9. Mallory, M.L.; Gaston, A.J.; Forbes, M.R.; Gilchrist, H.G. Influence of weather on reproductive success of northern fulmars in the Canadian high Arctic. Polar Biol. 2009, 32, 529–538. [Google Scholar] [CrossRef]
  10. Wilcox, L.R. A twenty year banding study of the piping plover. Auk 1959, 76, 129–152. [Google Scholar] [CrossRef]
  11. Cohen, J.B.; Houghton, L.M.; Fraser, J.D. Nesting density and reproductive success of piping plovers in response to storm- and human-created habitat changes. Wildl. Monogr. 2009, 173, 1–24. [Google Scholar] [CrossRef]
  12. Walker, K.M.; Fraser, J.D.; Catlin, D.H.; Ritter, S.J.; Robinson, S.G.; Bellman, H.A.; DeRose-Wilson, A.; Karpanty, S.M.; Papa, S.T. Hurricane Sandy and engineered response created habitat for a threatened shorebird. Ecosphere 2019, 10, e02771. [Google Scholar] [CrossRef]
  13. Catlin, D.H.; Fraser, J.D.; Felio, J.H. Demographic responses of piping plovers to habitat creation on the Missouri river. Wildl. Monogr. 2015, 192, 1–42. [Google Scholar] [CrossRef]
  14. Hunt, K.L.; Fraser, J.D.; Karpanty, S.M.; Catlin, D.H. Body condition of piping plovers (Charadrius melodus) and prey abundance on flood-created habitat on the Missouri River, USA. Wilson J. Ornithol. 2017, 129, 754–764. [Google Scholar] [CrossRef]
  15. Hunt, K.L.; Fraser, J.D.; Friedrich, M.J.; Karpanty, S.M.; Catlin, D.H. Demographic response of piping plovers suggests that engineered habitat restoration is no match for natural riverine processes. Condor 2018, 120, 149–165. [Google Scholar] [CrossRef]
  16. Dorsey, S.S.L. Factors Affecting Piping Plover Nest Site Selection Following Landscape and Predator Community Changes. MSc Thesis, Virginia Polytech Institute & State University, Blacksburg, VA, USA, 2024; 107p. [Google Scholar]
  17. Robinson, S.; Fraser, J.; Catlin, D.; Karpanty, S.; Altman, J.; Boettcher, R.; Holcomb, K.; Huber, C.; Hunt, K.; Wilke, A. Irruptions: Evidence for breeding season habitat limitation in Piping Plover (Charadrius melodus). Avian Conserv. Ecol. 2019, 14, 19. [Google Scholar] [CrossRef]
  18. Robinson, S.G.; Gibson, D.; Riecke, T.V.; Fraser, J.D.; Bellman, H.A.; DeRose-Wilson, A.; Karpanty, S.M.; Walker, K.M.; Catlin, D.H. Piping Plover population increase after Hurricane Sandy mediated by immigration and reproductive output. Condor 2020, 122, duaa041. [Google Scholar] [CrossRef]
  19. Schupp, C.A.; Winn, N.T.; Pearl, T.L.; Kumer, J.P.; Carruthers, T.J.B.; Zimmerman, C.S. Restoration of overwash processes creates piping plover (Charadrius melodus) habitat on a barrier island (Assateague Island, Maryland). Estuar. Coast. Shelf Sci. 2013, 113, 11–20. [Google Scholar] [CrossRef]
  20. McIntyre, A.F.; Heath, J.A.; Jannsen, J. Trends in piping plover reproduction at Jones Beach State Park, NY, 1995–2007. Northeast. Nat. 2010, 17, 493–504. [Google Scholar] [CrossRef]
  21. Swaisgood, R.R.; Nordstrom, L.A.; Schuetz, J.G.; Boylan, J.T.; Fournier, J.J.; Shemai, B. A management experiment evaluating nest-site selection by beach-nesting birds. J. Wildl. Manag. 2017, 82, 192–201. [Google Scholar] [CrossRef]
  22. Le Fer, D.; Fraser, J.; Kruse, C. Piping plover foraging-site selection on the Missouri River. Waterbirds 2008, 31, 587–592. [Google Scholar] [CrossRef]
