For many marine species and habitats, climate change and overfishing present a double threat. To manage marine resources effectively, it is necessary to adapt management to changes in the physical environment. Simple relationships between environmental conditions and fish abundance have long been used in both fisheries and fishery management. In many cases, however, physical, biological, and human variables feed back on each other. For these systems, associations between variables can change as the system evolves in time. This can obscure relationships between population dynamics and environmental variability, undermining our ability to forecast changes in populations tied to physical processes. Here we present a methodology for identifying physical forcing variables based on nonlinear forecasting and show how the method provides a predictive understanding of the influence of physical forcing on Pacific sardine.ecosystem-based management | physical-biological interactions | state space reconstruction | complex systems | time series analysis E cosystem-based management (EBM) is an essential challenge that places strong demands on our understanding of coupled social-ecological systems. EBM requires an understanding of how human activities such as fishing influence and are influenced by other parts of the ecosystem. This includes accounting for the effects of the physical environment on exploited populations. However, the interactions between ecosystem components can be complex, and unraveling physical-biological interactions remains a challenge. For instance, a retrospective study of 35 exploited and unexploited species in the California Current shows that fishing pressure can amplify the influence of environmental forcing on populations by truncating the age structure (1). This study and others (2-4) demonstrate that the effect of environmental forcing on populations can be contingent on fishing effort, current abundance, and age structure. This raises an important issue: Ecosystem variables are not separate, decomposable forces. Instead, their interactions are state-dependent, meaning that the impact of one variable on another depends on the state of the variables.State-dependent behavior can confound many traditional statistical methods. Witness that valid correlations between physical and biological variables can be difficult to find (5) and can appear and disappear with time (6). In fact, nonlinear systems (systems with state-dependent interactions) can produce mirage correlations: variables that seem positively correlated over one period in time may seem negatively correlated or unrelated over another period (7). A meta-analysis of environment-recruitment relationships in marine populations shows that these correlations hold up poorly when retested with new data (6). Consequently, EBM requires more robust methods for identifying driving variables and understanding their influence on population and community dynamics. Here, we show that methods based on multivariate state space reconstruction (SSR) (8) offer ...

No abstract

Accurate predictions of species abundance remain one of the most vexing challenges in ecology. This observation is perhaps unsurprising, because population dynamics are often strongly forced and highly nonlinear. Recently, however, numerous statistical techniques have been proposed for fitting highly parameterized mechanistic models to complex time series, potentially providing the machinery necessary for generating useful predictions. Alternatively, there is a wide variety of comparatively simple model-free forecasting methods that could be used to predict abundance. Here we pose a rather conservative challenge and ask whether a correctly specified mechanistic model, fit with commonly used statistical techniques, can provide better forecasts than simple model-free methods for ecological systems with noisy nonlinear dynamics. Using four different control models and seven experimental time series of flour beetles, we found that Markov chain Monte Carlo procedures for fitting mechanistic models often converged on best-fit parameterizations far different from the known parameters. As a result, the correctly specified models provided inaccurate forecasts and incorrect inferences. In contrast, a model-free method based on state-space reconstruction gave the most accurate short-term forecasts, even while using only a single time series from the multivariate system. Considering the recent push for ecosystem-based management and the increasing call for ecological predictions, our results suggest that a flexible model-free approach may be the most promising way forward. U nderstanding fluctuations in species abundance is a longstanding goal of ecology and is particularly important for the conservation of severely depleted populations (1, 2). Dramatic changes in species abundance are common (3) and large declines can have disastrous impacts on ecosystem users (4). Accurate forecasts could allow for improved conservation efforts and increased fishery yields, however, ecological surprises are common (5), and accurate predictions remain a major challenge (6).As ecological time series continue to lengthen, and computerintensive statistics become more convenient, there has been an increasing trend toward fitting complex mechanistic models that incorporate both process and observation error (i.e., state-space models) (7). A Bayesian fitting procedure for such a model typically involves computationally intensive Markov chain Monte Carlo (MCMC) sampling of the joint posterior probability distribution to find the best fitting parameters (8). Previous work concluded that misspecified mechanistic models can produce poor forecasts (9), but it has been suggested that state-space models may be more robust to misspecification due to their inclusion of a process-error term (8); however, their forecast accuracy remains largely untested.Model-free time series analysis provides an alternative method for generating forecasts that, unlike most state-space fitting procedures, does not require intensive computations. Although there exists a vas...

While ecologists have long recognized the influence of spatial resolution on species distribution models (SDMs), they have given relatively little attention to the influence of temporal resolution. Considering temporal resolutions is critical in distribution modelling of highly mobile marine animals, as they interact with dynamic oceanographic processes that vary at time-scales from seconds to decades. We guide ecologists in selecting temporal resolutions that best match ecological questions and ecosystems, and managers in applying these models. We group the temporal resolutions of environmental variables used in SDMs into three classes: instantaneous, contemporaneous and climatological. We posit that animal associations with fine-scale and ephemeral | 1099MANNOCCI et Al. | INTRODUCTIONHighly mobile marine animals such as marine mammals, seabirds, sea turtles and fish are unevenly distributed in the ocean. Ecologists have long sought to understand and predict their patterns of distributions, particularly for commercially valuable species subject to exploitation (Lehodey, Bertignac, Hampton, Lewis, & Picaut, 1997) and for protected species vulnerable to incidental harm (Reilly, 1990). They often employ species distribution models (SDMs) that statistically relate distribution patterns to environmental conditions by linking animal observations to environmental variables. SDMs have been successfully used to examine many ecological, management and conservation questions (Elith & Leathwick, 2009). In particular, they have been widely used to explain and predict distribution patterns of highly mobile marine animals in a variety of ecosystems (Benson et al., 2011;Forney, Becker, Foley, Barlow, & Oleson, 2015;Hartog, Hobday, Matear, & Feng, 2011;Mannocci et al., 2014).It has become apparent that the hierarchical structure of processes in the marine environment drives the distribution and movement patterns of marine animals at multiple spatio-temporal scales (Benoit-Bird, Battaile, Nordstrom, & Trites, 2013;Fauchald, Erikstad, & Skarsfjord, 2000;Fauchald & Tveraa, 2006;Fritz, Said, & Weimerskirch, 2003;Pinaud & Weimerskirch, 2005) (Figure 1). At fine scales, animals track ephemeral prey patches that extend over tens of metres to satisfy their energy requirements (Goldbogen et al., 2008;Heaslip, Iverson, Bowen, & James, 2012 (Benson et al., 2011;Hobday & Hartog, 2014;Tew Kai & Marsac, 2010). At broad scales, animals associate with persistent water masses and current systems that extend over thousands of kilometres and delimit their geographic ranges or migration routes (Jaquet, Whitehead, & Lewis, 1996;Reygondeau et al., 2012;Shillinger et al., 2008). Thus, the distributions of highly mobile marine animals appear determined by both short-term ocean variability and persistent patterns of longer-term ocean climate.Researchers use a variety of methods to obtain synoptic data on marine animal distributions and the marine environment at a wide range of spatial and temporal extents ( Figure 2, see Appendix S1 in Supporti...