Dataset Information

Gbm.auto: A software tool to simplify spatial modelling and Marine Protected Area planning.

ABSTRACT:

Boosted regression trees. excellent for data-poor spatial management but hard to use

Marine resource managers and scientists often advocate spatial approaches to manage data-poor species. Existing spatial prediction and management techniques are either insufficiently robust, struggle with sparse input data, or make suboptimal use of multiple explanatory variables. Boosted Regression Trees feature excellent performance and are well suited to modelling the distribution of data-limited species, but are extremely complicated and time-consuming to learn and use, hindering access for a wide potential user base and therefore limiting uptake and usage.

Brts automated and simplified for accessible general use with rich feature set

We have built a software suite in R which integrates pre-existing functions with new tailor-made functions to automate the processing and predictive mapping of species abundance data: by automating and greatly simplifying Boosted Regression Tree spatial modelling, the gbm.auto R package suite makes this powerful statistical modelling technique more accessible to potential users in the ecological and modelling communities. The package and its documentation allow the user to generate maps of predicted abundance, visualise the representativeness of those abundance maps and to plot the relative influence of explanatory variables and their relationship to the response variables. Databases of the processed model objects and a report explaining all the steps taken within the model are also generated. The package includes a previously unavailable Decision Support Tool which combines estimated escapement biomass (the percentage of an exploited population which must be retained each year to conserve it) with the predicted abundance maps to generate maps showing the location and size of habitat that should be protected to conserve the target stocks (candidate MPAs), based on stakeholder priorities, such as the minimisation of fishing effort displacement.

Gbm.auto for management in various settings

By bridging the gap between advanced statistical methods for species distribution modelling and conservation science, management and policy, these tools can allow improved spatial abundance predictions, and therefore better management, decision-making, and conservation. Although this package was built to support spatial management of a data-limited marine elasmobranch fishery, it should be equally applicable to spatial abundance modelling, area protection, and stakeholder engagement in various scenarios.

SUBMITTER: Dedman S

PROVIDER: S-EPMC5720763 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Gbm.auto: A software tool to simplify spatial modelling and Marine Protected Area planning.

Dedman Simon S Officer Rick R Clarke Maurice M Reid David G DG Brophy Deirdre D

PloS one 20171207 12

<h4>Boosted regression trees. excellent for data-poor spatial management but hard to use</h4>Marine resource managers and scientists often advocate spatial approaches to manage data-poor species. Existing spatial prediction and management techniques are either insufficiently robust, struggle with sparse input data, or make suboptimal use of multiple explanatory variables. Boosted Regression Trees feature excellent performance and are well suited to modelling the distribution of data-limited spec ...[more]

PMID: 29216310

Similar Datasets

Project description:BackgroundImplementation of control of parasitic diseases requires accurate, contemporary maps that provide intervention recommendations at policy-relevant spatial scales. To guide control of soil transmitted helminths (STHs), maps are required of the combined prevalence of infection, indicating where this prevalence exceeds an intervention threshold of 20%. Here we present a new approach for mapping the observed prevalence of STHs, using the example of Kenya in 2009.Methods and findingsObserved prevalence data for hookworm, Ascaris lumbricoides and Trichuris trichiura were assembled for 106,370 individuals from 945 cross-sectional surveys undertaken between 1974 and 2009. Ecological and climatic covariates were extracted from high-resolution satellite data and matched to survey locations. Bayesian space-time geostatistical models were developed for each species, and were used to interpolate the probability that infection prevalence exceeded the 20% threshold across the country for both 1989 and 2009. Maps for each species were integrated to estimate combined STH prevalence using the law of total probability and incorporating a correction factor to adjust for associations between species. Population census data were combined with risk models and projected to estimate the population at risk and requiring treatment in 2009. In most areas for 2009, there was high certainty that endemicity was below the 20% threshold, with areas of endemicity ≥ 20% located around the shores of Lake Victoria and on the coast. Comparison of the predicted distributions for 1989 and 2009 show how observed STH prevalence has gradually decreased over time. The model estimated that a total of 2.8 million school-age children live in districts which warrant mass treatment.ConclusionsBayesian space-time geostatistical models can be used to reliably estimate the combined observed prevalence of STH and suggest that a quarter of Kenya's school-aged children live in areas of high prevalence and warrant mass treatment. As control is successful in reducing infection levels, updated models can be used to refine decision making in helminth control.