  23. Hecht, A.; Melvin, S.M. Population trends of Atlantic Coast piping plovers, 1986–2006. Waterbirds 2009, 32, 64–72. [Google Scholar] [CrossRef]
  24. Mahoney, M. Assessing Short Term Coastal Dynamics to Predict Impacts from Severe Weather Events on Piping Plover Nesting Preferences. MSc Thesis, Mount Allison University, Sackville, NB, Canada, 2015. [Google Scholar]
  25. Maslo, B.; Handel, S.N.; Pover, T. Restoring beaches for Atlantic Coast piping plovers (Charadrius melodus): A classification and regression tree analysis of nest-site selection. Restor. Ecol. 2011, 19, 194–203. [Google Scholar] [CrossRef]
  26. Maslo, B.; Leu, K.; Pover, T.; Weston MAGilby, B.L.; Schlacher, T.A. Optimizing conservation benefits for threatened beach fauna following severe natural disturbances. Sci. Total Environ. 2019, 649, 661–671. [Google Scholar] [CrossRef] [PubMed]
  27. Robinson, S.G.; Walker, K.M.; Bellman, H.A.; Gibson, D.; Catlin, D.H.; Karpanty, S.M.; Ritter, S.J.; Fraser, J.D. Piping plover chick ecology following landscape-level disturbance. J. Wildl. Manag. 2023, 87, e22325. [Google Scholar] [CrossRef]
  28. Zeigler, S.L.; Gutierrez, B.T.; Sturdivant, E.J.; Catlin, D.H.; Fraser, J.D.; Hecht, A.; Karpanty, S.M.; Plant, N.G.; Thieler, E.R. Using a Bayesian network to understand the importance of coastal storms and undeveloped landscapes for the creation and maintenance of early successional habitat. PLoS ONE 2019, 14, e0209986. [Google Scholar] [CrossRef] [PubMed]
  29. Zeigler, S.L.; Gutierrez, B.T.; Hecht, A.; Plant, N.G.; Sturdivant, E.J. Piping plovers demonstrate regional differences in nesting habitat selection patterns along the U.S. Atlantic coast. Ecosphere 2021, 12, e03418. [Google Scholar] [CrossRef]
  30. Cohen, J.B. Factors Limiting Piping Plover Nesting Pair Density and Reproductive Output on Long Island, New York. Ph.D. Dissertation, Faculty of the Virginia Polytechnic Institute and State University, Blacksburg, VA, USA, 2005; 251p. [Google Scholar]
  31. Houghton, L.M. Piping Plover Population Dynamics and Effects of Beach Management Practices on Piping Plovers at West Hampton Dunes and Westhampton Beach, New York. Ph.D. Dissertation, Faculty of Virginia Polytechnic Institute and State University, Blacksburg, VA, USA, 2005; 176p. [Google Scholar]
  32. Bourque, N.R.; Villard, M.; Mazerolle, M.J.; Amirault-Langlais, D.; Tremblay, E.; Jolicoeur, S. Piping Plover response to coastal storms occurring during the nonbreeding season. Avian Conserv. Ecol. 2015, 10, 12. [Google Scholar] [CrossRef]
  33. Czaplewski, M.M.; Peyton, M.; Jenniges, J. Observation of hailstorm-caused mortality of least terns and piping plovers on the Niobrara River, Nebraska. Neb. Bird Rev. 2007, 76, 129–130. [Google Scholar]
  34. Gratto-Trevor, C.L.; Abbot, S. Conservation of piping plover (Charadrius melodus) in North America: Science, successes, and challenges. Can. J. Zool. 2011, 89, 401–418. [Google Scholar] [CrossRef]
  35. Cameron, J.E. Assessing Climate-Impact Risk to Piping Plover (Charadrius melodus melodus) Breeding Sites in Nova Scotia, Canada. MES Thesis, Dalhousie University, Halifax, NS, Canada, 2022; 170p. [Google Scholar]
  36. Amirault-Langlais, D.L.; Imlay, T.L.; Boyne, A.W. Dispersal patterns suggest two breeding populations of piping plovers in Eastern Canada. Wilson J. Ornithol. 2014, 126, 352–359. [Google Scholar] [CrossRef]
  37. Calvert, A.M.; Amirault-Langlais, D.L.; Shaffer, F.; Elliot, R.; Hanson, A.; McKnight, J.; Taylor, P.D. Population assessment of an endangered shorebird: The piping plover (Charadrius melodus melodus) in Eastern Canada. Avian Conserv. Ecol. 2006, 1, 4. [Google Scholar] [CrossRef]
  38. Rioux, S.; Amirault-Langlais, D.L.; Shaffer, F. Piping Plovers make decisions regarding dispersal based on personal and public information in a variable coastal ecosystem. J. Field Ornithol. 2011, 82, 32–43. [Google Scholar] [CrossRef]
  39. NSDLF (Nova Scotia Department of Lands and Forestry). Recovery Plan for Piping plover (Charadrius melodus melodus) in Nova Scotia [Final]; Nova Scotia Endangered Species Act Recovery Plan Series; Environment Canada: Ottawa, ON, Canada, 2021; 304p. [Google Scholar]
  40. ECCC (Environment and Climate Change Canada). Recovery Strategy (Amended) and Action Plan for the Piping Plover melodus subspecies (Charadrius melodus melodus) in Canada [Proposed]; Species at Risk Act Recovery Strategy Series; Environment and Climate Change Canada: Ottawa, ON, Canada, 2021; viii + 124p. [Google Scholar]
  41. Murphy, K.A. The Impacts of Human Recreational Activities, Habitat Quality and Weather Conditions on the Foraging Behaviour and Fledging Success of Breeding Piping Plovers (Charadrius melodus). MSc Thesis, St. Mary’s University, Halifax, NS, Canada, 2007; 147p. [Google Scholar]
  42. Guild, R.G.; Wang, X.; Hirtle, S.; Mader, S. Spatiotemporal and weather effects on the reproductive success of piping plovers on Prince Edward Island, Canada. Ecol. Evol. 2024, 14, e11581. [Google Scholar] [CrossRef]
  43. Boyne, A.W.; Amirault-Langlais, D.L.; McCue, A.J. Characteristics of piping plover nesting habitat in the Canadian maritime provinces. Northeast. Nat. 2014, 21, 164–173. [Google Scholar] [CrossRef]
  44. Thomas, L.; Lajeunesse, D. Recovery efforts for Piping Plovers in Prince Edward Island National Park of Canada. In Proceedings of the Species at Risk 2004 Pathways to Recovery Conference, Victoria, BC, Canada, 2–6 March 2024; Hooper, T.D., Ed.; 12p. [Google Scholar]
  45. Stewart, J.I. A Multiscale Habitat Suitability Assessment for Piping Plover (Charadrius melodus) on Prince Edward Island. MES Thesis, Dalhousie University, Halifax, NS, Canada, 2004; 189p. [Google Scholar]
  46. Davis, K.L.; Schoenemann, K.L.; Catlin, D.H.; Hunt, K.L.; Friedrich, M.J.; Ritter, S.J.; Fraser, J.D.; Karpanty, S.M. Hatch-year piping plover (Charadrius melodus) prospecting and habitat quality influence second-year nest site selection. Auk 2017, 134, 92–103. [Google Scholar] [CrossRef]
  47. Shaw, J.; Taylor, R.B.; Forbes, D.L.; Ruz, M.H.; Solomon, S. Sensitivity of the Coasts of Canada to Sea-Level Rise; Report for Geological Survey of Canada Bulletin 505; Natural Resources Canada: Ottawa, ON, Canada, 1998; 96p. [Google Scholar]
  48. Forbes, D.L.; Parkes, G.S.; Manson, G.K.; Ketch, L.A. Storms and shoreline retreat in the southern Gulf of St. Lawrence. Mar. Geol. 2004, 210, 169–204. [Google Scholar] [CrossRef]
  49. Davies, M. Geomorphic Shoreline Classification of Prince Edward Island; Report Prepared by Coldwater Consulting Ltd. for Prince Edward Island’s Department of Energy, Environment & Forestry: Charlottetown, PE, Canada, 2011; 70p. [Google Scholar]
