Increasing marsh bird abundance in coastal wetlands of the Great Lakes, 2011–2021, likely caused by increasing water levels

ABSTRACT Wetlands of the Laurentian Great Lakes of North America (i.e., lakes Superior, Michigan, Huron, Erie, and Ontario) provide critical habitat for marsh birds. We used 11 years (2011–2021) of data collected by the Great Lakes Coastal Wetland Monitoring Program at 1,962 point-count locations in 792 wetlands to quantify the first-ever annual abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands throughout the entire Great Lakes. Nine species (50%) increased by 8–37% per year across all of the Great Lakes combined, whereas none decreased. Twelve species (67%) increased by 5–50% per year in at least 1 of the 5 Great Lakes, whereas only 3 species (17%) decreased by 2–10% per year in at least 1 of the lakes. There were more positive trends among lakes and species (n = 34, 48%) than negative trends (n = 5, 7%). These large increases are welcomed because most of the species are of conservation concern in the Great Lakes. Trends were likely caused by long-term, cyclical fluctuations in Great Lakes water levels. Lake levels increased over most of the study, which inundated vegetation and increased open water-vegetation interspersion and open water extent, all of which are known to positively influence abundance of most of the increasing species and negatively influence abundance of all of the decreasing species. Coastal wetlands may be more important for marsh birds than once thought if they provide high-lake-level-induced population pulses for species of conservation concern. Coastal wetland protection and restoration are of utmost importance to safeguard this process. Future climate projections show increases in lake levels over the coming decades, which will cause “coastal squeeze” of many wetlands if they are unable to migrate landward fast enough to keep pace. If this happens, less habitat will be available to support periodic pulses in marsh bird abundance, which appear to be important for regional population dynamics. Actions that allow landward migration of coastal wetlands during increasing lake levels by removing or preventing barriers to movement, such as shoreline hardening, will be useful for maintaining marsh bird breeding habitat in the Great Lakes. LAY SUMMARY We calculated the first-ever abundance estimates for 18 marsh-breeding bird species over 11 years (2011–2021) in coastal wetlands throughout all of the North American Great Lakes. Abundance of 9 (50%) species increased by large amounts, but none decreased, which is welcomed because the abundance of most species was already low. Increases were likely caused by rising water levels, which created high-quality marsh bird habitat consisting of patches of wet, standing vegetation (such as cattails) interspersed with pools of open water. Coastal wetlands may provide critical, periodic boosts to regional populations of marsh birds during rising water levels, but this may be threatened in the future because extremely high-water levels due to climate change could cause reductions in marsh bird habitat due to flooding. Protection and restoration of coastal wetlands and actions that allow landward migration of coastal wetlands by removing or preventing barriers to movement will be needed. RÉSUMÉ Les zones humides des Grands Lacs laurentiens d'Amérique du Nord (c.-à-d. les lacs Supérieur, Michigan, Huron, Érié et Ontario) constituent un habitat essentiel pour les oiseaux de marais. Nous avons utilisé 11 ans (2011-2021) de données recueillies par le Programme de surveillance des zones humides côtières des Grands Lacs à 1 962 points d'écoute dans 792 zones humides pour quantifier les tout premiers indices annuels d'abondance et les tendances chez 18 espèces d'oiseaux de marais dans les zones humides côtières de l'ensemble des Grands Lacs. Neuf espèces (50 %) ont augmenté de 8 à 37 % par année dans l'ensemble des Grands Lacs, tandis qu'aucune n'a diminué. Douze espèces (67 %) ont augmenté de 5 à 50 % par année dans au moins un des cinq Grands Lacs, alors que seulement trois espèces (17 %) ont diminué de 2 à 10 % par année dans au moins un des lacs. Il y avait plus de tendances positives parmi les lacs et les espèces (n = 34, 48 %) que de tendances négatives (n = 5, 7 %). Ces fortes augmentations sont les bienvenues car la plupart des espèces sont préoccupantes sur le plan de la conservation dans les Grands Lacs. Les tendances ont probablement été causées par des fluctuations cycliques à long terme des niveaux d'eau des Grands Lacs. Le niveau des lacs a augmenté pendant la majeure partie de l'étude, ce qui a inondé la végétation et augmenté la juxtaposition de l'eau libre et de la végétation, ainsi que l'étendue de l'eau libre, qui sont reconnus pour influencer positivement l'abondance de la plupart des espèces en augmentation et d'influencer négativement l'abondance de toutes les espèces en diminution. Les zones humides côtières peuvent être plus importantes pour les oiseaux de marais qu'on ne le pensait auparavant si elles fournissent des impulsions démographiques induites par le niveau élevé des lacs pour les espèces dont la conservation est préoccupante. La protection et la restauration des zones humides côtières sont d'une importance capitale pour préserver ce processus. Les projections climatiques futures indiquent une augmentation des niveaux d'eau au cours des prochaines décennies, ce qui entraînera une « compression côtière» de plusieurs zones humides si elles ne sont pas en mesure de migrer vers l'intérieur des terres assez rapidement pour suivre le rythme. Dans ce cas, il y aura moins d'habitat disponible pour soutenir les pulsations périodiques de l'abondance des oiseaux de marais, qui semblent importantes pour la dynamique des populations régionales. Les mesures qui permettent la migration vers l'intérieur des terres des zones humides côtières lorsque le niveau des lacs augmente, en éliminant ou en empêchant la formation d'obstacles au mouvement, comme le durcissement des rives, seront utiles pour maintenir l'habitat de reproduction des oiseaux de marais dans les Grands Lacs.


LAY SUMMARY
• We calculated the first-ever abundance estimates for 18 marsh-breeding bird species over 11 years (2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021) in coastal wetlands throughout all of the North American Great Lakes.• Abundance of 9 (50%) species increased by large amounts, but none decreased, which is welcomed because the abundance of most species was already low.• Increases were likely caused by rising water levels, which created high-quality marsh bird habitat consisting of patches of wet, standing vegetation (such as cattails) interspersed with pools of open water.• Coastal wetlands may provide critical, periodic boosts to regional populations of marsh birds during rising water levels, but this may be threatened in the future because extremely high-water levels due to climate change could cause reductions in marsh bird habitat due to flooding.• Protection and restoration of coastal wetlands and actions that allow landward migration of coastal wetlands by removing or preventing barriers to movement will be needed.

