Data-poor Methods Research Articles

This is what we know: Assessing the stock status of the data-poor skipjack tuna (Katsuwonus pelamis) fishery in the South China Sea

Skipjack tuna (Katsuwonus pelamis) is a highly migratory species of significant value in global marine fisheries. The South China Sea (SCS) is the largest marginal sea in the northwestern Pacific Ocean, and many tuna and tuna-like species occur within it. Although a recent Western and Central Pacific Fisheries Commission report identified skipjack tuna in the western and central Pacific to be sustainably fished, the exploitation status of skipjack tuna in the SCS is still unclear, due to limited data. We apply two data-poor methods, length-based Bayesian biomass estimation (LBB) and length-based spawning potential ratio (LBSPR), to assess the status of skipjack tuna stock in the SCS. We use electronic length frequency analysis to estimate von Bertalanffy growth parameters (asymptotic lengthLinfand growth coefficientK) to serve as priors for LBB and LBSPR estimation. Estimates are calculated for SCS skipjack tunaLinf(68.3 cm),K(0.325 year−1), natural mortality (0.49 year−1) and mean fork length at 50% sexual maturity (36.7 cm). LBB analysis reveals the estimated relative stock size (0.29) indicates that the SCS skipjack tuna stock maybe heavily overfished. LBSPR analysis reveals the estimated spawning potential ratio for this fishery to be 3%, which is significantly below the limit reference point of 20%. Accordingly, for the SCS skipjack tuna stock we identify a need to reduce fishing mortality by controlling fishing effort and increasing catchable size.

Performance Comparison of Three Data-Poor Methods With Various Types of Data on Assessing Southern Atlantic Albacore Fishery

In the world, more than 80% of the fisheries by numbers and about half of the catches have not been formally analyzed and evaluated due to limited data. It has led to the fast growth of data-poor evaluation methods. There have been various studies carried out on the comparative performance of data-poor and data-moderate methods in evaluating fishery exploitation status. However, most studies to date have focused on coastal fish stocks with simple data sources. It is important to pay attention to high sea fisheries because they are exploited by multiple countries, fishing gears and data may be divrsified and inconsistent. Furthermore, a comparison of the performance of catch-based, length-based, and abundance-based methods to estimate fishery status is needed. This study is the first attempt to apply catch-based, length-based, and abundance-based data-poor methods to stock assessment for an oceanic tuna fishery and to compare the performance with a data-moderate model. Results showed that the three data-poor methods with various types of data did not produce an entirely consistent stock status of the southern Atlantic albacore (Thunnus alalunga) fishery in 2005, as the estimated B2005/BMSY ranged from 0.688 to 1.3 and F2005/FMSY ranged from 0.708 to 1.6. The Monte Carlo Catch maximum sustainable yield model (CMSY) produced a similar time series of B/BMSY and F/FMSY and stock status (recovering) to the Bayesian state-space Schaefer model (BSM). The abundance-based method (AMSY) gave the most conservative condition (overfished) of this fishery. Sensitivity analysis showed the results of the length-based Bayesian biomass estimation method (LBB) are sensitive to Linf settings, and the results with higher Linf were similar to those of other models. However, the LBB results with setting Linf at lower levels produced more optimistic conditions (healthy). Our results highlight that attention should be paid to the settings of model parameter priors and different trends implied in various types of data when using these data-poor methods.

Effect of a precautionary management measure on the vulnerability and ecological risk of elasmobranchs captured as target fisheries

The impact of fisheries on elasmobranch populations has been widely documented; however, baseline information on shark and batoid management remains limited, particularly in tropical areas and in developing countries. The use of precautionary measures and data-poor methods have been recommended as valuable tools to guide the management and conservation of marine taxa when information is scarce and the effect of fisheries on populations is evident. A Productivity Susceptibility Analysis (PSA) and intrinsic rebound potential (r) estimates were used to evaluate the vulnerability and resilience of elasmobranchs before and after a precautionary measure (closure of the fishery) was implemented in the Archipelago of San Andrés, Providencia and Santa Catalina, Colombia (Seaflower Biosphere Reserve). We identified locally large information gaps, specifically on the demographic and life history parameters of the 15 elasmobranch species (13 sharks and 2 rays) evaluated for the Archipelago. Including fishing effects, nine large commercial shark species were identified as the most vulnerable and with the lowest intrinsic rebound potential, while the least vulnerable species was the whale shark (Rhincodon typus), which presented the third highest value of r2M. Once the fishery was closed, the vulnerability of large sharks decreased greatly. Our findings highlight the importance of: involving communities to achieve effective management processes; implementing precautionary measures for high impact and targeted fisheries; and using valuable data-poor tools for the study of populations as an alternative for evaluating and suggesting management measures. Finally, based in PSA and intrinsic rebound potential results, it is suggested to maintain the fishing ban in consensus with the fishermen, and to evaluate economic alternatives already being generated in the area such as recreational diving and ecological tourism, among others.

Estimating blue marlin (Makaira nigricans) sustainable yield in the Indian Ocean using a data-poor approach

Blue marlin (Makaira nigricans) is a common bycatch species in the global tuna longline fishery. In this study, we applied a common data-poor approach, i.e., depletion-corrected average catch (DCAC) to assess stock status of the Indian Ocean blue marlin. Sustainable yield (Ysust), one reference point in this case, was estimated, and its uncertainty was integrated by using Monte Carlo simulation. The results revealed the estimate of Ysust by DCAC was lower than MSY of 11,926 t by BSP-SS and is also lower than the provisional reference point of 11,704 t by the management proposal. DCAC is reliable for blue marlin in driving precautionary management quantity based on the CPUE of Japan (1980–2015). This study also implies that DCAC could be applied to other billfish stocks and uncertainty be estimated for sustainable yield. However, data-poor methods could be adjusted with precautionary approaches.

Integrating reproductive ecology, early life dynamics and mesoscale oceanography to improve albacore tuna assessment in the Western Mediterranean

Albacore tuna is a pelagic top predator of high ecological and commercial relevance worldwide. Despite the importance of this species, there is only limited knowledge of its life cycle, reproductive dynamics and early life ecology. In the Mediterranean, where the assessment is based on data-poor methods, the status of the albacore population remains uncertain. Here we combine information from commercial fisheries, ichthyoplankton surveys and oceanographic data in the Western Mediterranean Sea to identify the geographical location of the main spawning areas and larval habitats. We applied this information to develop a habitat-corrected larval abundance index providing information on early life dynamics. The larval index was compared with fisheries-dependent indices used for the stock assessment of the Mediterranean albacore. The results allowed the identification of the main spawning grounds, located around the Balearic Sea, and the role of oceanographic conditions when studying larval abundances. The larval abundance index shows a significant negative trend during the last decade, indicating a potential reduction in the spawning stock biomass, which is consistent with the trend observed from the fisheries-dependent indices. We discuss how the systematic integration of information from ichthyoplankton surveys and hydrography to improve assessment is an example of operational fisheries oceanography in practice.

Assessing the vulnerability of demersal elasmobranchs to a data-poor shrimp trawl fishery in Costa Rica, Eastern Tropical Pacific

The impact of shrimp trawl fisheries on slow-growing demersal sharks, skates and rays has been widely documented. Yet, a lack of catch records and biological information has hindered improvements in elasmobranch bycatch management, particularly in tropical regions. When information is scarce, data–poor methods can be valuable tools to guide the management and conservation of vulnerable marine taxa. Here we combined an Ecological Risk Assessment commonly used in data-deficient fisheries (Productivity Susceptibility Analysis or PSA) with a spatial analysis (Hotspot Analysis) to (i) identify which elasmobranch species are most vulnerable to the Costa Rican shrimp trawl fishery, and (ii) locate areas and seasons with high concentrations of vulnerable elasmobranchs. According to the results of our PSA, the scalloped hammerhead (Sphyrna lewini), prickly shark (Echinorhinus cookei), and long–tail stingray (Hypanus longus) were the most vulnerable species, while the least vulnerable species were small batoids of the Urotrygonidae family. We identified large information gaps, namely demographic parameter estimates and long–term bycatch trends. Spatial aggregations of vulnerable elasmobranchs occurred near highly productive estuaries and coastal habitats, especially at depths of 50–100m. Based on our findings, PSA and Hotspot analyses can be powerful tools to identify spatial areas where trawl fishing grounds overlap with the habitat of vulnerable bycatch species, and thus can be used to design spatial trawling closures that protect species at risk. We therefore conclude that our methodological approach may aid in the implementation of ecosystem–based fisheries management in data–poor situations.

Predicting reference points and associated uncertainty from life histories for risk and status assessment

To assess status of fish populations and the risks of overexploitation, management bodies compare fishing mortality rates and abundance estimates with reference points (RP). Generic, “data-poor” methods for estimating RP are garnering attention because they are faster and cheaper to implement than those based on extensive life history data. Yet data-poor RP are subject to many unquantified uncertainties. Here, we predict fishing mortality RP based on five levels of increasingly comprehensive data, to quantify effects of parameter and structural uncertainty on RP. Level I RP (least data) are estimated solely from species' maximum size and generic life history relationships, while level V RP (most data) are estimated from population-specific growth and maturity data. By estimating RP at all five data levels, for each of 12 North Sea populations, we demonstrate marked changes in the median RP values, and to a lesser extent uncertainty, when growth parameters come from data rather than life history relationships. As a simple rule, halving the median level I RP gives almost 90% probability that a level V median RP is not exceeded. RP and uncertainty were substantially affected by assumed gear selectivity; plausible changes in selectivity had a greater effect on RP than adding level V data. Calculations of RP using data for successive individual years from 1984 to 2014 showed that the median RP based on data for any given year would often fall outside the range of uncertainty for RP based on data from earlier or later years. This highlighted the benefits of frequent RP updates when suitable data are available. Our approach provides a quantitative method to inform risk-based management and decisions about acceptable targets for data collection and quality. Ultimately, however, the utility and extent of adoption of data-poor methods for estimating RP will depend on the risk aversion of managers.

Current methods for setting catch limits for data-limited fish stocks in the United States

This paper examines how the requirement for annual catch limits (ACLs) has been implemented for data-limited stocks in all federally-managed fisheries in the United States. The legal mandate to establish ACLs in the U.S. has spurred substantial scientific advances, including the development and adoption of at least 16 methods for establishing catch limits for data-limited fisheries. This study analyzed the assessment methods that form the basis of ACLs, those which determine the overfishing limits (OFLs) and the acceptable biological catches (ABCs). Nationally, 30% (150) of OFLs/ABCs are currently calculated using conventional data-rich assessment methods, 11% (59) using data-moderate methods, and 59% (295) using data-poor approaches. There is substantial variation in the proportion of stocks that are currently managed with data-rich versus data-limited methods across regions, and there are clear geographical patterns in the types and diversity of methods being utilized to calculate OFLs/ABCs. Data-poor methods are the most commonly used OFL/ABC-setting methods in the U.S., particularly in the Southeast, Atlantic highly migratory species (HMS), Pacific, and Western Pacific regions. The Southeast and Atlantic HMS regions use some form of catch scalar or an ABC of zero landings for each data-limited stock. The Pacific and North Pacific regions currently employ a higher diversity of data-limited methods than any other region; these include both data-moderate methods and data-poor methods. Regional disparities in data-limited method development and implementation are attributed to regional differences in the number of stocks being managed, the data types and lengths of the time series available, and the resources dedicated to data processing and stock assessment. Recommendations for improving management of data-limited stocks include establishing a complete inventory of all available data for each managed stock, dedicating resources and expertise to data-limited method development and evaluation, and developing a more streamlined assessment process to handle the expanded volume of stocks requiring ACLs.

Testing the robustness of data-poor assessment methods to uncertainty in catch and biology: a retrospective approach

Abstract The quality and quantity of data affect the reliability of all stock assessments. Over time, we expect data to improve and assessment predictions to become more reliable. There is a potential for strong bias in estimates of sustainable yield if the available data are not a good representation of stock dynamics, particularly for catch-based data-poor methods that rely on limited information and assumptions about stock status. We retrospectively investigated the interaction of data quantity and quality through time using the “real-world” data for a stock as it progressed from data-poor to data-rich. For this analysis, we chose a currently data-rich and overfished stock with historical assessments representing both a data-poor and data-moderate state, the canary rockfish (Sebastes pinniger). We asked how changes in the catch history and biological parameters over time affected the estimates of sustainable yield and the overfishing limit (OFL) predicted by two data-poor assessment tools, depletion-corrected average catch (DCAC) and depletion-based stock reduction analysis (DB-SRA). We found that both of these methods underestimated the “true” OFL in simulations with catch error alone. While there was slightly less bias for DB-SRA than DCAC, increasing error in the catch led to a more rapid increase in the variance of the DB-SRA harvest limit (HL). Our retrospective analysis showed that the expectation for a more accurate HL estimate between the data-poor and data-moderate canary rockfish assessments does not come from an increase in the quantity or quality of the catch data alone; a decrease in the quality of the biological data between assessments had the greatest impact. By evaluating these methods with historical data, our retrospective approach highlighted the impact of change in data quality and quantity on HL estimates for a long-lived rockfish, and could be used to define the amount and type of error included in simulation studies that further evaluate data-poor methods.

A new role for effort dynamics in the theory of harvested populations and data-poor stock assessment

Research shows that population status can be predicted using catch data, but there is little justification for why these predictions work or how they account for changes in fisheries management. We demonstrate that biomass can be reconstructed from catch data whenever fishing mortality follows predictable dynamics over time (called “effort dynamics”), and we develop a state-space catch only model (SSCOM) for this purpose. We use theoretical arguments and simulation modeling to demonstrate that SSCOM can, in some cases, estimate population status from catch data. Next, we use meta-analysis to estimate effort dynamics for US West Coast groundfishes before and after fisheries management changes in the mid-1990s. We apply the SSCOM using meta-analytic results to data for eight assessed species and compare results with stock assessment and data-poor methods. Results indicate general agreement among all three methods. We conclude that effort dynamics provides a theoretical basis for using catch data to reconstruct biomass and has potential for conducting data-poor assessments. However, we still recommend that index and compositional data be collected to allow application of data-rich methods.

Asymptotic size and natural mortality of long-lived fish for data poor stock assessments

An assessment of the California sheephead (Semicossyphus pulcher) was conducted in 2004 despite the lack of critical data to estimate life history parameters, as often occurs for other species in need of assessment. Among other parameters, natural mortality, M, and the parameters of the von Bertalanffy growth equation were estimated using data poor methods. We assess the difference between those results and the results of applying a more recent set of data poor methods to the same data. We use a Bayesian model with only sheephead length data to estimate asymptotic size; our estimate is similar to previous work using the von Bertalanffy growth equation but includes statements about uncertainty in asymptotic size. We use weight-based methods for calculating age-specific mortality and compare the estimate to the constant of 0.2 used in the sheephead assessment. Finally, since the sheephead is a long-lived fish, we determine the age at which fish must recruit to the fishery in order for a constant M to be appropriate. Our estimate of constant mortality stabilizes if the fish recruit after age two.

Choosing a fishery’s governance structure using data poor methods

Multi-species fisheries are complex to manage and the ability to develop an appropriate governance structure is often seriously impeded because trading between sustainability objectives at the species level, economic objectives at the fleet level, and social objectives at the community scale, is complex. Many of these fisheries also tend to have a mix of information, with stock assessments available for some species and almost no information on other species. The fleets themselves comprise fishers from small family enterprises to large vertically integrated businesses. The Queensland trawl fishery in Australia is used as a case study for this kind of fishery. It has the added complexity that a large part of the fishery is within a World Heritage Area, the Great Barrier Reef Marine Park, which is managed by an agency of the Australian Commonwealth Government whereas the fishery itself is managed by the Queensland State Government. A stakeholder elicitation process was used to develop social, governance, economic and ecological objectives, and then weight the relative importance of these. An expert group was used to develop different governance strawmen (or management strategies) and these were assessed by a group of industry stakeholders and experts using multi-criteria decision analysis techniques against the different objectives. One strawman clearly provided the overall set of outcomes given the multiple objectives, but was not optimal in terms of every objective, demonstrating that even the best strawman may be less than perfect. © 2012.

From clinical narratives to ICD codes: automatic text categorization for medico-economic encoding

Summary We report on the design and preliminary evaluation of a system to help assignment ICD (International Classification of Disease) codes to clinical narratives. The objective of the project is to deliver an operational system to assist professional encoders of the University Hospital of Geneva in their coding task. We combine a set of machine learning and data-poor methods to generate a single automatic text categorizer, which returns a ranked list of ICD codes for any given output. In its current settings, the system is accepting three different textual fields as input: anamnesis, diagnosis and prescription fields. The combined ranking system currently obtains a precision of 74% at high ranks and a recall of about 63% for the top twenty returned codes. Although fairly promising, the impact of the tool on the institution coding efforts as well as on the billing figures (DRG: Diagnosis Related Groups) must now be evaluated.