Project description:Assessments of the effectiveness of marine protected areas (MPAs) usually assume that fishing patterns change exclusively due to the implementation of an MPA. This assumption increases the risk of erroneous conclusions in assessing marine zoning, and consequently counter-productive management actions. Accordingly, it is important to understand how fishers respond to a combination of the implementation of no-take zones, and various climatic and human drivers of change. Those adaptive responses could influence the interpretation of assessment of no-take zone effectiveness, yet few studies have examined these aspects. Indeed, such analysis is often unfeasible in developing countries, due to the dominance of data-poor fisheries, which precludes full examination of the social-ecological outcomes of MPAs. In the Galapagos Marine Reserve (Ecuador), however, the availability of long-term spatially explicit fishery monitoring data (1997-2011) for the spiny lobster fishery allows such an analysis. Accordingly, we evaluated how the spatiotemporal allocation of fishing effort in this multiple-use MPA was affected by the interaction of diverse climatic and human drivers, before and after implementation of no-take zones. Geographic information system modelling techniques were used in combination with boosted regression models to identify how these drivers influenced fishers' behavior. Our results show that the boom-and-bust exploitation of the sea cucumber fishery and the global financial crisis 2007-09, rather than no-take zone implementation, were the most important drivers affecting the distribution of fishing effort across the archipelago. Both drivers triggered substantial macro-scale changes in fishing effort dynamics, which in turn altered the micro-scale dynamics of fishing patterns. Fishers' adaptive responses were identified, and their management implications analyzed. This leads to recommendations for more effective marine and fishery management in the Galapagos, based on improved assessment of the effectiveness of no-take zones.

Project description:BackgroundMarine soundscape is the aggregation of sound sources known as geophony, biophony, and anthrophony. The soundscape analysis, in terms of collection and analysis of acoustic signals, has been proposed as a tool to evaluate the specific features of ecological assemblages and to estimate their acoustic variability over space and time. This study aimed to characterise the Capo Caccia-Isola Piana Marine Protected Area (Italy, Western Mediterranean Sea) soundscape over short temporal (few days) and spatial scales (few km) and to quantify the main anthropogenic and biological components, with a focus on fish biophonies.MethodsWithin the MPA, three sites were chosen each in a different protection zone (A for the integral protection, B as the partial protection, and C as the general protection). In each site, two underwater autonomous acoustic recorders were deployed in July 2020 at a depth of about 10 m on rocky bottoms. To characterise the contribution of both biophonies and anthrophonies, sea ambient noise (SAN) levels were measured as sound pressure level (SPL dB re: 1 μ Pa-rms) at eight 1/3 octave bands, centred from 125 Hz to 16 kHz, and biological and anthropogenic sounds were noted. Fish sounds were classified and counted following a catalogue of known fish sounds from the Mediterranean Sea based on the acoustic characteristic of sound types. A contemporary fish visual census had been carried out at the test sites.ResultsSPL were different by site, time (day vs. night), and hour. SPLs bands centred at 125, 250, and 500 Hz were significantly higher in the daytime, due to the high number of boats per minute whose noise dominated the soundscapes. The loudest man-made noise was found in the A zone, followed by the B and the C zone, confirming that MPA current regulations do not provide protection from acoustic pollution. The dominant biological components of the MPA soundscape were the impulsive sounds generated by some invertebrates, snapping shrimps and fish. The vast majority of fish sounds were recorded at the MPA site characterized by the highest sound richness, abundance, and Shannon-Wiener index, coherently with the results of a fish visual census. Moreover, the acoustic monitoring detected a sound associated with a cryptic species (Ophidion spp.) never reported in the study area before, further demonstrating the usefulness of passive acoustic monitoring as a complementary technique to species census. This study provides baseline data to detect future changes of the marine soundscapes and some suggestions to reduce the impact of noise on marine biodiversity.

Dataset Information

Gbm.auto: A software tool to simplify spatial modelling and Marine Protected Area planning.

Boosted regression trees. excellent for data-poor spatial management but hard to use

Brts automated and simplified for accessible general use with rich feature set

Gbm.auto for management in various settings

Publications

Gbm.auto: A software tool to simplify spatial modelling and Marine Protected Area planning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets