Eric J. Ward’s research while affiliated with National Oceanic and Atmospheric Administration and other places

As global climate change and anthropogenic activities amplify widespread environmental variability, there is a strong need for management strategies that incorporate relationships between ecosystem components. This need is especially apparent when changes in environmental drivers cause threshold responses (abrupt, nonlinear changes) in ecosystems. Such ecological thresholds can provide useful reference points for management decisions. However, methods for detecting thresholds in empirical datasets may fail to find an existing threshold, find one that does not exist, or be biased in their estimates of threshold locations. These types of threshold misspecifications can result in high conservation and socioeconomic costs. Simulation studies can mitigate these risks by providing information about method performance across different scenarios. Here, we constructed a series of simulations to evaluate the robustness of threshold detection with generalized additive models (GAMs) when exposed to a variety of common, real‐world data characteristics. GAMs generally performed best when time series were long, observation error was low, thresholds were crossed fairly frequently, and covariates were accounted for. Over realistic ranges of values, observation error and frequency of threshold crossing had stronger effects on threshold detectability than time series length. Importantly, detectability was found to depend on both the shape of the threshold relationship and the statistical definition of the threshold location. As a case study, we applied this threshold detection method to an empirical dataset relating ocean temperature and the spatial distribution of Pacific hake (Merluccius productus), the largest volume fishery on the US West Coast. While the data suggest no statistical evidence for a threshold relationship, our simulations indicated approximately equal chances of true and false threshold detection given currently available data. Our results provide general guidelines for where threshold detection with GAMs is likely to be robust and are useful in the context of indicator development for ecosystem‐based management in a variable world.

Fisheries management faces challenges due to political, spatial, and ecological complexities, which are further exacerbated by variation or shifts in species distributions. Effective management depends on the ability to integrate fisheries data across political and geographic boundaries. However, such efforts may be hindered by inconsistent data formats, limited data sharing, methodological differences in sampling, and regional governance differences. To address these issues, we introduce the surveyjoin R package, which combines and provides public access to bottom trawl survey data collected by NOAA Fisheries and Fisheries and Oceans Canada in the Northeast Pacific Ocean. This initial database integrates over 3.3 million observations from 14 bottom trawl surveys spanning Alaska, British Columbia, Washington, Oregon, and California from the 1980s to present. This effort standardizes variables such as catch-per-unit-effort (CPUE), haul data, and in-situ measurements of bottom temperature. We demonstrate the utility of this database through three case studies. The first develops a coastwide biomass index for Pacific hake (Merluccius productus) using geostatistical index standardization, comparing results to independent acoustic survey estimates. The second examines changes in the spatial distribution of groundfish species across marine heatwave and non-heatwave years, highlighting species-specific and community-level responses to warming events. The third applies spatially varying coefficient models to assess sablefish (Anoplopoma fimbria) biomass trends, identifying regional variability in increases in occurrence and biomass. Together, these case studies demonstrate how the surveyjoin R package and database may improve species and ecosystem assessments by providing insights into population trends across geopolitical boundaries. This database and package represent an important step toward offering a scalable framework that can be extended to include additional data types, surveys, and species. By fostering collaboration, transparency, and data-driven decision-making, surveyjoin supports international efforts to sustainably manage shared marine resources under dynamic environmental conditions.

Diel vertical movement (DVM) is a widespread behavior in aquatic ecosystems, occurring across a variety of taxa and water bodies. The factors hypothesized to drive DVM can vary tremendously through time, yet little is known about how DVM changes at interannual timescales. Here we explore how cyclical prey abundance affects predator DVM. Higher consumption levels increase the optimal temperatures for growth in fishes. Thus, annual variation in prey abundance should generate corresponding variation in the depths and temperatures selected during predator DVM. In Crater Lake, one of the deepest and most oligotrophic lakes in the world, Daphnia zooplankton exhibit cyclical patterns of abundance. We compiled data spanning four distinct pulses of Daphnia and analyzed the response of their predator, kokanee salmon (Oncorhynchus nerka). Our data spanned 36 yr for Daphnia abundance and kokanee body condition, and 24 yr for kokanee DVM (measured by hydroacoustic surveys). Kokanee exhibited four pulses in body weight and condition that corresponded to the four Daphnia pulses, suggesting a strong bottom‐up response. Kokanee altered their DVM in years with Daphnia by occurring deeper during the day, where Daphnia were concentrated, and shallower at night, where temperatures were > 5°C warmer. By selecting warmer habitat in years with Daphnia, kokanee increased their estimated overnight digestion by ~ 25%. Understanding how predators alter DVM and other patterns of cyclical habitat use in response to variation in prey abundance has important implications for understanding predator–prey dynamics, which are highly sensitive to prey encounter rates and maximum consumption rates.