INTRODUCTION
Marsh bird habitat declined in the Laurentian Great Lakes basin of North America (i.e., in the watersheds of lakes Superior, Michigan, Huron, Erie, and Ontario) by 24% between 1984 and 2021 (Amani et al. 2022), and up to 90% of wetlands used by marsh birds has been lost in parts of the region since the 1800s (Ducks Unlimited Canada 2010).Despite these losses, 70,700 km 2 of marsh bird habitat remained in the Great Lakes basin as of 2021, comprising an estimated 14% of marsh bird habitat across the conterminous USA and Canada (Homer et al. 2020, Mahdianpari et al. 2021).Approximately 2,200 km 2 or 3% of Great Lakes basin wetlands are classified as coastal wetlands (United States Environmental Protection Agency 2022), which are defined as wetlands under direct hydrologic influence from Great Lakes waters or their connecting river systems (McKee et al. 1992).Some marsh bird species occur predominantly within coastal wetlands, or at greater densities within coastal wetlands, compared to other wetlands (Tozer 2020, Tozer et al. 2020, Studholme et al. 2023).Given their ability to support healthy wildlife populations, coastal wetlands typically have higher conservation value compared to other wetlands (Government of Ontario 2020, Ramsar Sites Information Service 2023).Despite the importance of coastal wetlands, over 38% (60 km 2 ) of wetland losses to development in the Great Lakes basin between 1992 and 2001 occurred within 10 km of a coastal area and most within 1 km (Wolter et al. 2006).Thus, Great Lakes wetlands, particularly coastal wetlands, act as critical but vulnerable reservoirs of habitat for migrating and breeding marsh birds throughout their annual cycles (Prince et al. 1992, Ross et al. 2012, Grand et al. 2020, Tozer et al. 2020).
Due to a suite of anthropogenic threats and stressors, especially in lakes Michigan, Erie, and Ontario (Allan et al. 2013), at least 16 marsh bird priority species in the Great Lakes region require immediate conservation action and monitoring to assess their status (Collins andSmith 2014, Soulliere et al. 2018).This diverse group of species includes Least Bittern (Ixobrychus exilis), which nests and feeds in dense, emergent marsh vegetation, and is designated as threatened or endangered in 7 of the 9 (78%) states and 1 province bordering the Great Lakes (COSEWIC 2009).It also includes Black Tern (Chlidonias niger), which nests colonially on floating substrates in moderately dense marsh vegetation, feeds over relatively open water, and is 90% less common as a breeder in the Great Lakes now compared to 30 years ago (Wyman andCuthbert 2016, 2017).Most of the species are sensitive to anthropogenic stressors like wetland habitat reduction, urban development, and agriculture in nearby landscapes (Howe et al. 2023).Since the mid-1990s, more than 1,200 volunteers in the Great Lakes Marsh Monitoring Program (GLMMP) delivered by Birds Canada (birdscanada.org/gl_mmp) have monitored the status of this group of species and others at coastal and inland wetlands throughout the developed, southern portion of the Great Lakes basin in the USA and Canada (i.e.,in Bird Conservation Regions 13,22,and 23).The effort revealed that abundance of about onethird of 18 marsh-breeding bird species declined by 2-8% per year, 1995-2018 (total decline: 41-85%; Tozer 2020).The declines were attributed to wetland loss and fragmentation, increased urban and agricultural land use surrounding wetlands, reduced open water-vegetation interspersion, spread of non-native invasive plants, habitat degradation within Great Lakes Areas of Concern, and potential problems at migratory staging areas or on the wintering grounds (Tozer 2016).The following have been suggested as key for successful marsh bird conservation in the region: (1) fluctuating water levels, (2) high amounts of wetland cover in the surrounding landscape, and (3) low amounts of urban land cover in the surrounding area (Timmermans et al. 2008, Chin et al. 2014, Tozer 2016, 2020, Tozer et al. 2018, 2020, Studholme et al. 2023).Naturally fluctuating water levels across years, especially rising levels, were particularly important for the maintenance of inundated marsh vegetation, open water-vegetation interspersion, and associated increases in nesting and foraging substrates (Steen et al. 2006), as long as relatively stable water levels occurred during the breeding period to avoid flooded or stranded nests (Meyer et al. 2006).Indeed, abundance of some species that decreased during receding and low lake levels (mid-1990s to early-2010s) began to increase again during rising and high lake levels (early-2010s to early-2020s; Tozer 2020, Birds Canada 2022).However, the GLMMP has never effectively monitored marsh birds throughout the relatively undeveloped, northern portion of the Great Lakes basin (i.e., in Bird Conservation Regions 8 and 12) due primarily to low availability of volunteer scientists (Weeber and Vallianatos 2000, Crewe et al. 2006, Tozer 2016, Tozer et al. 2022).
To help fill the marsh bird sampling gap in the undeveloped, northern portion of the Great Lakes and to monitor the status and trends of the ecological condition of coastal wetlands throughout the entire Great Lakes basin, the Great Lakes Coastal Wetland Monitoring Program (CWMP) was implemented in 2011 by Central Michigan University and dozens of partner organizations in the USA and Canada in collaboration with the United States Environmental Protection Agency (greatlakeswetlands.org; Uzarski et al. 2017Uzarski et al. , 2019)).Each year, trained CWMP technicians survey marsh-breeding birds, as well as anurans (frogs/toads), fish, macroinvertebrates, vegetation, and water quality, at coastal wetlands throughout the Great Lakes (Uzarski et al. 2017(Uzarski et al. , 2019)).Marsh bird data from the GLMMP and CWMP have been combined, where appropriate, to answer various questions of conservation interest and to provide critical monitoring and status assessments (Tozer et al. 2017, 2022, Tozer and Mackenzie 2019, Elliott et al. 2023).CWMP data have also been used to extend findings of the GLMMP to the entire Great Lakes basin (Hohman et al. 2021) and to examine findings of the GLMMP in areas where GLMMP data are sparse (Gnass Giese et al. 2018).Thus, the GLMMP and CWMP provide flexible options for guiding marsh bird conservation at inland and coastal wetlands throughout the entire Great Lakes basin.With over a decade of data collection by the CWMP to date, we can now calculate the first-ever abundance indices and trends for marsh bird species at coastal wetlands throughout all of the Great Lakes, which will yield a more complete assessment of the status of marsh birds in the region.
Our objective was to quantify annual abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes using 11 years of data (2011-2021) collected at 1,962 point-count locations in 792 wetlands by the CWMP.We used Bayesian hierarchical models with spatial dependencies to estimate annual abundance indices and trends for each lake (Thogmartin et al. 2004, Bled et al. 2013, Ethier and Nudds 2015, Smith et al. 2015, Meehan et al. 2019, Ethier et al. 2022).The models included the following environmental predictors of marsh bird abundance: (1) local wetland cover (as a proxy for wetland size); (2) detrended, standardized Great Lakes water levels (to avoid correlation with year); (3) urban land cover in the surrounding watershed; and (4) agricultural land cover in the surrounding watershed.In doing so, we extended and improved previous, similar analyses completed in the developed, southern portion of the Great Lakes (Timmermans et al. 2008, Tozer 2016, Bianchini and Tozer 2023).Based on these previous studies and life histories of individual wetland bird species, we predicted that abundance of most species would increase over time due to concurrent increases in lake levels during the study period (e.g., American Coot [Fulica americana], Common Gallinule [Gallinula galeata], and Pied-billed Grebe [Podilymbus podiceps]), whereas abundance of only a few species would decrease due to increasing lake levels (e.g., Sandhill Crane [Antigone canadensis], Common Yellowthroat [Geothlypis trichas], Swamp Sparrow [Melospiza georgiana]; Timmermans et al. 2008, Gnass Giese et al. 2018, Kirchin et al. 2020, Tozer 2020, Hohman et al. 2021, Denomme-Brown et al. 2023).We also predicted that most species would have (1) higher abundance in certain lakes compared to others (e.g., Marsh Wren [Cistothorus palustris], Common Grackle [Quiscalus quiscula], and Red-winged Blackbird [Agelaius phoeniceus] lower in lakes Superior and Huron; Sedge Wren [Cistothorus stellaris] higher in lakes Superior and Michigan; Hanowski et al. 2007, Timmermans et al. 2008, Panci et al. 2017, Tozer 2020, Hohman et al. 2021); (2) higher abundance where local wetland cover was higher (e.g., Black Tern, Forster's Tern [Sterna forsteri] and Sora [Porzana carolina]; Tozer 2016, Saunders et al. 2019, Grand et al. 2020, Tozer et al. 2020, Malone et al. 2021); (3) lower abundance where urban land cover in the surrounding watershed was higher (e.g., American Bittern [Botaurus lentiginosus] and Virginia Rail [Rallus limicola]; Panci et al. 2017, Wyman and Cuthbert 2017, Jung et al. 2020, Rahlin et al. 2022, Studholme et al. 2023); and (4) variable responses where agricultural land cover in the surrounding watershed was higher (e.g., Wilson's Snipe [Gallinago delicata]; Tozer 2016, Saunders et al. 2019, Tozer et al. 2020).An exception was Mute Swan (Cygnus olor), which we predicted would increase with increasing urban land cover in the surrounding area (Gehring et al. 2020).

