The role of spatial structure in at-risk metapopulation recoveries.

Metapopulations are often managed as a single contiguous population despite the spatial structure underlying their local and regional dynamics. Disturbances from human activities can also be spatially structured with mortality impacts concentrated to just a few local populations among the aggregate. Scale transitions between local and regional processes can generate emergent properties whereby the whole system can fail to recover as quickly as expected for an equivalent single population. Here, we draw on theory and empirical case studies to ask: what is the consequence of spatially structured ecological and disturbance processes on metapopulation recoveries? We suggest that exploring this question could help address knowledge gaps for managing metapopulations including: Why do some metapopulations recover quickly while others remain collapsed? And, what risks are unaccounted for when metapopulations are managed at aggregate scales? First, we used model simulations to examine how scale transitions among ecological and disturbance conditions interact to generate emergent metapopulation recovery outcomes. In general, we found that the spatial structure of disturbance was a strong determinant of recovery outcomes. Specifically, disturbances that unevenly impacted local populations consistently generated the slowest recoveries and highest conservation risks. Ecological conditions that dampened metapopulation recoveries included low dispersal, variable local demography, sparsely connected habitat networks, and spatially and temporally correlated stochastic processes. Second, we illustrate the unexpected challenges of managing metapopulations by examining the recoveries of three USA federally listed endangered species: Florida Everglade snail kites, California and Alaska sea otters, and Snake River Chinook salmon. Overall, our results show the pivotal role of spatial structure in metapopulation recoveries whereby the interplay between local and regional processes shapes the resilience of the whole system. With this understanding, we provide guidelines for resource managers tasked with conserving and managing metapopulations and identify opportunities for research to support the application of metapopulation theory to real-world challenges.

Marine and freshwater regime changes impact a community of migratory Pacific salmonids in decline.

Marine and freshwater ecosystems are increasingly at risk of large and cascading changes from multiple human activities (termed "regime shifts"), which can impact population productivity, resilience, and ecosystem structure. Pacific salmon exhibit persistent and large fluctuations in their population dynamics driven by combinations of intrinsic (e.g., density dependence) and extrinsic factors (e.g., ecosystem changes, species interactions). In recent years, many Pacific salmon have declined due to regime shifts but clear understanding of the processes driving these changes remains elusive. Here, we unpacked the role of density dependence, ecosystem trends, and stochasticity on productivity regimes for a community of five anadromous Pacific salmonids (Steelhead, Coho Salmon, Pink Salmon, Dolly Varden, and Coastal Cutthroat Trout) across a rich 40-year time-series. We used a Bayesian multivariate state-space model to examine whether productivity shifts had similarly occurred across the community and explored marine or freshwater changes associated with those shifts. Overall, we identified three productivity regimes: an early regime (1976-1990), a compensatory regime (1991-2009), and a declining regime (since 2010) where large declines were observed for Steelhead, Dolly Varden, and Cutthroat Trout, intermediate declines in Coho and no change in Pink Salmon. These regime changes were associated with multiple cumulative effects across the salmon life cycle. For example, increased seal densities and ocean competition were associated with lower adult marine survival in Steelhead. Watershed logging also intensified over the past 40 years and was associated with (all else equal) ≥97% declines in freshwater productivity for Steelhead, Cutthroat, and Coho. For Steelhead, marine and freshwater dynamics played approximately equal roles in explaining trends in total productivity. Collectively, these changing environments limited juvenile production and lowered future adult returns. These results reveal how changes in freshwater and marine environments can jointly shape population dynamics among ecological communities, like Pacific salmon, with cascading consequences to their resilience.

Assessing conservation status with extensive but low-resolution data: Application of frequentist and Bayesian models to endangered Athabasca River rainbow trout.

Use of extensive but low-resolution abundance data is common in the assessment of species at-risk status based on quantitative decline criteria under International Union for Conservation of Nature (IUCN) and national endangered species legislation. Such data can be problematic for 3 reasons. First, statistical power to reject the null hypothesis of no change is often low because of small sample size and high sampling uncertainty leading to a high frequency of type II errors. Second, range-wide assessments composed of multiple site-specific observations do not effectively weight site-specific trends into global trends. Third, uncertainty in site-specific temporal trends and relative abundance are not propagated at the appropriate spatial scale. A common result is the propensity to underestimate the magnitude of declines and therefore fail to identify the appropriate at-risk status for a species. We used 3 statistical approaches, from simple to more complex, to estimate temporal decline rates for a designatable unit (DU) of rainbow trout in the Athabasca River watershed in western Canada. This DU is considered a native species for purposes of listing because of its genetic composition characterized as >0.95 indigenous origin in the face of continuing introgressive hybridization with introduced populations in the watershed. Analysis of abundance trends from 57 time series with a fixed-effects model identified 33 sites with negative trends, but only 2 were statistically significant. By contrast, a hierarchical linear mixed model weighted by site-specific abundance provided a DU-wide decline estimate of 16.4% per year and a 3-generation decline of 93.2%. A hierarchical Bayesian mixed model yielded a similar 3-generation decline trend of 91.3% and the posterior distribution showed that the estimate had a >99% probability of exceeding thresholds for an endangered listing. We conclude that the Bayesian approach was the most useful because it provided a probabilistic statement of threshold exceedance in support of an at-risk status recommendation.

El uso de datos extensivos, pero de baja resolución, de la abundancia es una práctica común en la evaluación del estado de riesgo de una especie con base en los criterios cuantitativos de declinación establecidos por la Unión Internacional para la Conservación de la Naturaleza (UICN) y la legislación nacional sobre especies en peligro extinción. Dicha información puede ser problemática por tres razones: primero, el poder estadístico para rechazar la hipótesis nula de ningún cambio es frecuentemente bajo debido a un tamaño pequeño de la muestra y a la elevada incertidumbre del muestreo, lo que resulta en una frecuencia elevada de errores de tipo II; segundo, las evaluaciones de amplia variedad compuestas de varias observaciones específicas de sitio no sopesan efectivamente las tendencias específicas de sitio dentro de las tendencias globales; y tercero, la incertidumbre en las tendencias temporales específicas de sitio y en la abundancia relativa no se propagan a la escala espacial apropiada. Un resultado común del uso de esta información es la propensión a subestimar la magnitud de las declinaciones, y por lo tanto equivocarse en la identificación del estado de riesgo apropiado para la especie. Usamos tres estrategias estadísticas, de simples a más complejas, para estimar las tasas de declinación temporal para una unidad designable (UD) de trucha arcoíris en la cuenca del río Athabasca al oeste de Canadá. Esta UD es considerada una especie nativa por razones de listado debido a su composición genética, caracterizada como >0-95 de origen nativo de frente a la continua hibridación introgresiva con poblaciones introducidas a la cuenca. El análisis de las tendencias de abundancia de 57 series de tiempo con un modelo de efectos fijos identificó 33 sitios con tendencias negativas, pero sólo dos fueron estadísticamente significativas. En contraste, un modelo lineal mixto de jerarquías sopesado por abundancia específica de sitio proporcionó una estimación de declinación en toda la UD de 16.4% año−1 y una declinación a tres generaciones de 93.2%. Un modelo bayesiano de jerarquías produjo una tendencia de declinación a tres generaciones de 91.3% y la distribución posterior mostró que el estimado tuvo una probabilidad >99% de exceder los umbrales para la categorización como especie en peligro. Concluimos que la estrategia bayesiana fue la más útil porque proporcionó una afirmación probabilística de la superación del umbral a favor de una recomendación de categorizar el estado como en riesgo.

Hotspots for rockfishes, structural corals, and large-bodied sponges along the central coast of Pacific Canada.

Biological hotspots are places with outstanding biodiversity features, and their delineation is essential to the design of marine protected areas (MPAs). For the Central Coast of Canada's Northern Shelf Bioregion, where an MPA network is being developed, we identified hotspots for structural corals and large-bodied sponges, which are foundation species vulnerable to bottom contact fisheries, and for Sebastidae, a fish family which includes species that are long-lived (> 100 years), overexploited, evolutionary distinctive, and at high trophic levels. Using 11 years of survey data that spanned from inland fjords to oceanic waters, we derived hotspot indices that accounted for species characteristics and abundances and examined hotspot distribution across depths and oceanographic subregions. The results highlight previously undocumented hotspot distributions, thereby informing the placement of MPAs for which high levels of protection are warranted. Given the vulnerability of the taxa that we examined to cumulative fishery impacts, prospective MPAs derived from our data should be considered for interim protection measures during the protracted period between final network design and the enactment of MPA legislations. These recommendations reflect our scientific data, which are only one way of understanding the seascape. Our surveys did not cover many locations known to Indigenous peoples as biologically important. Consequently, Indigenous knowledge should also contribute substantially to the design of the MPA network.

Life-history variation along environmental and harvest clines of a northern freshwater fish: Plasticity and adaptation.