Many species distribution models (SDMs) incorporate external information, such as environmental or habitat features, yet the majority overlook age-specific information. Including age information may be particularly valuable for species exhibiting age-related spatial patterns driven by ontogenetic shifts, recruitment dynamics, and selective fishing pressures. Here, we develop an age-structured SDM framework using data from the west coast of the USA and concentrate on two species with distinct life histories: North Pacific hake (Merluccius productus) and sablefish (Anoplopoma fimbria). We validate our approach by forecasting age classes across survey locations in future years; these results highlight that forecast models have predictive skill (sablefish more so than hake) and the predictive ability is highest for older age classes. Using predictions of age-1 sablefish as an example, we demonstrate how our models may be used to forecast future bycatch risk in space, helping increase the efficiency of fisheries. Our framework is applicable to a wide range of survey types and platforms around the world and supports the development of spatially explicit management strategies and optimal allocation of fishing effort.

Environmental and biological factors influencing fish larvae can drive fishery cohort strength, yet larval abundance is typically a better indicator of spawning biomass. Under a changing ocean, studies that explore the relationships between environmental variables, larval abundance, and fishery recruitment remain valuable areas for ongoing research. We focus on a popular, recreational-only, multispecies saltwater bass fishery (genus Paralabrax) whose population status and recovery potential are uncertain. We resolved Paralabrax spp. larval data to species over a 54-year period (1963-2016) and used species distribution models to (i) generate and test species-specific standardized indices of larval abundance as indicators of adult stock status and fishery recruitment and (ii) evaluate long-term spa-tiotemporal trends in their population dynamics relative to environmental variables and climate forcing. Contrary to initial hypotheses, species-specific larval abundance predicted future catches, with higher recent larval abundances suggesting potential for fishery recovery. Temperature, zooplankton biomass, and isothermal depth were important predictors of bass larval abundance, indicating these variables could also be valuable for predicting fishery recruitment and anticipating population change. Our findings paint a path for-w ard tow ards a more ecosystem-based fishery management approach for this important fishery and may serve as a template for data-or assessment-limited fisheries.

Species distribution modeling is increasingly used to describe and anticipate consequences of a warming ocean. These models often identify statistical associations between distribution and environmental conditions such as temperature and oxygen, but rarely consider the mechanisms by which these environmental variables affect metabolism. Oxygen and temperature jointly govern the balance of oxygen supply to oxygen demand, and theory predicts thresholds below which population densities are diminished. However, parameterizing models with this joint dependence is challenging because of the paucity of experimental work for most species, and the limited applicability of experimental findings in situ. Here we ask whether the temperature‐sensitivity of oxygen can be reliably inferred from species distribution observations in the field, using the U.S. Pacific Coast as a model system. We developed a statistical model that adapted the metabolic index — a compound metric that incorporates these joint effects on the ratio of oxygen supply and oxygen demand by applying an Arrhenius equation — and used a non‐linear threshold function to link the index to fish distribution. Through simulation testing, we found that our statistical model could not precisely estimate the parameters due to inherent features of the distribution data. However, the model reliably estimated an overall metabolic index threshold effect. When applied to case studies of real data for two groundfish species, this new model provided a better fit to spatial distribution of one species, sablefish Anoplopoma fimbria, than previously used models, but did not for the other, longspine thornyhead Sebastolobus altivelis. This physiological framework may improve predictions of species distribution, even in novel environmental conditions. Further efforts to combine insights from physiology and realized species distributions will improve forecasts of species' responses to future environmental changes.

Understanding the dynamic relationship between marine species and their changing environments is critical for ecosystem based management, particularly as coastal ecosystems experience rapid change (e.g., general warming, marine heat waves). In this paper, we present a novel statistical approach to robustly estimate and track the thermal niches of 30 marine fishes along the west coast of North America. Leveraging three long-term fisheries-independent datasets, we use spatiotemporal modeling tools to capture spatiotemporal variation in species densities. Estimates from our models are then used to generate species-specific estimates of thermal niches through time at several scales: coastwide and for each of the three regions. By synthesizing data across regions and time scales, our modeling approach provides insights into how these marine species may be tracking or responding to changes in temperature. While we did not find evidence of consistent temperature-density relationships among regions, we are able to contrast differences across species: Dover sole and shortspine thornyhead have relatively broad thermal niche estimates that are static over time, whereas several semi-pelagic species (e.g., Pacific hake, walleye pollock) have niches that are both becoming warmer over time and simultaneously narrowing. This illustrates how several economically and ecologically valuable species are facing contrasting fates in a changing environment, with potential consequences for fisheries and ecosystems. Our modeling approach is flexible and can be easily extended to other species or ecosystems, as well as other environmental variables. Results from these models may be broadly useful to scientists, managers, and stakeholders—monitoring trends in the direction and variability of thermal niches may be useful in identifying species that are more susceptible to environmental change, and results of this work can form quantitative metrics that may be included in climate vulnerability assessments, estimation of dynamic essential fish habitat, and assessments of climate risk posed to fishing communities.