Study Area and Design
We conducted our study in coastal wetlands throughout the entire Great Lakes basin (Figure 1).We selected coastal wetlands using a stratified, random sampling protocol (Uzarski et al. 2017(Uzarski et al. , 2019)).Further details regarding the study design are in Burton et al. (2008).The sampling universe was all coastal wetlands greater than 4 ha in size with a permanent or periodic surface-water connection to an adjacent Great Lake or their connecting river systems (Uzarski et al. 2017).We stratified our selection of wetlands for the study by (1) wetland hydrogeomorphic type (riverine, lacustrine, barrier protected; Albert et al. 2005), (2) region (northern or southern; Danz et al. 2005), and (3) lake (i.e., the watershed of 1 of the 5 Great Lakes).We sampled ~20% of all wetlands in each stratum each year, so that nearly all coastal wetlands within the Great Lakes basin meeting the selection criteria were sampled at least once every 5 years.In addition, we resampled 10% of wetlands between years according to a rotating panel design.Sampled wetlands were dominated by emergent, herbaceous vegetation and shallow water (<2 m deep) containing floating and/or submerged vegetation.We surveyed a mean of 369 point-count locations (range: 309-423) for marsh birds in each year throughout the Great Lakes basin, and the mean annual number of point-count locations in each lake was 65 points for Erie (range: 31-98 points), 99 for Huron (76-116), 56 for Michigan (36-71), 95 for Ontario (77-117), and 55 for Superior (35-78; Figure 2).

Bird Surveys
We conducted surveys at 1-8 fixed point-count locations at the edge of, or within, each wetland in each year that a wetland was selected for surveys.Point-count locations were >250 m apart to avoid double-counting individuals.We surveyed each point-count location twice per year, at least 15 days apart, between May 20 and July 10, which was the peak breeding period for marsh birds in the study area.Surveys took place either in the morning (30 min before sunrise to 4 hr after sunrise) or the evening (4 hr before sunset to 30 min after sunset), with 1 or both of the 2 surveys being in the morning each year (Tozer et al. 2017).We conducted surveys only when there was no precipitation and wind was <20 km hr -1 (Beaufort 3 or less).Each point-count survey lasted 10 min, consisting of an initial 5-min passive listening period followed by a 5-min call broadcast period.The call broadcast period was intended to increase detections of secretive species by eliciting auditory responses and was composed of 30 s of vocalizations followed by 30 s of silence for each of the following: (1) Least Bittern, (2) Sora, (3) Virginia Rail, (4) a mixture of American Coot and Common Gallinule, and (5) Pied-billed Grebe, in that order.We trained observers so they thoroughly understood the field protocols, and we required each observer to pass an aural and visual bird identification test in order to collect data.CWMP bird surveys were 15 min in duration from 2011 to 2018 but were reduced to 10 min from 2019 to 2021 (Tozer et al. 2017).To accommodate changes in survey protocol, we filtered the data to only include birds detected in the first 10 min of point counts from 2011 to 2018.For a detailed description of the sampling protocol visit greatlakeswetlands.org/Sampling-protocols.

Response Variable
The response variable for each species was the maximum number of individuals observed during either of the 2 surveys at each point-count location in each year (Tozer 2020, Hohman et al. 2021).We viewed these counts as indices of true density, meaning our modeled values estimated relative abundance (Thogmartin et al. 2004).We assumed that variation in species-specific detection was uncorrelated with the predictors in our models, including year.This was sufficient in our case because our objective was to quantify relative differences and changes in abundance and not to quantify actual density.Our assumption was warranted because our data were collected using standardized methods designed to reduce heterogeneity in detection (e.g., observer training and testing) as well as restrictions on survey date, time of day, and wind (Conway 2011, Uzarski et al. 2017).It was further justified by other long-term, broad-scale studies of birds based on point counts conducted using similar standardized approaches, which found no differences in year or covariate effects based on counts that were adjusted or unadjusted for detection (Etterson et al. 2009, Zlonis et al. 2019).We note that long-term (1996-2013) marsh-breeding bird monitoring data collected throughout the developed, southern portion of the Great Lakes basin showed no systematic trends in detectability over time for 14 of 15 (93%) species (Tozer 2016).We also found no trends in detectability across years for all of the species in our dataset (Supplementary Material Figure 1), meaning that differences in detection did not bias our estimates of annual abundance indices or trends.Therefore, we did not adjust for detectability, which has been supported, for instance, by Hutto (2016) and Johnson (2008).
The dataset consisted of 8,120 surveys completed at 1,962 point-count locations in 792 coastal wetlands in 599 watersheds (defined by Forsyth et al. 2016Forsyth et al. ) over 11 years (2011Forsyth et al. -2021; Figures 1 and 2; Supplementary Material Table 1).There were 2.2 ± 1.6 (mean ± SD) point-count locations per wetland (range: 1-8) and 1.3 ± 0.9 wetlands per watershed (range: 1-9).In total, we analyzed 18 species: (1) American Bittern, (2) American Coot, (3) Black Tern, (4) Common Gallinule, (5) Common Grackle, (6) Common Yellowthroat, (7) Forster's Tern, (8) Least Bittern, (9) Marsh Wren, (10) Mute Swan, (11) Pied-billed Grebe, (12) Red-winged Blackbird, (13) Sandhill Crane, ( 14) Sedge Wren, (15) Sora, ( 16) Swamp Sparrow, (17) Virginia Rail, and (18) Wilson's Snipe.We chose these species because they were of conservation interest in the Great Lakes region (Bianchini and Tozer 2023) and regularly nested or foraged in Great Lakes coastal wetlands.We attempted to model abundance and trends for Trumpeter Swan (Cygnus buccinator) and Yellow-headed Blackbird (Xanthocephalus xanthocephalus), but data were too sparse for the models to converge.We considered some regions of our study area to be out of range for some species.We accounted for this by dividing our study area into 10 regions and dropped any of them from species-specific analyses if naive occupancy was <5% (Supplementary Material Table 2).By excluding out-ofrange point-count locations, we reduced the number of zero counts and focused our analysis on point-count locations where zero counts were most likely to represent legitimate absences.As a result, the number of marsh-breeding bird species for which we quantified abundance and trends varied by lake due to uneven species occurrences across the study area: Superior (n = 10), Ontario (n = 12), Erie (n = 16), Huron (n = 16), and Michigan (n = 17).

Environmental Predictors
We included the following environmental predictors in our models, which were known to influence the abundance of marsh-breeding birds in the Great Lakes: ( 1 (3) percent urban land cover in the surrounding watershed (Rahlin et al. 2022); and (4) percent agricultural land cover in the surrounding watershed (Saunders et al. 2019).The land cover predictors were static covariates (i.e., they were the same for all years), whereas detrended, standardized Great Lakes water level was a dynamic covariate (i.e., it varied annually).Land cover and water-level information at finer spatial and temporal scales would have been preferred, but such data were unavailable.Nonetheless, it is reasonable to assume that the land cover and water-level data we used provided useful approximations of the true values, particularly at the watershed scale (Michaud et al. 2022).The percent local wetland cover was based on the coastal wetland layer built by the Great Lakes Coastal Wetland Consortium (Burton et al. 2008, Uzarski et al. 2017), and the percent urban and agricultural land cover were from Host et al. (2019) with watersheds defined by Forsyth et al. (2016); coastal wetland and watershed data are available at glahf.org/data.We used ArcGIS 10.8.1 to overlay CWMP sample points onto the land cover layers and extracted the relevant predictors for each point (Figure 3).Yearly water levels were from the Great Lakes Environmental Research Laboratory of the National Oceanic and Atmospheric Administration available at glerl.noaa.gov/data/wlevels.We used the mean yearly water level from May to July since these months overlapped with our survey period.We detrended water levels from year by using the residuals from a line of best fit for each lake, given that water levels generally increased in all lakes over the course of the study.Water levels were also standardized across  lakes by subtracting the long-term mean (2011-2021) for each lake from the annual value for each lake and dividing by the standard deviation, given the reference value is the same for all lakes (International Great Lakes Datum 1985).Our detrended, standardized water levels therefore represent water levels without being confounded with year (Figure 4).The environmental predictors were not correlated (-0.2 < r < 0.05; Supplementary Material Figure 2).

