Identifying fishing trip behaviour and estimating fishing effort from VMS data using Bayesian Hidden Markov Models

Recent advances in technologies have lead to a vast influx of data on movements, based on discrete recorded position of animals or fishing boats, opening new horizons for future analyses. However, most of the potential interest of tracking data depends on the ability to develop suitable modelling strategies to analyze trajectories from discrete recorded positions. A serious modelling challenge is to infer the evolution of the true position and the associated spatio-temporal distribution of behavioural states using discrete, error-prone and incomplete observations. In this paper, a Bayesian Hierarchical Model (HBM) using Hidden Markov Process (HMP) is proposed as a template for analyzing fishing boats trajectories based on data available from satellite-based vessel monitoring systems (VMS). The analysis seeks to enhance the definition of the fishing pressure exerted on fish stocks, by discriminating between the different behavioural states of a fishing trip, and also by quantifying the relative importance of each of these states during a fishing trip. The HBM approach is tested to analyse the behaviour of pelagic trawlers in the Bay of Biscay. A hidden Markov chain with a regular discrete time step is used to model transitions between successive behavioural states (e.g., fishing, steaming, stopping (at Port or at sea)) of each vessel. The parameters of the movement process (speed and turning angles) are defined conditionally upon the behavioural states. Bayesian methods are used to integrate the available data (typically VMS position recorded at discrete time) and to draw inferences on any unknown parameters of the model. The model is first tested on simulated data with different parameters structures. Results provide insights on the potential of HBM with HMP to analyze VMS data. They show that if VMS positions are recorded synchronously with the instants at which the process switch from one behavioural state to another, the estimation method provides unbiased and precise inferences on behavioural states and on associated movement parameters. However, if the observations are not gathered with a sufficiently high frequency, the performance of the estimation method could be drastically impacted when the discrete observations are not synchronous with the switching instants. The model is then applied to real pathways to estimate variables of interest such as the number of operations per trip, time and distance spent fishing or travelling.     

A model of fishing vessel accident probability

Problem: Commercial fishing is one of the least safe occupations. Method: The researchers develop a fishing vessel accident probability model for fishing areas off the northeastern United States using logit regression and daily data from 1981 to 1993. Results: The results indicate that fishing vessel accident probability declined over the study period. Higher wind speed is associated with greater accident probability. Medium-size vessels have the highest accident probability, while small vessels have the lowest. Within the study region, accident probability is lower in the southwestern section than in the northeastern section. Accidents are likely to occur closer to shore than offshore. Accident probability is lowest in spring. Impact on Industry: The probability model is an important building block in development and quantitative assessment of management mechanisms related to safety in the commercial fishing industry.     

Losses to the environment from the multilevel cycle of anthropogenic lead

As a component of a multi-level study of the anthropogenic lead cycle for year 2000 (52 countries, 8 regions, and the planet), we have estimated the lead flows in seven emission streams: tailings, slag, fabrication and manufacturing, dissipation from use, hibernation, landfilling, and dispersion following product discard. For every 1 kg of lead put into end use, 0.5 kg is lost to the environment, largely due to landfilling and dissipation from use. From the standpoint of the receiving media, 1/3 of the losses are to uncontained solids on land and 48% of the losses are to containment facilities on land. On a country basis, the largest losses occur in the United States and China, which between them are responsible for about 32% of total global lead losses. On a per capita basis, the highest lead losses occurred in the United Kingdom, Belgium-Luxembourg, and Ireland.     

Interactions between cod, herring and sprat in the changing environment of the Baltic Sea: A dynamic model analysis

The interactions between cod (Gadus morhua), herring (Clupea harengus) and sprat (Sprattus sprattus) in the Central Baltic Sea were examined with a simple dynamic model, an alternative to more complicated and data-demanding multispecies and ecosystem models. The main aims of the study were to compare the effect of alternative structures on the model output and examine the control relationships in the fish assemblage under different environmental conditions. The effect of environmental conditions was modelled using a stock-recruitment equation for cod incorporating an environmental index. The model output was especially sensitive to the functional response in predation by cod on herring and sprat. The type II functional response led to a collapse of the clupeid stocks when cod was abundant, while the type III response produced more realistic stock dynamics. According to the simulations, an abundant cod stock was able to keep the sprat stock at a low level, while the herring stock was less affected and benefited from the decreased density of sprat. Simulation of different fishing scenarios indicated that reducing fishing mortality to the level currently advised by ICES would allow the recovery of the cod stock even in unfavourable environmental conditions.     

Optimization of an artificial neural network for identifying fishing set positions from VMS data: An example from the Peruvian anchovy purse seine fishery

The spatial behavior of numerous fishing fleets is nowadays well documented thanks to satellite Vessel Monitoring Systems (VMS). Vessel positions are recorded on a frequent and regular basis which opens promising perspectives for improving fishing effort estimation and management. However, no specific information is provided on whether the vessel is fishing or not. To answer that question, existing works on VMS data usually apply simple criteria (e.g. threshold on speed). Those simple criteria generally focus in detecting true positives (a true fishing set detected as a fishing set); conversely, estimation errors are given no attention. For our case study, the Peruvian anchovy fishery, those criteria overestimate the total number of fishing sets by 182%. To overcome this problem an artificial neural network (ANN) approach is presented here. In order to set both the optimal parameterization and use “rules” for this ANN, we perform an extensive sensitivity analysis on the optimization of (1) the internal structure and training algorithm of the ANN and (2) the “rules” used for choosing both the relative size and the composition of the databases (DBs) used for training and inferring with the ANN. The “optimized” ANN greatly improves the estimates of the number and location of fishing events. For our case study, ANN reduces the total estimation error on the number of fishing sets to 1% (in average) and obtains 76% of true positives. This spatially explicit information on effort, provided with error estimation, should greatly reduce misleading interpretations of catch per unit effort and thus significantly improve the adaptive management of fisheries. While fitted on Peruvian anchovy fishery data, this type of neural network approach has wider potential and could be implemented in any fishery relying on both VMS and at-sea observer data. In order to increase the accuracy of the ANN results, we also suggest some criteria for improving sampling design by at-sea observers and VMS data.     

Effect of variable fishing strategy on fisheries under changing effort and pressure: An agent-based model application

An agent-based model was used to evaluate the response of a two-species fish community to fishing boat exploration strategies, namely: boats following high-yield boats (Cartesian); boats fishing at random sites (stochast-random); and boats fishing at least exploited sites (stochast-pressure). At low fishing pressure, the stochast-random mode yielded a high average catch per boat while sustaining fish biomass. At high fishing pressure, the Cartesian mode was more effective. For the Cartesian strategy, fish biomass exhibited four distinct behaviors with increasing number of boats. In the first phase, the fish biomass dropped with increasing number of boats due to a corresponding rise in biomass extraction. Rapid exploitation occurred in the second phase, when two or more boats occupied the same initial area, that led to the faster abandonment of those sites which then underwent biomass recovery. In the third phase, adding more boats resulted in a fluctuating stock biomass, where the combined effects of initial spatial distribution of boats and rapid localization led to either full stock recovery when boats were eventually confined to a single location due to spillovers, or stock extirpation when the entire area became fully occupied. Beyond the third phase, stock extirpation was assured. In order to break the pattern of localization (bandwagon effect), we introduced stochast-random intruders in a Cartesian-dominated fishery. Adding a single intruder changed the patchy-structured stock biomass pattern of a purely Cartesian fishery to a uniformly explored stock biomass pattern because of the additional spatial information provided by the intruder. Consequently, the average catch per boat increased but at the expense of a disproportionate decline in equilibrium biomass.     

Effects of local fisheries and ocean productivity on the northeastern Ionian Sea ecosystem

To better understand the effects of fisheries and ocean productivity on the northeastern Ionian Sea we constructed an Ecopath with Ecosim model with 22 functional groups. Data on biomass, production/biomass, consumption/biomass, and diet for each group were estimated or extrapolated from the literature. Fisheries landings and discards were also included. Temporal trajectories were simulated using Ecosim. The model was fitted with time-series data for the most important groups from 1964 to 2006. Simulations highlighted a decline of top predators and of most of the commercial species since the late 1970s. The model shows that the decline of fish resources was mainly caused by an intensive fishing pressure that occurred in the area until the end of the 1990s and also by changes in primary production that impacted the trajectories of the main functional groups. In particular, simulated changes through time in PP impacted the abundance trends of all the commercial species, showing a cascade-up effect through the ecosystem. The application of Ecopath with Ecosim was a useful tool for understanding the trends of the main functional groups of the northeastern Ionian Sea. The model underlined that management actions are needed to restore and protect target species including marine mammals, pelagic and demersal fishes. In particular, measures to reduce overfishing, illegal fishing activities and to respect existing legislations are in need. Moreover, the adoption of marine protected areas could be an effective management measure to guarantee prey survival and to sustain marine predators.     

Quality standard codes of reference of Jordanian coastal waters of the Gulf of Aqaba,Red Sea

Back in 1992, the Gulf of Aqaba Environmental Action Plan (GAEAP), a collaboration between the Aqaba Region Authority (ARA), Jordan and the World Bank, gave considerable emphasis to the environmental protection of the Gulf of Aqaba [The World Bank. Gulf of Aqaba Environmental Action Plan. Report No. 12244JO (1993).]. The document recommended the establishment of a marine reserve and the long term monitoring of the coastal habitats’ environmental quality. The combination of a dedicated follow up, the collaborative efforts of ARA and the Marine Science Station (MSS), and the founding of the Aqaba Special Economic Zone Authority (ASEZA) have turned the recommendations into reality. A comprehensive monitoring program of the Jordanian coastal habitats commenced in 1999. The first three years of the program were financed by a donation from The Global Environmental Facility (GEF). In return, Jordan has committed itself to the maintenance of the monitoring program as an ongoing tool for sustainable coastal management. The monitoring program includes observations on benthic habitat, fish communities, bottom sediments and seawater quality. This paper focuses on the results of seawater-quality monitoring in the first three years. Records of weather conditions, coastal currents, seawater temperature, transparency, salinity, density, pH, alkalinity, dissolved oxygen, ammonia, nitrate, nitrite, phosphate, silicate, particulate matter, chlorophyll a, zooplankton biomass, total coliform, fecal coliform, hydrocarbons and sedimentation rate have been generated monthly since January 1999 at six coastal stations, and one offshore reference station, in the Jordanian waters of the Gulf of Aqaba. The coastal stations are located at sites with different benthic habitats and are occupied by different human activities. Offshore records of density (thermohaline structure), nutrients and chlorophyll a depicted two well-defined seasons; a nutrient-/chlorophyll a-rich, mixed water winter from December to April and a nutrient-/chlorophyll a-poor, stratified water summer from June to October. Short transition seasons appeared in May and November. The mixing and stratification seasons were also clearly depicted in the coastal waters. Statistical analysis of the three-year data collected at the offshore station revealed no significant inter-annual differences in the upper 125 m of the water column with respect to any of the measured parameters. At coastal stations, the water quality at the two northernmost stations was significantly different in comparison to the upper 125 m at the offshore station and to the other coastal stations, with respect to the two key indicator parameters: inorganic nitrogen and chlorophyll a. The three-year findings of the monitoring program are employed to suggest standard codes of reference for the coastal water quality.     

Science, Policy, Stakeholders, and Fish Consumption Advisories: Developing a Fish Fact Sheet for the Savannah River

In recent years there has been a startling rise in the issuance of fish consumption advisories. Unfortunately, compliance by the public is often low. Low compliance can be due to a number of factors, including confusion over the meaning of advisories, conflicting advisories issued by different agencies, controversies involving health benefits versus the risks from consuming fish, and an unwillingness to act on the advisories because of personal beliefs. In some places, such as along the Savannah River, one state (South Carolina) had issued a consumption advisory while the other (Georgia) had not, although at present, both states now issue consumption advisories for the Savannah River. Herein we report on the development of a fish fact sheet to address the confusing and conflicting information available to the public about consuming fish from the Savannah River. The process involved interviewing fishers to ascertain fishing and consumption patterns, evaluating contaminant levels and exposure pathways, discussing common grounds for the provision of information, and consensus-building among different regulatory agencies (US Environmental Protection Agency, South Carolina Department of Health and Environmental Control, Georgia Department of Natural Resources) and the Department of Energy. Consensus, a key ingredient in solving many different types of “commons” problems, was aided by an outside organization, the Consortium for Risk Evaluation with Stakeholder Participation (CRESP). The initial role for CRESP was to offer scientific data as a basis for groups with different assumptions about risks to reach agreement on a regulatory response action. The process was an example of how credible science can be used to implement management and policies and provide a basis for consensus-building on difficult risk communication issues. The paper provides several lessons for improving the risk process from stakeholder conflicts, through risk assessment, to risk management. It also suggests that consensus-building and risk communication are continuing processes that involve assimilation of new information on contaminants and food-chain processes, state and federal law, public policy, and public response.     

Simulation of socio-ecological impacts: Modeling a fishing village

The interrelationship of society and environment is addressed here through the study of a remote fishing village of 750 people. An interdisciplinary study evaluated demographic, economic, and social aspects of the community, and simulation modeling was used to integrate these societal characteristics with environmental factors.The population of the village had grown gradually until the 1960's, when a decline began. Out-migration correlated with declining fish harvests and with increased communications with urban centers. Fishing had provided the greatest economic opportunity, followed by logging. A survey was conducted to investigate the costs and revenues of village fishermen. Diversification characterized the local fleet, and analysis showed that rates of return on investment in the current year were equal between vessel types.The variable levels and rate parameters of the demographic, economic, and social components of the model were specified through static and time series data. Sensitivity analysis to assess the effects of uncertainty, and validation tests against known historical changes were also conducted. Forecast scenarios identified the development options under several levels of fish abundance and investment. The weight given to ecological versus economic resource management registered disproportionate effects due to the interaction between investment and migration rates and resource stochasticity. This finding argues against a golden mean rule for evaluating policy trade-offs and argues for the importance of using a dynamic, socio-ecological perspective in designing development policies for rural communities.