  50. Mathew, S.; Davidson-Arnott, R.G.; Ollerhead, J. Evolution of a beach–dune system following a catastrophic storm overwash event: Greenwich Dunes, Prince Edward Island, 1936–2005. Can. J. Earth Sci. 2010, 47, 273–290. [Google Scholar] [CrossRef]
  51. Parkes, G.S.; Forbes, D.L.; Ketch, L.A. Sea-level Rise. In Coastal Impacts of Climate Change and Sea-Level Rise on Prince Edward Island; Forbes, D.L., Shaw, R.W., Eds.; Geological Survey of Canada: Ottawa, ON, Canada, 2002; Open File 4261. [Google Scholar]
  52. Pasch, R.J.; Reinhart, B.J.; Alaka, L. National Hurricane Center Tropical Cyclone Report: Hurricane Fiona; Joint Publication by National Oceanic and Atmospheric Administration & National Weather Service: Miami, FL, USA, 2023; 60p. Available online: https://www.nhc.noaa.gov/data/tcr/AL072022_Fiona.pdf (accessed on 5 September 2024).
  53. Ollerhead, J.; Davidson-Arnott, R.; Bauer, B.O. The Importance of Coastal Foredunes as a Nature-Based Solution for Shoreline Protection: What Hurricane Fiona Tells Us. Bulletin of the Canadian Meteorological and Oceanographic Society. 2022. Available online: https://bulletin.cmos.ca/the-importance-of-coastal-foredunes-as-a-nature-based-solution-for-shoreline-protection-what-hurricane-fiona-tells-us/ (accessed on 5 June 2023).
  54. Jardine, D. Post Tropical Storm Fiona Highwater Mark and Shoreline Erosion Field Notes with Photos. Report Prepared for the Government of Prince Edward Island by DE Jardine Consulting. 2023. 117p. Available online: https://www.princeedwardisland.ca/sites/default/files/publications/post_tropical_storm_fiona_highwater_mark_and_shoreline_erosion.pdf (accessed on 1 September 2024).
  55. Pang, T.; Wang, X.; Basheer, S.; Guild, R. Landcover-based detection of rapid impacts of extreme storm on coastal landscape. Sci. Total Environ. 2024, 932, 173099. [Google Scholar] [CrossRef]
  56. Zharikov, Y.; Skilleter, G.A.; Loneragan, N.R.; Taranto, T.; Cameron, B.E. Mapping and characterising subtropical estuarine landscapes using aerial photography and GIS for potential application in wildlife conservation and management. Biol. Conserv. 2005, 125, 87–100. [Google Scholar] [CrossRef]
  57. Uhl, F.; Græsdal Rasmussen, T.; Oppelt, N. Classification ensembles for beach cast and drifting vegetation mapping with Sentinel-2 and PlanetScope. Geosciences 2021, 12, 15. [Google Scholar] [CrossRef]
  58. McAllister, E.; Payo, A.; Novallino, A.; Dolphin, T.; Medina-Lopez, E. Shoreline extraction using high resolution satellite imagery at Start Bay, UK. Proc. IAHR World Congr. 2022, 2022, 5811–5820. [Google Scholar]
  59. Robinson, S.; Bellman, H.; Walker, K.; Catlin, D.; Karpanty, S.; Ritter, S.; Fraser, J. Adult piping plover habitat selection varies by behaviour. Ecosphere 2021, 12, e03870. [Google Scholar] [CrossRef]
  60. Elliot-Smith, E.; Bidwell, M.; Holland, A.E.; Haig, S.M. Data from the 2011 International Piping Plover Census; Report Prepared Jointly by the U.S. Geological Survey and Environment Canada; Geological Survey Data Series 922; Geological Survey and Environment Canada: Reston, VA, USA, 2015; 296p. [Google Scholar] [CrossRef]
  61. R Core Team. R: A Language and Environment for Statistical Computing; R Foundational for Statistical Computing: Vienna, Austria, 2024; Available online: https://www.R-project.org (accessed on 5 November 2024).