Statistical Modeling
We fit models in a Bayesian framework with Integrated Nested Laplace Approximation (INLA) using the R-INLA package (Rue et al. 2009) for R statistical computing (version 4.2.0;R Core Team 2022).For each species, we modeled the expected (predicted mean) number of individuals per point-count location in each Great Lake in each year, as well as the trend in these values across years in each lake, and then pooled the lake-specific trends to obtain Great Lakes-wide estimates.We included spatial structure in the models using an intrinsic conditional autoregressive (iCAR) structure (Besag et al. 1991), which allowed for informa-tion on relative abundance to be shared across lakes sharing basin boundaries.By accounting for this spatial structure in counts, the model allowed abundance and trend information to be shared among adjacent lakes (as described later), which improved estimates for lakes with limited sample sizes (Bled et al. 2013) and reduced the amount of spatial autocorrelation in model residuals (Zuur et al. 2017).
We modeled counts у i,j,t using the maximum number of individuals observed at a point-count location within a given wetland j, lake i, and year t.The expected counts per lake within a given year µ i,t for each of the 18 species took the form: where α is the random lake intercept; T is the year, indexed to 2021; τ is the random lake slope effect; κ is the random wetland effect; ρ is the random wetland type effect (i.e., riverine, lacustrine, barrier protected; Albert et al. 2005); and у is the random lake-year effect.Environmental predictors included: W is the percent local wetland cover within 250 m; L is the detrended, standardized water level; U is the percent urban land cover in the surrounding water-  shed; and A is the percent agricultural land cover in the surrounding watershed.
The random lake intercept (α i ) had an iCAR structure, where values of α i came from a normal distribution with a mean value related to the average of adjacent lakes.The random lake intercept also had a conditional variance proportional to the variance across adjacent lakes and inversely proportional to the number of adjacent lakes.We modeled the random lake slopes (τ i ) as spatially structured, lakespecific, random slope coefficients for the year effect, using the iCAR structure, with conditional means and variances as described earlier.We incorporated spatial structure into the random lake slopes (τ i ) to allow information about year effects to be shared across neighboring lakes, and to allow year effects to vary among lakes.We transformed year (T) such that the maximum year was 0, and each preceding year was a negative integer.This scaling meant that the estimates of the random lake intercepts (α i ) could be interpreted as the lake-specific expected counts (i.e., index of abundance) during the final year of the time series.We accounted for differences in relative abundance among wetlands (κ) and wetland types (ρ) with an independent and identically distributed (i.d.d.) random effect.To derive an annual index of abundance per lake, we included a random effect per lakeyear (у) with an i.d.d., and combined these effects with α and τ.Β 1 , β 2 , β 3 , and β 4 were given normal priors with mean of zero and precision equal to 0.001.We scaled the spatial structure parameters α and τ such that the geometric mean of marginal variances was equal to one (Sørbye and Rue 2014, Riebler et al. 2016, Freni-Sterrantino et al. 2018), and priors for precision parameters were penalized complexity (PC) priors, with parameter values UPC = 1 and PC = 0.01 (Simpson et al. 2017).We also assigned precision for the random wetland, wetland type, and lake-year effects with a PC prior with parameter values previously stated.In general, the weakly informed priors used here tend to shrink the structured and unstructured random effects towards zero in the absence of a strong signal (Simpson et al. 2017).
We validated distributional assumptions with simulation to ensure models could handle the large number of zero counts for some species.The abundance of most species was modeled using a zero-inflated Poisson (ZIP) distribution.Common Grackle and Red-winged Blackbird, which were more frequently detected compared to the other species, better fit a negative binomial distribution, and Common Yellowthroat better fit a Poisson distribution.We further validated models by visually inspecting (1) the fit versus raw counts; (2) residuals versus predictors; and (3) the estimate for Ф, the dispersion parameter (Zuur and Ieno 2016).Our visual inspections of fit versus raw counts suggested models were not overfitted and were able to capture the variation of the raw counts.In general, residuals versus fit values behaved randomly around the zero line, and residuals appeared to behave randomly with each predictor, suggesting the models fit well.The dispersion statistics were around 1 for all species, ranging lowest for Common Yellowthroat (0.72) and highest for Mute Swan (3.38), suggesting some residual under and overdispersion, respectively.Mute Swan had some high counts (outliers) which may have contributed to this.Following model analysis, we computed posterior estimates of trends (τ) and associated credible intervals for the full extent of the study area (i.e., by pooling lake-specific trends) using lake watershed size to calculate area-weighted averages (Link and Sauer 2002).

RESULTS
Nine species (50%) increased by 8-37% per year across all of the Great Lakes combined, whereas none decreased (Table 1).Twelve species (67%) increased by 5-50% per year in at least 1 of the 5 Great Lakes, whereas only 3 species Trends for each species are pooled across the 5 Great Lakes (Figure 6).Species are listed in decreasing order by trend with lower and upper 95% credible limits (CL).Species with CLs that do not overlap zero are shown in bold.
(17%) decreased by 2-10% per year in at least 1 of the lakes (Common Yellowthroat, Sandhill Crane, Swamp Sparrow; Figure 5).There were more positive trends across lakes and species (n = 34, 48%) than negative trends (n = 5, 7%), whereas credible intervals overlapped zero in the remainder (n = 32, 45%; Figure 5).Trends among lakes for most species involved a mix of increasing, decreasing, and/or stable trends (i.e., credible intervals overlapped zero), although trends were positive in all lakes for Common Gallinule, Pied-billed Grebe, Sora, and Virginia Rail.Sandhill Crane was the only species that significantly increased in one lake (Erie) and decreased in another (Michigan; Figure 5).Relationships with environmental predictor variables varied among species.Abundance of nearly half of the species was influenced by detrended, standardized water levels (8 of 18 species, 44%), with 5 species increasing as detrended, standardized water levels increased (American Bittern, American Coot, Common Gallinule, Sora, and Virginia Rail) FIGURE 5. Trends in abundance indices (percent change per year) mostly increased for 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.Horizontal lines are 95% credible intervals.Note that some lakes were omitted for some species due to lack of data.Lake names are provided in the left vertical axis and in the legend for ease of interpretation.and 3 species decreasing (Common Yellowthroat, Sandhill Crane, and Swamp Sparrow; Figure 6).Abundance of most species increased as local wetland cover increased (12 of 18 species, 67%), whereas no species decreased in relation to this predictor (Figure 6).Abundance of about one-third of the species was associated with the amount of urban land cover in the surrounding watershed (6 of 18 species, 33%), with 4 species decreasing as urban land cover increased (American Bittern, Common Yellowthroat, Sandhill Crane, and Swamp Sparrow) and 2 species increasing (Common Grackle and Red-winged Blackbird; Figure 6).Abundance of only a few species was associated with the amount of agricultural land use in the surrounding watershed (3 of 18 species, 17%), with 2 species significantly increasing as agricultural land cover increased (Common Grackle and Least Bittern) and Common Yellowthroat decreasing (Figure 6).
Credible intervals overlapped extensively among lakes for most species in most years, although some differences were worth noting, given that they appeared consistently across several or all of the years.Swamp Sparrow was more abundant in Lake Ontario than the other lakes; and Least Bittern, Marsh Wren, and Mute Swan tended to be more abundant in Lake Ontario, and in, at least, some years, Pied-billed Grebe tended to be less abundant in Lake Ontario, compared to the other lakes (Figure 7).American Bittern and Common Grackle were less abundant in Lake Superior in the earlier part of the study period than in the other lakes, but less so thereafter (Figure 7).Abundance of American Coot and Sora in all years, and abundance of Sandhill Crane in the earlier part of the study period, tended to be higher in Lake Michigan compared to the other lakes (Figure 7).Abundance of Marsh Wren and Red-winged Blackbird tended to be lower in Lake Huron and Lake Superior compared to the other lakes (Figure 7).Abundance was notably more similar among lakes across all years for Common Gallinule and Virginia Rail compared to the other species (Figure 7).