Plasticity, local adaptation and evolutionary trade-offs drive clinal variation in traits associated with lifetime growth. Disentangling the processes and determinants that cause these traits to vary helps to understand species' responses to changing environments. This is particularly urgent for exploited populations, where size-selective harvest can induce life-history evolution. Lake trout (Salvelinus namaycush) are an exploited fish with a life history adapted to low-productivity freshwaters of northern North America, which makes them highly vulnerable to ecosystem changes and overfishing. We characterized life-history variation across a broad and diverse landscape for this iconic northern freshwater fish and evaluated whether clinal variation was consistent with hypotheses for local adaptation or growth plasticity. We estimated growth-associated traits for 90 populations exposed to a diversity of environments using a Bayesian multivariate hierarchical model. We tested for clinal variation in their somatic growth, size at maturity and reproductive allocation along environmental gradients of lake productivity, climate, prey and exploitation clines under competing hypotheses of plasticity and local adaptation. Clinal life-history variation was consistent with growth plasticity and local adaptations but not harvest-induced evolution. Variation in somatic growth was explained by exploitation, climate and prey fish occurrence. Increased exploitation, from pristine to fully exploited conditions, led to increased somatic growth (from 32 to 45 mm/year) and adult life spans, and reduced age at maturity (from 11 to 8 years). Variation in size at maturity was explained by climate and, less certainly, prey fish occurrence, while reproductive allocation was explained by evolutionary trade-offs with mortality and other traits, but not environment. Lake trout life-history variation within this range was as wide as that observed across dozens of other freshwater species. Lake trout life histories resulted from evolutionary trade-offs, growth plasticity and local adaptations along several environmental clines. Presuming a plastic response, we documented ~1.4-fold growth compensation to exploitation-lower growth compensation than observed in many freshwater fishes. These results suggest that harvested species exposed to spatially structured and diverse environments may have substantial clinal variation on different traits, but due to different processes, and this has implications for their resilience and successful management.

Interaction of ecological and angler processes: experimental stocking in an open access, spatially structured fishery.

Effective management of socioecological systems requires an understanding of the complex interactions between people and the environment. In recreational fisheries, which are prime examples of socioecological systems, anglers are analogous to mobile predators in natural predator-prey systems, and individual fisheries in lakes across a region are analogous to a spatially structured landscape of prey patches. Hence, effective management of recreational fisheries across large spatial scales requires an understanding of the dynamic interactions among ecological density dependent processes, landscape-level characteristics, and angler behaviors. We focused on the stocked component of the open access rainbow trout (Oncorhynchus mykiss) fishery in British Columbia (BC), and we used an experimental approach wherein we manipulated stocking densities in a subset of 34 lakes in which we monitored angler effort, fish abundance, and fish size for up to seven consecutive years. We used an empirically derived relationship between fish abundance and fish size across rainbow trout populations in BC to provide a measure of catch-based fishing quality that accounts for the size-abundance trade off in this system. We replicated our experimental manipulation in two regions known to have different angler populations and broad-scale access costs. We hypothesized that angler effort would respond to variation in stocking density, resulting in spatial heterogeneity in angler effort but homogeneity in catch-based fishing quality within regions. We found that there is an intermediate stocking density for a given lake or region at which angler effort is maximized (i.e., an optimal stocking density), and that this stocking density depends on latent effort and lake accessibility. Furthermore, we found no clear effect of stocking density on our measure of catch-based fishing quality, suggesting that angler effort homogenizes catch-related attributes leading to an eroded relationship between stocking density and catch-based fishing quality at the timescale of annual surveys. We conclude that declines in fishing quality resulting from understocking (due to declines in catch rate with low fish abundance) and overstocking (due to suppressed growth and limited recruitment at high density) give an optimal stocking rate that depends on accessibility and latent effort.

Supply-demand equilibria and the size-number trade-off in spatially structured recreational fisheries.

Recreational fishing effort varies across complex inland landscapes (e.g., lake-districts) and appears influenced by both angler preferences and qualities of the fishery resource, like fish size and abundance. However, fish size and abundance have an ecological trade-off within a population, thereby structuring equal-quality isopleths expressing this trade-off across the fishing landscape. Since expressed preferences of recreational anglers (i.e., site-selection of high-quality fishing opportunities among many lakes) can be analogous to optimal foraging strategies of natural predators, adopting such concepts can aid in understanding scale-dependence in fish-angler interactions and impacts of fishing across broad landscapes. Here, we assumed a fish supply-angler demand equilibria and adapted a novel bivariate measure of fishing quality based on fish size and catch rates to assess how recreational anglers influence fishing quality among a complex inland landscape. We then applied this metric to evaluate (1) angler preferences for caught and released fish compared to harvested fish, (2) the nonlinear size-numbers trade-off with uncertainty in both traits, and (3) the spatial-scale of the equilibria across 62 lakes and four independent management regions in British Columbia's (BC) rainbow trout Oncorhynchus mykiss fishery. We found anglers had low preference for caught and released fish (~10% of the value compared to harvested fish), which modified anglers' perception of fishing quality. Hence, fishing quality and angler effort was not influenced simply by total fish caught, but largely by harvested fish catch rates. Fishing quality varied from BC's northern regions (larger fish and more abundant) compared to southern regions (smaller fish and less abundant) directly associated with a 2.5 times increase in annual fishing effort in southern regions, suggesting that latent fishing pressure can structure the size-numbers trade-off in rainbow trout populations. The presence of two different equal-quality isopleths suggests at least two effective landscapes support co-occurring ideal free distributions of recreational fishing effort in BC's rainbow fishery. Anglers' expressed preferences among lakes interacted with density dependent growth and survival within lakes to structure a size-numbers trade-off influencing how anglers perceive fishing quality and, ultimately, distribute across complex inland landscapes.