  62. Wickham, H.; François, R.; Henry, L.; Müller, K.; Vaughan, D. dplyr: A Grammar of Data Manipulation. 2023. Available online: https://CRAN.R-project.org/package=dplyr (accessed on 5 October 2024).
  63. Wickham, H.; Vaughan, D.; Girlich, M. tidyr: Tidy Messy Data. R Package Version 1.3.1. 2024. Available online: https://github.com/tidyverse/tidyr (accessed on 5 October 2024).
  64. Wickham, H. ggplot2: Elegant Graphics for Data Analysis; Springer: New York, NY, USA, 2016; Available online: https://ggplot2.tidyverse.org (accessed on 5 October 2024).
  65. Bartoń, K. MuMIn: Multi-Model Inference. R Package Version 1.48.4. 2024. Available online: https://CRAN.R-project.org/package=MuMIn (accessed on 2 October 2024).
  66. Robin, X.; Turck, N.; Hainard, A.; Tiberti, N.; Lisacek, F.; Sanchez, J.C.; Müller, M. pROC: An open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinform. 2011, 12, 77. [Google Scholar] [CrossRef] [PubMed]
  67. Catlin, D.H.; Fraser, J.D.; Felio, J.H.; Cohen, J.B. Piping plover habitat selection and nest success on natural, managed, and engineered sandbars. J. Wildl. Manag. 2011, 75, 305–310. [Google Scholar] [CrossRef]
  68. Gieder, K.; Karpanty, S.M.; Fraser, J.D.; Catlin, D.H.; Gutierrez, B.T.; Plant, N.G.; Turecek, A.M.; Thieler, E.R. A Bayesian network approach to predicting nest presence of the federally threatened piping plover (Charadrius melodus) using barrier island features. Ecol. Model. 2014, 276, 38–50. [Google Scholar] [CrossRef]
  69. Espie, R.H.; James, P.C.; Brigham, R. The effects of flooding on piping plover Charadrius melodus reproductive success at Lake Diefenbaker, Saskatchewan, Canada. Biol. Conserv. 1998, 86, 215–222. [Google Scholar] [CrossRef]
  70. Prindiville-Gaines, E.; Ryan, M.R. Piping plover habitat use and reproductive success in North Dakota. J. Wildl. Manag. 1988, 52, 266. [Google Scholar] [CrossRef]
  71. Ogden, J.; Dowding, J.E. Population estimates and conservation of the New Zealand dotterel (Charadrius obscurus) on Great Barrier Island, New Zealand. Notornis 2013, 60, 210–223. [Google Scholar]
  72. Johnston-González, R.; Abril, E. Predation risk and resource availability explain roost locations of Whimbrel Numenius phaeopus in a tropical mangrove delta. Ibis 2019, 161, 839–853. [Google Scholar] [CrossRef]
  73. Donnelly, C.; Kraus, N.; Larson, M. State of knowledge on measurement and modeling of coastal overwash. J. Coast. Res. 2006, 224, 965–991. [Google Scholar] [CrossRef]
  74. Kossin, J.P.; Knapp, K.R.; Olander, T.L.; Velden, C.S. Global increase in major tropical cyclone exceedance probability over the past four decades. Proc. Natl. Acad. Sci. USA 2020, 117, 11975–11980. [Google Scholar] [CrossRef]
  75. Ashton, A.D.; Lorenzo-Trueba, J. Morphodynamics of barrier response to sea-level rise. In Barrier Dynamics and Response to Changing Climate; Moore, L.J., Murray, A.B., Eds.; Springer International Publishing: Berlin/Heidelberg, Germany, 2018; pp. 277–304. [Google Scholar] [CrossRef]
  76. Harris, W.C.; Duncan, D.C.; Franken, R.J.; McKinnon, D.T.; Dundas, H.A. Reproductive success of piping plovers at Big Quill Lake, Saskatchewan. Wilson Bull. 2005, 117, 165–171. [Google Scholar] [CrossRef]
  77. Burger, J. Physical and social determinants of nest-site selection in piping plover in New Jersey. Condor 1987, 89, 811. [Google Scholar] [CrossRef]
  78. Kumer, J. Status of the endangered piping plover, Charadrius melodus, population in the Maryland coastal bays. In Maryland’s Coastal Bays: Ecosystem Health Assessment 2004; Wazniak, C.E., Hall, M.R., Eds.; Maryland Department of Natural Resources, Tidewater Ecosystem Assessment: Annapolis, MD, USA, 2004; pp. 8.94–8.99. [Google Scholar]