DISCUSSION
We found 9 (50%) species increased by 8-37% per year across all of the Great Lakes combined, whereas none decreased.We also found more positive trends among lakes and species (n = 34, 48%) than negative trends (n = 5, 7%).The FIGURE 6. Relationships with environmental predictors varied among 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.Horizontal lines are 95% credible intervals.Wetland cover was within 250 m; water level was detrended, standardized water level; and urban and agricultural land cover were within the surrounding watershed, defined by Forsyth et al. (2016).
overwhelmingly positive trends we found are in stark contrast to the mostly negative trends reported for marsh-breeding birds in parts of the Great Lakes during earlier time periods (Crewe et al. 2006, Timmermans et al. 2008, Tozer 2013, 2016).These earlier reports, however, were from periods with mostly declining and low Great Lakes water levels.As we elaborate further later, the highly positive trends we found were likely caused by increasing lake levels and associated increases in marsh bird habitat quality during most of our study.Our findings potentially have large implications for conservation because if Great Lakes coastal wetlands provide high-lake-level-induced population pulses for most marsh bird species, then coastal wetlands may be more important than once thought for recovering marsh-breeding bird species of conservation concern in the Great Lakes.
Water levels of the Great Lakes are driven by 3 key factors: (1) over-lake precipitation, (2) lake evaporation, and (3) basin run-off (Kayastha et al. 2022), all of which are influenced by various global climate oscillations and other factors (Saber et al. 2023).Variation in these factors resulted in declining and then low lake levels from the mid-1990s to early-2010s, followed by rising and then very high lake levels from the early-2010s to early-2020s, which is when our study took place (Gronewold et al. 2021, Lam and Dokoska 2022, Mayne et al. 2022).Trends for most marsh-breeding birds in the Great Lakes were negative during the low-water period.For example, occupancy of 9 of 15 (60%) marsh-breeding bird species decreased in the southern portion of the Great Lakes basin 1996-2013, whereas only 1 (7%) increased (Tozer 2016).By contrast, trends tended to be positive during the high-water period, as we found in this study.For example, abundance of 9 species that had been declining during the low-water period, then reversed and increased during the high-water period (Crewe et al. 2006, Timmermans et al. 2008, Tozer 2013, 2016, 2020, Kirchin et al. 2020, Birds Canada 2022, Bianchini and Tozer 2023).This makes sense, because abundance of these species, which included (1) American Bittern, (2) American Coot, (3) Black Tern, (4) Common Gallinule, (5) Least Bittern, (6) Marsh Wren, (7) Pied-billed Grebe, (8) Sora, and (9) Virginia Rail, has been shown to decline with receding lake levels and increase with rising lake levels (Timmermans et al. 2008, Gnass Giese et al. 2018, Kirchin et al. 2020, Hohman et al. 2021, Denomme-Brown et al. 2023).On the other hand, abundance of 3 species that had been increasing during the low-water period, then reversed and decreased during the high-water period (Crewe et al. 2006, Timmermans et al. 2008, Tozer 2013, 2016, 2020, Kirchin et al. 2020, Bianchini and Tozer 2023).This also makes sense, because abundance of these species, which included Common Yellowthroat, Sandhill Crane, and Swamp Sparrow, has been shown to increase with receding lake levels and decrease with rising lake levels (Gnass Giese et al. 2018, Hohman et al. 2021).Thus, the marsh-breeding bird trends we observed in this study during the high-water period were likely caused by lake-level-induced changes in marsh-breeding bird habitat quality, which, depending on the species, meant either a gain in habitat quality, which was the case for the majority of the species we analyzed (e.g., American Bittern), or a loss in habitat quality (e.g., Common Yellowthroat).
Increasing water levels likely increased habitat quality for most of the marsh-breeding bird species we analyzed by inundating marsh vegetation, given most of these species prefer water depths of >20-30 cm (Desgranges et al. 2006, Lor and Malecki 2006, Tozer et al. 2010).Habitat quality was likely also improved by increasing water levels due to associated increases in open water-vegetation interspersion (i.e., the amount of emergent vegetation edge) and extent of open water (Rehm and Baldassarre 2007).This is partly because availability of prey, such as invertebrates and fish, is influenced by emergent vegetation edges (Voigts 1976) and the availability of submerged vegetation for herbivorous species is influenced by the extent of open water (Brackney and Bookhout 1982).The abundance of at least 7 of the 12 (58%) species that increased in our study increased by several hundred percent in response to increasing interspersion and open water extent (Steen et al. 2006, Rehm and Baldassarre 2007, Bolenbaugh et al. 2011, Wyman and Cuthbert 2017, Dinehart et al. 2022), and interspersion increased by 20% and open water extent by 43% in coastal wetlands throughout the Great Lakes between a low-water year (2013) and a high-water year (2018; Hohman et al. 2021).Therefore, the predominantly positive marsh-breeding bird trends we observed in this study were likely caused by increasing water levels, which increased habitat quality for most species, not only by inundating marsh vegetation, but also by increasing the interspersion of, and open water extent within, marsh vegetation.
The overwhelmingly positive trends we found potentially have large implications for conservation because most of the species that increased in this study were designated as marsh bird priority species of conservation concern in the Great Lakes (Collins andSmith 2014, Soulliere et al. 2018).Monitoring by the GLMMP showed increasing trends for some of these species in coastal wetlands from the early-2010s to early-2020s, just like we observed in this study, but no corresponding increases or much smaller increases in inland wetlands (Tozer 2020).This suggests that high-lake-level-induced increases in abundance in Great Lakes coastal wetlands may be important for boosting regional populations of marsh bird priority species of conservation concern.Coastal wetland protection and restoration are of utmost importance to safeguard this process.More information on how observed increases in abundance of these species in coastal wetlands may translate into regional population increases would be useful for conservation planning.For instance, if abundance remains high enough for long enough in coastal wetlands do individuals eventually "spill over" to inland wetlands causing an even larger boost to the total population in the Great Lakes region?Continued, standardized monitoring at coastal and inland wetlands by the CWMP and GLMMP will be useful for gauging these patterns at broad scales.
Any value that Great Lakes coastal wetlands provide as critical reservoirs of high-quality marsh-breeding bird habitat will be challenged in the future due to climate change.Future climate projections for the Great Lakes show large increases in lake levels over the next several decades, including more frequent extremes (i.e., higher highs and lower lows; Kayastha et al. 2022, Lam and Dokoska 2022, Mayne et al. 2022).This may cause "coastal squeeze" of wetlands (Pontee 2013) if wetlands are unable to migrate landward fast enough to keep pace with increasing lake levels (Smith et al. 2021, Hartsock et al. 2022, Rutherford et al. 2022, Tozer et al. 2022, Anderson et al. 2023).As a result, there will be less coastal marsh bird habitat available to support periodic pulses in marsh bird abundance, which could have negative consequences for regional population dynamics.Actions that remove or prevent barriers to landward movement of wetlands, such as shoreline armoring, riprap, retaining walls, road infrastructure, or other obstructions, will help mitigate the negative influence of future high-water extremes on coastal marsh bird habitat due to climate change (Mayne et al. 2022); for instance, by creating upslope "climate refugia" in key areas where coastal wetlands would be free to migrate as water levels rise (Mayne et al. 2022).Likewise, actions that facilitate lakeward movement and persistence of coastal wetlands during low lake level extremes will also help marsh-breeding birds resist the effects of climate change (Snell et al. 2006).Some of the differences in abundance that we found among lakes for certain species have been found or discussed previously by others.We found that Swamp Sparrow was clearly more abundant, and Least Bittern, Marsh Wren, and Mute Swan tended to be more abundant, in Lake Ontario compared to the other lakes (Figure 5).Mute Swan might be more abundant in Lake Ontario due to lower frequency of control efforts and the presence of several large, attractive wetlands for breeding (Meyer et al. 2012, Gehring et al. 2020).Least Bittern, Marsh Wren, and Swamp Sparrow might be more abundant in Lake Ontario due to the widespread presence of large, unbroken stands of the invasive, hybrid cattail (Typha × glauca).This is preferred breeding habitat for these species and has spread throughout the lake's coastal wetlands because of increased nutrient inputs and water-level regulation (Wilcox et al. 2008, Jobin et al. 2011, Harms and Dinsmore 2015, Panci et al. 2017, Bansal et al. 2019, Heminway and Wilcox 2022).We also found that Pied-billed Grebe tended to be less abundant in some years in Lake Ontario (Figure 5), probably due to reductions in open water-vegetation interspersion in the lake's wetlands due to cattail invasion (Rehm and Baldassarre 2007, Carson et al. 2018, Hohman et al. 2021).We found that American Coot and Sora in all years, and Sandhill Crane in the earlier part of the study period, tended to be more abundant in Lake Michigan compared to the other lakes (Figure 5), probably, at least in part, because the highest North American breeding densities of these species occur in the Plains and Prairie regions to the west of the lake (Tacha and Braun 1994, Tozer et al. 2016, Fink et al. 2022).Lastly, we found that Marsh Wren and Red-winged Blackbird tended to be less abundant in Lake Huron and Lake Superior compared to the other lakes (Figure 5).This is likely because wetlands in the northern portion of the Great Lakes are less productive, have fewer extensive cattail stands, and offer less insect food for these species due to fewer nutrients as a function of hard, underlying, granitic bedrock and much lower anthropogenic inputs (Environment Canada 2002, Hanowski et al. 2007, Wilcox 2012).
The abundance of 12 of 18 (67%) species in this study increased with increasing local wetland cover within 250 m, a proxy for wetland size.This was expected and has been shown previously by many others for marsh-breeding birds in the Great Lakes (Grand et al. 2020).Each species depends to varying degrees on different types and structural components of marsh vegetation for nesting and feeding during the breeding season, and many are considered area-sensitive with respect to wetland size (Riffell et al. 2001, Su 2003, Bolenbaugh et al. 2011, Glisson et al. 2015, Harms and Dinsmore 2015, Tozer 2016, Panci et al. 2017, Saunders et al. 2019, Gehring et al. 2020, Grand et al. 2020, Tozer et al. 2020, Malone et al. 2021, Rahlin et al. 2022, Studholme et al. 2023).Also in this study, the abundance of 6 of 18 (33%) species increased (2 species) or decreased (4 species) with increasing urban land cover in the surrounding watershed, and the abundance of 3 of 18 (17%) species increased (2 species) or decreased (1 species) with increasing agricultural land cover in the surrounding watershed.Results similar to these have also been reported for marsh-breeding birds in the Great Lakes (Smith and Chow-Fraser 2010, Panci et al. 2017, Saunders et al. 2019, Jung et al. 2020, Tozer et al. 2020, Malone et al. 2021, Rahlin et al. 2022, Studholme et al. 2023).The negative influence of surrounding urban land use on abundance of marsh birds has been especially noted for some species (Panci et al. 2017, Jung et al. 2020, Malone et al. 2021) and may apply to reproductive success as well.For instance, nest failure of Redwinged Blackbirds increased with increasing urban and residential land cover within 1 km of coastal wetlands in Lake Superior (Grandmaison et al. 2007), although the abundance of this species increased with increasing urban land cover in the surrounding watershed in this study.The results from this study highlight recommendations made previously by several others: multiple marsh-breeding bird species will benefit from the protection and restoration of large Great Lakes coastal wetlands or wetland complexes, which contain diverse communities of native marsh vegetation interspersed with numerous shallow, open-water pools, and which are surrounded by low amounts of urban land use in the surrounding watershed or landscape (Smith and Chow-Fraser 2010, Tozer et al. 2010, 2020, Tozer 2016, 2020, Panci et al. 2017, Grand et al. 2020, Malone et al. 2021, Studholme et al. 2023).
In this study, we quantified annual abundance indices and trends of marsh-breeding birds in coastal wetlands throughout all of the Great Lakes using CWMP data.We did not combine or integrate these data with those from the GLMMP or other long-term, broad-scale, bird monitoring programs, although the methods to do so are becoming more developed (Zipkin et al. 2019).Many of the species we studied occur at low frequencies in Great Lakes coastal wetlands (Supplementary Material Table 1), which in turn yield numerous zero counts, small sample sizes, and low associated statistical power for quantifying abundance indices and trends (Steidl et al. 2013).Low power is evident in some of our results, which have relatively wide credible intervals for some species (Figures 5 and  7).Power and precision could be improved by combining or integrating CWMP marsh bird data with those from other monitoring programs to increase sample sizes (Hertzog et al. 2021).For instance, Bianchini and Tozer (2023) combined GLMMP, Breeding Bird Survey, and eBird data to increase sample sizes and associated precision of abundance indices and trends of marsh-dependent breeding birds at inland and Great Lakes coastal marshes throughout southern Ontario.A major challenge with data integration is to control for biases inherent to each dataset, such as differences in field protocols and experimental designs, although data filtering, covariate, and hierarchical joint likelihood modeling procedures are increasingly being developed and improved to successfully achieve this (Massimino et al. 2008, Zipkin et al. 2021).Integrating CWMP data with GLMMP and eBird data especially, and potentially other monitoring datasets, is a promising avenue for future research that will likely improve the precision, without biasing the accuracy, of abundance indices and trends for marsh-breeding birds.

Conclusion and Conservation Implications
We found overwhelmingly positive trends in abundance for 18 marsh-breeding bird species in coastal wetlands across all of the Great Lakes combined, 2011-2021.Increasing trends were likely caused by rising Great Lakes water levels and associated increases in marsh bird habitat quality over most of the study.Therefore, Great Lakes coastal wetlands may provide critical, periodic boosts to regional populations of priority species of conservation concern.Coastal wetland protection and restoration are of utmost importance to safeguard this process.Any value that Great Lakes coastal wetlands provide for marsh-breeding birds will be challenged in the future due to climate change because "coastal squeeze" may cause reductions in habitat.Actions that allow landward migration of coastal wetlands during increasing water levels by removing or preventing barriers to movement, such as shoreline hardening, will be useful for maintaining marsh bird breeding habitat in the Great Lakes.

FIGURE 1 .
FIGURE 1. Point-count locations (n = 1,962) used to quantify abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.
) percent local wetland cover within 250 m of point-count locations (as a proxy for wetland size; Studholme et al. 2023); (2) detrended, standardized Great Lakes water levels (to avoid correlation with year; Hohman et al. 2021, Denomme-Brown et al. 2023);

FIGURE 2 .
FIGURE 2. Number of point-count locations used to quantify abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands of each of the 5 Great Lakes (leftmost panel) and throughout the entire Great Lakes (rightmost panel), 2011-2021.See Supplementary MaterialTable 1 for exact values.

FIGURE 3 .
FIGURE 3. Distribution of land cover predictors (percent cover within 250 m or the surrounding watershed) used to quantify abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.Watersheds were defined by Forsyth et al. (2016).

FIGURE 4 .
FIGURE 4. "Raw" water levels increased throughout the study in each of the Great Lakes (leftmost panels; m IGLD85 = meters International Great Lakes Datum 1985), which were used to calculate detrended, standardized water level predictors (rightmost panel) for quantifying abundance indices and trends of 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.Note that water levels were identical for Lake Huron and Lake Michigan (Hur-Mich) because they are connected.See Methods-Environmental predictors for details.

FIGURE 7 .
FIGURE 7. Abundance indices (individuals per point-count location) mostly increased for 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.Curved lines are LOESS smoothers, and vertical lines are 95% credible intervals.Note that, for each species, the upper portion of the vertical axis is shrunk, starting just above the highest annual index (indicated by horizontal black lines), to facilitate interpretation of differences among lakes while also showing the full upper extent of credible intervals.
Table 1 for exact values.

Table 1 .
Trends in abundance (percent change per year) increased for one-half of 18 marsh-breeding bird species in coastal wetlands throughout the Great Lakes, 2011-2021.