Generating vast arrays of genetic markers for evolutionary ecology studies has become routine and cost-effective. However, analyzing data from large numbers of loci associated with a small number of finite chromosomes introduces a challenge: loci on the same chromosome do not assort independently, leading to pseudoreplication. Previous studies have demonstrated that pseudoreplication can substantially reduce precision of genetic analyses (and make confidence intervals wider), such as FST and linkage disequilibrium (LD) measures between pairs of loci. In LD analyses, another type of dependency (overlapping pairs of the same loci) also creates pseudoreplication. Building on previous work, we explore the potential of entropy metrics to improve the status quo, particularly total correlation (TC), to assess pseudoreplication in LD studies. Our simulations, performed on a monoecious population with a range of effective population sizes (Ne) and numbers of loci, attempted to isolate the overlapping-pairs-of-loci effect by considering unlinked loci and using entropy to quantify inter-locus relationships. We hypothesized a positive correlation between TC and the number of loci (L), and a negative correlation between TC and Ne. Results from our statistical models predicting TC demonstrate a strong effect of the number of loci, and muted effects of Ne and other predictors, adding support to the use of entropy-based metrics as a tool for estimating the statistical information of complex genetic datasets. Our results also highlight a challenge regarding scalability; computational limitations arise as the number of loci grows, making our current approach limited to smaller datasets. Despite these challenges, this work further refines our understanding of entropy measures, and offers insights into the complex dynamics of genetic information in evolutionary ecology research.

Growth variability is a key contributor to the dynamic productivity of populations. Thus, better understanding and accounting for variation in growth can improve both tactical and strategic management. Using ocean shrimp (Pandalus jordani) from the U.S. West Coast as a case study, we demonstrate interactions between growth and optimal fishery opening dates. While the fishery opens on 1 April, industry often delays fishing to minimize catches of small shrimp. Understanding drivers of size-at-recruitment can help managers optimize opening dates and shrimpers plan their participation in this and other fisheries. Using three decades of fishery-dependent sampling, we built a spatially, temporally, and environmentally explicit Bayesian state-space model for shrimp size-at-age. Model outputs were then used to parameterize a revenue-per-recruit model and explore how variability in size-at-recruitment impacted optimal opening dates. Shrimp recruited at smaller sizes farther north. Delaying opening would likely benefit shrimpers in areas and years with smaller shrimp and higher fishing mortality. Broadly, choosing when to open the fishery is a complex decision requiring understanding of growth, but also recruitment, economic incentives, and life history.

The study of animal diets and the proportional contribution that different foods make to their diets is an important task in ecology. Stable Isotope Mixing Models (SIMMs) are an important tool for studying an animal's diet and understanding how the animal interacts with its environment. We present cosimmr, a new R package designed to include covariates when estimating diet proportions in SIMMs, with simple functions to produce plots and summary statistics. The inclusion of covariates allows for users to perform a more in-depth analysis of their system and to gain new insights into the diets of the organisms being studied. A common problem with the previous generation of SIMMs is that they are very slow to produce a posterior distribution of dietary estimates, especially for more complex model structures, such as when covariates are included. The widely-used Markov chain Monte Carlo (MCMC) algorithm used by many traditional SIMMs often requires a very large number of iterations to reach convergence. In contrast, cosimmr uses Fixed Form Variational Bayes (FFVB), which we demonstrate gives up to an order of magnitude speed improvement with no discernible loss of accuracy. We provide a full mathematical description of the model, which includes corrections for trophic discrimination and concentration dependence, and evaluate its performance against the state of the art MixSIAR model. Whilst MCMC is guaranteed to converge to the posterior distribution in the long term, FFVB converges to an approximation of the posterior distribution, which may lead to sub-optimal performance. However we show that the package produces equivalent results in a fraction of the time for all the examples on which we test. The package is designed to be user-friendly and is based on the existing simmr framework.