  79. Nordstrom, L.H.; Ryan, M.R. Invertebrate abundance at occupied and potential piping plover nesting beaches: Great Plains alkali wetlands vs. the Great Lakes. Wetlands 1996, 16, 429–435. [Google Scholar] [CrossRef]
  80. Casagrande, G.; Bezzi, A.; Fracaros, S.; Martinucci, D.; Pillon, S.; Salvador, P.; Sponza, S.; Fontolan, G. Quantifying transgressive coastal changes using UAVs: Dune migration, overwash recovery, and barrier flooding assessment and interferences with human and natural assets. J. Mar. Sci. Eng. 2023, 11, 1044. [Google Scholar] [CrossRef]
  81. Brenner, O.T.; Lentz, E.E.; Hapke, C.J.; Henderson, R.E.; Wilson, K.E.; Nelson, T.R. Characterizing storm response and recovery using the beach change envelope: Fire Island, New York. Geomorphology 2018, 300, 189–202. [Google Scholar] [CrossRef]
  82. Walker, I.J.; Davidson-Arnott, R.G.; Bauer, B.O.; Hesp, P.A.; Delgado-Fernandez, I.; Ollerhead, J.; Smyth, T.A. Scale-dependent perspectives on the geomorphology and evolution of beach-dune systems. Earth Sci. Rev. 2017, 171, 220–253. [Google Scholar] [CrossRef]
  83. Ezer, T. Analysis of the changing patterns of seasonal flooding along the U.S. East Coast. Ocean Dyn. 2020, 70, 241–255. [Google Scholar] [CrossRef]
Remotesensing 16 04764 g001
Figure 1. Map of all nesting sites on PEI with breeding activity since 2011.
Figure 1. Map of all nesting sites on PEI with breeding activity since 2011.
Remotesensing 16 04764 g001
Remotesensing 16 04764 g002
Figure 2. Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical barrier island habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circle on the island map indicates nesting site with no data (ND).
Figure 2. Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical barrier island habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circle on the island map indicates nesting site with no data (ND).
Remotesensing 16 04764 g002
Remotesensing 16 04764 g003
Figure 3. Classified landcover pre-storm (A) and post-storm (B) and total change in open sand area (C) over critical sandspit/bar habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND). Additional classified sandspit/bar nesting sites are displayed in Figure S3.
Figure 3. Classified landcover pre-storm (A) and post-storm (B) and total change in open sand area (C) over critical sandspit/bar habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND). Additional classified sandspit/bar nesting sites are displayed in Figure S3.
Remotesensing 16 04764 g003
Remotesensing 16 04764 g004
Figure 4. Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical mainland beach habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND).
Figure 4. Classified landcover pre-storm (A) and 1-year post-storm (B) and total change in open sand area (C) over critical mainland beach habitats for PIPL on PEI. Counts of breeding pairs (BP) and fledglings (FL) from each site indicated by light and dark blue bars, respectively. Black circles on the island map indicate nesting sites with no data (ND).
Remotesensing 16 04764 g004
Remotesensing 16 04764 g005
Figure 5. Nest locations and outcomes during three breeding seasons preceding (top three panels) and the initial season following (fourth panel) PTC Fiona over Conway Sandhills, PEI. Classified change in dry and wet sand area after one-year post-storm depicted in bottom panel, with colours representing change classes (in Figure 2, Figure 3 and Figure 4).
Figure 5. Nest locations and outcomes during three breeding seasons preceding (top three panels) and the initial season following (fourth panel) PTC Fiona over Conway Sandhills, PEI. Classified change in dry and wet sand area after one-year post-storm depicted in bottom panel, with colours representing change classes (in Figure 2, Figure 3 and Figure 4).
Remotesensing 16 04764 g005
Remotesensing 16 04764 g006
Figure 6. Fledging rate across common nesting sites with at least three years of nesting attempts between 2020 and 2023. Horizontal line at 1.65 fledglings/pair indicates ECCC productivity target for the Eastern Canadian recovery unit. Vertical line distinguishes between pre- and post-storm breeding seasons.
Figure 6. Fledging rate across common nesting sites with at least three years of nesting attempts between 2020 and 2023. Horizontal line at 1.65 fledglings/pair indicates ECCC productivity target for the Eastern Canadian recovery unit. Vertical line distinguishes between pre- and post-storm breeding seasons.
Remotesensing 16 04764 g006
Remotesensing 16 04764 g007
Figure 7. Model-averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of binary hatch success (left) and data summaries of hatch outcomes across habitat metrics from 2020–2023 on PEI (right). Number in white indicates nest counts in each respective category; error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Figure 7. Model-averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of binary hatch success (left) and data summaries of hatch outcomes across habitat metrics from 2020–2023 on PEI (right). Number in white indicates nest counts in each respective category; error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Remotesensing 16 04764 g007
Remotesensing 16 04764 g008
Figure 8. Summaries of binary hatch success by year across habitat measures on PEI from 2020–2023. Number in white indicates the number of nests in each respective category. D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Figure 8. Summaries of binary hatch success by year across habitat measures on PEI from 2020–2023. Number in white indicates the number of nests in each respective category. D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Remotesensing 16 04764 g008
Remotesensing 16 04764 g009
Figure 9. Model averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of flooding and predation occurrences (left) and data summaries of nest outcomes across habitat metrics from 2020–2023 on PEI (right). Error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Figure 9. Model averaged coefficient estimates (log-odds scale) from the top-ranked logistic regression GLMs of flooding and predation occurrences (left) and data summaries of nest outcomes across habitat metrics from 2020–2023 on PEI (right). Error bars in GLM output display 95% confidence intervals (SE * 1.96). D2 ACCESS is represented as a categorical and continuous variable to convey complimentary insights.
Remotesensing 16 04764 g009
Table 1. Variables used in nest data analysis sourced from standardized annual monitoring (*) or derived from high-resolution satellite imagery (†).
Table 1. Variables used in nest data analysis sourced from standardized annual monitoring (*) or derived from high-resolution satellite imagery (†).
Habitat MetricAbbreviated TermVariable TypeDescription
Habitat Type *HABITATCategoricalNesting substrate; either cobble, open sand, or sparse vegetation
Renest Status *RENESTCategoricalConfirmed renest or confirmed/assumed original nest
Nest Outcome *-CategoricalHatched, flooded, predated, abandoned/buried, or unknown loss reason
Geomorphological Type †GEO (TYPE)CategoricalGeomorphology of nesting site; either barrier island, mainland beach/inlet, or sandspit/sandbar
Beach Zone †ZNCategoricalGeophysical region of beach; either backshore, foredune, washover, or bayside
Terminal End Habitat †TERMINAL HAB.CategoricalBinary; if nest is at the terminal end of a barrier island/sandspit/sandbar
Access to Low-Energy Shoreline Intertidal Zone †ACCESS LES MOSHCategoricalBinary; if nest is within 250 m of low-energy shoreline (bayside or tidal inlet) moist-sand habitat (MOSH)
Change in Habitat Area from Previous Year †Δ HABITAT (1-YR)CategoricalBinary; if nesting site experienced significant visible change in habitat area since last year
Recreational Access †REC. ACCESSCategoricalBinary; “closed” if access is either physically impeded (e.g., remote barrier island) or restricted by the National Park Authority during breeding season, else “open”
Beach Width †-ContinuousNesting beach width (m) measured from ocean high tide line to either dune toe, high tide of low-energy shoreline, or dense vegetation
Distance to Beach Access †D2 ACCESSContinuousLeast-cost path distance from nest to nearest beach access point for humans
Distance to Conspecific †D2 CONSPECIFICContinuousDistance from nest to nearest active nest
Distance to Previous Nest †D2 PREV NESTContinuousDistance from nest to previous nesting attempt by same breeding pair in current season
Distance to Low-Energy Shoreline Intertidal Zone †D2 LES ITZContinuousLeast-cost path distance from nest to the nearest low-energy shoreline intertidal zone
Distance to Ocean Intertidal Zone †D2 OCEAN ITZContinuousLeast-cost path distance from nest to the nearest ocean intertidal zone
Table 2. Mean ( X ¯ ) and standard deviation (σ) of producer’s accuracy (PA), user’s accuracy (UA), and overall accuracy (OA) for classification results across nesting sites.
