Suzanne Kohin

Prediction of fishing effort distributions using boosted regression trees.

Concerns about bycatch of protected species have become a dominant factor shaping fisheries management. However, efforts to mitigate bycatch are often hindered by a lack of data on the distributions of fishing effort and protected species. One approach to overcoming this problem has been to overlay the distribution of past fishing effort with known locations of protected species, often obtained through satellite telemetry and occurrence data, to identify potential bycatch hotspots. This approach, however, generates static bycatch risk maps, calling into question their ability to forecast into the future, particularly when dealing with spatiotemporally dynamic fisheries and highly migratory bycatch species. In this study, we use boosted regression trees to model the spatiotemporal distribution of fishing effort for two distinct fisheries in the North Pacific Ocean, the albacore (Thunnus alalunga) troll fishery and the California drift gillnet fishery that targets swordfish (Xiphias gladius). Our results suggest that it is possible to accurately predict fishing effort using < 10 readily available predictor variables (cross-validated correlations between model predictions and observed data -0.6). Although the two fisheries are quite different in their gears and fishing areas, their respective models had high predictive ability, even when input data sets were restricted to a fraction of the full time series. The implications for conservation and management are encouraging: Across a range of target species, fishing methods, and spatial scales, even a relatively short time series of fisheries data may suffice to accurately predict the location of fishing effort into the future. In combination with species distribution modeling of bycatch species, this approach holds promise as a mitigation tool when observer data are limited. Even in data-rich regions, modeling fishing effort and bycatch may provide more accurate estimates of bycatch risk than partial observer coverage for fisheries and bycatch species that are heavily influenced by dynamic oceanographic conditions.

Age validation of the blue shark (Prionace glauca) in the eastern Pacific Ocean

The blue shark (Prionace glauca) is subjected to high levels of fishery catch and by-catch worldwide; thus, knowledge of their productivity and population status is vital, yet basic assumptions of band-pair deposition rates in vertebrae used for age and growth models are being made without direct validation studies in the Pacific Ocean. As such, the purpose of the present study was to validate vertebral band-deposition rates of blue sharks tagged and recaptured in the eastern Pacific Ocean. Vertebrae of 26 blue sharks marked with oxytetracycline (OTC) were obtained from tag–recapture activities to determine timing of centrum growth-band deposition. Results from band counts distal to the OTC mark on each vertebra indicated that a single band pair (1 translucent and 1 opaque) is formed per year for blue sharks ranging from 1 to 8 years of age. Length–frequency modal analysis was also used to obtain growth estimates from a dataset spanning 26 years of research and commercial catch data. Results provide support for annual band-pair deposition in blue shark vertebrae and will aid in future blue shark age and growth studies in the Pacific Ocean.

State Space Model for Light Based Tracking of Marine Animals: Validation on Swimming and Diving Creatures

Determining light-based geolocations of electronically tagged marine animals by using irradiance measurements in an integrated random walk movement model is presented for improving the precision of the reconstructed geographical tracks. The model avoids making any irradiance threshold assumptions, or constraining the movement of the tag between dawn and dusk, which has limited the efficacy of previous models. Unlike previous models which relied on using previously calculated raw geolocations from outside sources, the new model uses irradiance data to produce the most probable track. The model generates two estimates of geographic positions per day (at dawn and dusk). Prior to this study, the model has been successfully evaluated for tags mounted on moorings, drifters, and in simulated cases. This paper documents the application of the model to wild caught swimming and diving mako sharks and blue marlin. The reconstructed tracks via the new model are compared to tracks adjusted using sea-surface temperature and to very accurate satellite-based tracks. The model performs well for the tested tags, which is surprising as these are pop-off satellite archival tags (PSATs), from which only a small fraction of the light record around approximately dusk and dawn is stored. The integrated model was vastly superior to classical purely light-based methods in all cases.

The political biogeography of migratory marine predators

During their migrations, marine predators experience varying levels of protection and face many threats as they travel through multiple countries’ jurisdictions and across ocean basins. Some populations are declining rapidly. Contributing to such declines is a failure of some international agreements to ensure effective cooperation by the stakeholders responsible for managing species throughout their ranges, including in the high seas, a global commons. Here we use biologging data from marine predators to provide quantitative measures with great potential to inform local, national and international management efforts in the Pacific Ocean. We synthesized a large tracking data set to show how the movements and migratory phenology of 1,648 individuals representing 14 species—from leatherback turtles to white sharks—relate to the geopolitical boundaries of the Pacific Ocean throughout species’ annual cycles. Cumulatively, these species visited 86% of Pacific Ocean countries and some spent three-quarters of their annual cycles in the high seas. With our results, we offer answers to questions posed when designing international strategies for managing migratory species.Daily location data on the individuals of 14 migratory marine species from 2000 to 2009 allow annual migratory cycles to be mapped to the time spent in the high seas and the exclusive economic zones of specific countries, providing a basis for international management strategies for these species.

Basking Shark (Cetorhinus maximus) Movements in the Eastern North Pacific Determined Using Satellite Telemetry

To fill data gaps on movements, behaviors and habitat use both near- and offshore, two programs were initiated to deploy satellite tags on basking sharks. Basking sharks are large filter feeding sharks that are second in size only to whale sharks. Similar to many megafauna populations, available data suggest that populations are below historic levels. In the northeast Pacific Ocean, the limited information on basking sharks comes from nearshore habitats where they forage. From 2010-2011, four sharks were tagged with pop-off satellite archival tags with deployments ranging from 9-240 days. The tags provided both transmitted and archived data on habitat use and geographic movement patterns. Nearshore, sharks tended to move north in the summer and prefer shelf and slope habitat around San Diego, Point Conception and Monterey Bay. The two sharks with 180 and 240 day deployments left the coast in the summer and fall. Offshore their paths diverged and by January one shark had moved to near the tip of the Baja Peninsula, Mexico and the other to the waters near Hawaii, USA. Vertical habitat use was variable both within and among individuals and changed as sharks moved offshore. Nearshore, most time was spent in the mixed layer but sharks did spend hours in cold waters below the mixed layer. Offshore vertical movements depended on location. The shark that went to Hawaii had a distinct diel pattern, with days spent at ~450-470 m and nights at ~250-300 m and almost no time in surface waters, corresponding with the diel migration of a specific portion of the deep scattering layer. The shark that moved south along the Baja Peninsula spent progressively more time in deep water but came to the surface daily. Movement patterns and shifts in vertical habitat and use are likely linked to shifts in prey availability. Data collected indicate the potential for large-scale movements and the need for international dialogue in any recovery efforts.

An analytical approach to sparse telemetry data

Horizontal behavior of highly migratory marine species is difficult to decipher because animals are wide-ranging, spend minimal time at the ocean surface, and utilize remote habitats. Satellite telemetry enables researchers to track individual movements, but population level inferences are rare due to data limitations that result from difficulty of capture and sporadic tag reporting. We introduce a Bayesian modeling framework to address population level questions with satellite telemetry data when data are sparse. We also outline an approach for identifying informative variables for use within the model. We tested our modeling approach using a large telemetry dataset for Shortfin Makos (Isurus oxyrinchus), which allowed us to assess the effects of various degrees of data paucity. First, a permuted Random Forest analysis is implemented to determine which variables are most informative. Next, a generalized additive mixed model is used to help define the relationship of each remaining variable with the response variable. Using jags and rjags for the analysis of Bayesian hierarchical models using Markov Chain Monte Carlo simulation, we then developed a movement model to generate parameter estimates for each of the variables of interest. By randomly reducing the tagging dataset by 25, 50, 75, and 90 percent and recalculating the parameter estimates, we demonstrate that the proposed Bayesian approach can be applied in data-limited situations. We also demonstrate how two commonly used linear mixed models with maximum likelihood estimation (MLE) can be similarly applied. Additionally, we simulate data from known parameter values to test each model’s ability to recapture those values. Despite performing similarly, we advocate using the Bayesian over the MLE approach due to the ability for later studies to easily utilize results of past study to inform working models, and the ability to use prior knowledge via informed priors in systems where such information is available.

Relative catchability (probability of presence) of broadbill swordfish Xiphias gladius in the California Current System. Hindcast predictions for November 2014 using UC Santa Cruz Regional Ocean Modelling System

Model-derived relative probability of swordfish presence, hind-cast on Regional Ocean Model outputs for the period 1st-31st November 2014.