Southwest Alaska Inventory and Monitoring Network

A framework for a sampling plan for monitoring marshbird populations in the contiguous 48 states is proposed here. The sampling universe is the breeding habitat (i.e. wetlands) potentially used by marshbirds. Selection protocols would be implemented within each of large geographical strata, such as Bird Conservation Regions. Site selection will be done using a two-stage cluster sample. Primary sampling units (PSUs) would be land areas, such as legal townships, and would be selected by a procedure such as systematic sampling. Secondary sampling units (SSUs) will be wetlands or portions of wetlands in the PSUs. SSUs will be selected by a randomized spatially balanced procedure. For analysis, the use of a variety of methods as a means of increasing confidence in conclusions that may be reached is encouraged. Additional effort will be required to work out details and implement the plan.

Abstract Dissolved oxygen (DO) is essential to the survival of almost all aquatic organisms. Here, we examine the possibility that abundant Pacific salmon ( Oncorhynchus spp.) and low streamflow combine to create hypoxic events in coastal rivers. Using high‐frequency DO time series from two similar watersheds in southeastern Alaska, we summarize DO regimes and the frequency of hypoxia in relationship to salmon density and stream discharge. We also employ a simulation model that links salmon oxygen respiration to DO dynamics and predicts combinations of salmon abundance, discharge, and water temperature that may result in hypoxia. In the Indian River, where DO was monitored hourly during the ice‐free season from 2010 to 2015, DO levels decreased when salmon were present. In 2013, a year with extremely high spawning salmon densities, DO dropped to 1.7 mg/L and 16% saturation, well below lethal limits. In Sawmill Creek, where DO was monitored every six minutes across an upstream–downstream gradient during the 2015 spawning season, DO remained fully saturated upstream of spawning reaches, but declined markedly downstream to 2.9 mg/L and 26% saturation during spawning. Modeled DO dynamics in the Indian River closely tracked field observations. Model sensitivity analysis illustrates that low summertime river discharge is a precursor to salmon‐induced oxygen depletion in our study systems. Our results provide compelling evidence that dense salmon populations and low discharge can trigger hypoxia, even in rivers with relatively cold thermal regimes. Although climate change modeling for southeastern Alaska predicts an increase in annual precipitation, snowfall in the winter and rainfall in the summer are likely to decrease, which would in turn decrease summertime discharge in rain‐ and snow‐fed streams and potentially increase the frequency of hypoxia. Our model template can be adapted by resource managers and watershed stakeholders to create real‐time predictive models of DO trends for individual streams. While preserving thermally suitable stream habitat for cold‐water taxa facing climate change has become a land management priority, managers should also consider that some protected watersheds may still be at risk of increasingly frequent hypoxia due to human impacts such as water diversion and artificially abundant salmon populations caused by hatchery straying.

Abstract Species distribution and abundance are critical population characteristics for efficient management, conservation, and ecological insight. Point process models are a powerful tool for modelling distribution and abundance, and can incorporate many data types, including count data, presence‐absence data, and presence‐only data. Aerial photographic images are a natural tool for collecting data to fit point process models, but aerial images do not always capture all animals that are present at a site. Methods for estimating detection probability for aerial surveys usually include collecting auxiliary data to estimate the proportion of time animals are available to be detected. We developed an approach for fitting point process models using an N ‐mixture model framework to estimate detection probability for aerial occupancy and abundance surveys. Our method uses multiple aerial images taken of animals at the same spatial location to provide temporal replication of sample sites. The intersection of the images provide multiple counts of individuals at different times. We examined this approach using both simulated and real data of sea otters ( Enhydra lutris kenyoni ) in Glacier Bay National Park, southeastern Alaska. Using our proposed methods, we estimated detection probability of sea otters to be 0.76, the same as visual aerial surveys that have been used in the past. Further, simulations demonstrated that our approach is a promising tool for estimating occupancy, abundance, and detection probability from aerial photographic surveys. Our methods can be readily extended to data collected using unmanned aerial vehicles, as technology and regulations permit. The generality of our methods for other aerial surveys depends on how well surveys can be designed to meet the assumptions of N ‐mixture models.

Global change has converted many structurally complex and ecologically and economically valuable coastlines to bare substrate. In the structural habitats that remain, climate-tolerant and opportunistic species are increasing in response to environmental extremes and variability. The shifting of dominant foundation species identity with climate change poses a unique conservation challenge because species vary in their responses to environmental stressors and to management. Here, we combine 35 y of watershed modeling and biogeochemical water quality data with species comprehensive aerial surveys to describe causes and consequences of turnover in seagrass foundation species across 26,000 ha of habitat in the Chesapeake Bay. Repeated marine heatwaves have caused 54% retraction of the formerly dominant eelgrass ( Zostera marina ) since 1991, allowing 171% expansion of the temperature-tolerant widgeongrass ( Ruppia maritima ) that has likewise benefited from large-scale nutrient reductions. However, this phase shift in dominant seagrass identity now presents two significant shifts for management: Widgeongrass meadows are not only responsible for rapid, extensive recoveries but also for the largest crashes over the last four decades; and, while adapted to high temperatures, are much more susceptible than eelgrass to nutrient pulses driven by springtime runoff. Thus, by selecting for rapid post-disturbance recolonization but low resistance to punctuated freshwater flow disturbance, climate change could threaten the Chesapeake Bay seagrass’ ability to provide consistent fishery habitat and sustain functioning over time. We demonstrate that understanding the dynamics of the next generation of foundation species is a critical management priority, because shifts from relatively stable habitat to high interannual variability can have far-reaching consequences across marine and terrestrial ecosystems.

Abstract Aim Sea otters ( Enhydra lutris ) are an apex predator of the nearshore marine community and nearly went extinct at the turn of the 20th century. Reintroductions and legal protection allowed sea otters to re‐colonize much of their former range. Our objective was to chronicle the colonization of this apex predator in Glacier Bay, Alaska, to help understand the mechanisms that governed their successful colonization. Location Glacier Bay is a tidewater glacier fjord in southeastern Alaska that was entirely covered by glaciers in the mid‐18th century. Since then, it has endured the fastest tidewater glacier retreat in recorded history. Methods We collected and analysed several data sets, spanning 20 years, to document the spatio‐temporal dynamics of an apex predator expanding into an area where they were formerly absent. We used novel quantitative tools to model the occupancy, abundance and colonization dynamics of sea otters, while accounting for uncertainty in the data collection process, the ecological process and model parameters. Results Twenty years after sea otters were first observed colonizing Glacier Bay, they became one of the most abundant and widely distributed marine mammal. The population grew exponentially at a rate of 20% per year. They colonized Glacier Bay at a maximum rate of 6 km per year, with faster colonization rates occurring early in the colonization process. During colonization, sea otters selected shallow areas, close to shore, with a steep bottom slope, and a relatively simple shoreline complexity index. Main conclusions The growth and expansion of sea otters in Glacier Bay demonstrate how legal protection and translocation of apex predators can facilitate their successful establishment into a community in which they were formerly absent. The success of sea otters was, in part, a consequence of habitat that was left largely unperturbed by humans for the past 250 years. Further, sea otters and other marine predators, whose distribution is limited by ice, have the potential to expand in distribution and abundance, reshaping future marine communities in the wake of deglaciation and global loss of sea ice.

Abstract Invasive species introductions in high latitudes are accelerating and elevating the need to address questions of their effects on Subarctic and Arctic ecosystems. As a driver of ecosystem function, submerged aquatic vegetation is one of the most deleterious biological invasions to aquatic food webs. The aquatic plant Elodea spp. has potential to be a widespread invader to Arctic and Subarctic ecosystems and is already established in 19 waterbodies in Alaska, USA. Elodea spp. has been found to alter ecosystem processes through multiple pathways; yet little is known about the impact of Elodea spp. on fish life history. A primary concern is the effect of Elodea spp. on juvenile Pacific salmon ( Oncorhynchus spp.), because this invading plant can form dense stands in littoral zones, potentially impacting important freshwater rearing habitats used by juvenile fish for foraging and refuge from predators. We used a field experiment to test the effect of Elodea spp. on juvenile coho salmon ( O. kisutch ) growth in an infested lake near Cordova, Alaska, USA. We found that Elodea spp. stands result in reduced growth and a lower trophic position for juvenile coho salmon over the summer compared to habitats dominated by a native assemblage of aquatic plants. While infested sites were not associated with significant changes in water condition or primary productivity compared to sites dominated by native vegetation, zooplankton densities were reduced, and Elodea spp. height and vegetation richness increased macroinvertebrate densities. Combined, these results indicate that Elodea spp. may alter the flow of energy to juvenile salmon by restructuring space and affecting prey resources for rearing fish. Furthermore, these results suggest that widespread establishment of Elodea spp. may alter the quality of habitat for juvenile salmon and, by affecting juvenile fish growth, could lead to population-level impacts on salmon returns.

Mercury (Hg) is a widespread element and persistent pollutant, harmful to fish, wildlife, and humans in its organic, methylated form. The risk of Hg contamination is driven by factors that regulate Hg loading, methylation, bioaccumulation, and biomagnification. In remote locations, with infrequent access and limited data, understanding the relative importance of these factors can pose a challenge. Here, we assessed Hg concentrations in an apex predator fish species, lake trout (Salvelinus namaycush), collected from 14 lakes spanning two National Parks in southwest Alaska, U.S.A. We then examined factors associated with the variation in fish Hg concentrations using a Bayesian hierarchical model. We found that total Hg concentrations in water were consistently low among lakes (0.11–0.50 ng L−1). Conversely, total Hg concentrations in lake trout spanned a thirty-fold range (101–3046 ng g−1 dry weight), with median values at 7 lakes exceeding Alaska's human consumption threshold. Model results showed that fish age and, to a lesser extent, body condition best explained variation in Hg concentration among fish within a lake, with Hg elevated in older, thinner lake trout. Other factors, including plankton methyl Hg content, fish species richness, volcano proximity, and glacier loss, best explained variation in lake trout Hg concentration among lakes. Collectively, these results provide evidence that multiple, hierarchically nested factors control fish Hg levels in these lakes.

Abstract In high‐latitude lakes, air temperature is an important driver of ice cover thickness and duration, which in turn influence water temperature and primary production supporting lake consumers and predators. In lieu of multidecadal observational records necessary to assess the response of lakes to long‐term warming, we used otolith‐based growth records from a long‐lived resident lake fish, lake trout ( Salvelinus namaycush), as a proxy for production. Lake trout were collected from seven deep, oligotrophic lakes in Lake Clark National Park and Preserve on in southwest Alaska that varied in the presence of marine‐derived nutrients (MDN) from anadromous sockeye salmon ( Oncorhynchus nerka) . Linear mixed‐effects models were used to partition variation in lake trout growth by age and calendar‐year and model comparisons tested for a mean increase in lake trout growth with sockeye salmon presence. Year effects from the best mixed‐effects model were subsequently compared to indices of temperature, lake ice, and regional indices of sockeye salmon escapement. A strong positive correlation between annual lake trout growth and temperature suggested that warmer springs, earlier lake ice break‐up, and a longer ice‐free growing season increase lake trout growth via previously identified bottom‐up increases in production with warming. Accounting for differences in the presence or annual escapement of sockeye salmon with available data did not improve model fit. Collectively with other studies, the results suggest that productivity of subarctic lakes has benefitted from warming spring temperatures and that temperature can synchronise otolith growth across lakes with and without sockeye salmon MDN.

Abstract Species distribution models (SDMs) are used to map and predict the geographic distributions of animals based on environmental covariates. Typically, SDMs require high‐resolution habitat data and time series information on animal locations. For data‐limited regions, defined as having scarce habitat or animal survey data, modeling is more challenging, often failing to incorporate important environmental attributes. For example, for sea otters ( Enhydra lutris ), a federally protected keystone species with variable population trends across the species' range, predictive modeling of distributions has been successfully conducted in areas with robust sea otter population and habitat data. We used open‐access data and employed a presence‐only model, maximum entropy (MaxEnt), to investigate subtidal habitat associations (substrate and algal cover, bathymetry, and rugosity) of northern sea otters ( E. lutris kenyoni ) for a data‐limited ecosystem, represented by Kachemak Bay, Alaska. Habitat association results corroborated previous findings regarding the importance of bathymetry and understory kelp as predictors of sea otter presence. Novel associations were detected as filamentous algae and shell litter were positively and negatively associated with northern sea otter presence, respectively, advancing existing knowledge of sea otter benthic habitat associations useful for predicting habitat suitability. This study provides a quantitative framework for conducting species distribution modeling with limited temporal and spatial animal distribution and abundance data. Utilizing drop camera information, our novel approach allowed for a better understanding of habitat requirements for a stable northern sea otter population, including bathymetry, understory kelp, and filamentous algae as positive predictors of sea otter presence in Kachemak Bay, Alaska.

Population dynamics vary in space and time. Survey designs that ignore these dynamics may be inefficient and fail to capture essential spatio-temporal variability of a process. Alternatively, dynamic survey designs explicitly incorporate knowledge of ecological processes, the associated uncertainty in those processes, and can be optimized with respect to monitoring objectives. We describe a cohesive framework for monitoring a spreading population that explicitly links animal movement models with survey design and monitoring objectives. We apply the framework to develop an optimal survey design for sea otters in Glacier Bay. Sea otters were first detected in Glacier Bay in 1988 and have since increased in both abundance and distribution; abundance estimates increased from 5 otters to >5,000 otters, and they have spread faster than 2.7 km/yr. By explicitly linking animal movement models and survey design, we are able to reduce uncertainty associated with forecasting occupancy, abundance, and distribution compared to other potential random designs. The framework we describe is general, and we outline steps to applying it to novel systems and taxa.