Table 2. Mean ( X ¯ ) and standard deviation (σ) of producer’s accuracy (PA), user’s accuracy (UA), and overall accuracy (OA) for classification results across nesting sites.
Land Cover Classes
Dry SandWet SandVegetationWaterOverall
Accuracy
X ¯ ± σ X ¯ ± σ X ¯ ± σ X ¯ ± σ X ¯
Pre-stormPA0.98 ± 0.040.94 ± 0.050.99 ± 0.020.99 ± 0.01
UA0.98 ± 0.030.97 ± 0.030.99 ± 0.010.99 ± 0.02
OA 0.98 ± 0.04
Post-stormPA0.99 ± 0.020.95 ± 0.030.92 ± 0.100.98 ± 0.03
UA0.98 ± 0.030.96 ± 0.060.97 ± 0.040.94 ± 0.06
OA 0.96 ± 0.08
1-Year RevisitPA0.99 ± 0.010.95 ± 0.050.96 ± 0.050.99 ± 0.02
UA0.98 ± 0.020.98 ± 0.030.99 ± 0.010.96 ± 0.05
OA 0.98 ± 0.05
Table 3. Island-wide nest success and productivity rates on Prince Edward Island between 2020–2023. Horizontal line separates pre-storm (above) from post-storm (below) breeding seasons.
Table 3. Island-wide nest success and productivity rates on Prince Edward Island between 2020–2023. Horizontal line separates pre-storm (above) from post-storm (below) breeding seasons.
YearNo. Breeding Pairs 1No.
Fledglings		Fledging Rate 2No.
Nests 3		Average Hatch
Success 4		Average Clutch SizeAverage Brood SizeAverage No. Fledglings per Brood
202026381.463036.7%3.672.802.53
202124391.633138.7%3.843.132.44
202234531.564243.8%3.453.052.41
202332541.694835.9%3.173.723.00
1 Those discovered and monitored, excluding single (unpaired) birds. 2 Total number of fledglings divided by total number of monitored breeding pairs. 3 Includes renests. 4 Proportion of eggs hatched per nest averaged across all nesting attempts.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Guild, R.; Wang, X. Piping Plover Habitat Changes and Nesting Responses Following Post-Tropical Cyclone Fiona on Prince Edward Island, Canada. Remote Sens. 2024, 16, 4764. https://doi.org/10.3390/rs16244764

AMA Style

Guild R, Wang X. Piping Plover Habitat Changes and Nesting Responses Following Post-Tropical Cyclone Fiona on Prince Edward Island, Canada. Remote Sensing. 2024; 16(24):4764. https://doi.org/10.3390/rs16244764

Chicago/Turabian Style

Guild, Ryan, and Xiuquan Wang. 2024. "Piping Plover Habitat Changes and Nesting Responses Following Post-Tropical Cyclone Fiona on Prince Edward Island, Canada" Remote Sensing 16, no. 24: 4764. https://doi.org/10.3390/rs16244764

APA Style

Guild, R., & Wang, X. (2024). Piping Plover Habitat Changes and Nesting Responses Following Post-Tropical Cyclone Fiona on Prince Edward Island, Canada. Remote Sensing, 16(24), 4764. https://doi.org/10.3390/rs16244764

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop