Performance evaluation of cetacean species distribution models developed using generalized additive models and boosted regression trees

Ecology and Evolution, Journal Year: 2020, Volume and Issue: 10(12), P. 5759 - 5784

Published: May 11, 2020

Abstract Species distribution models (SDMs) are important management tools for highly mobile marine species because they provide spatially and temporally explicit information on animal distribution. Two prevalent modeling frameworks used to develop SDMs generalized additive (GAMs) boosted regression trees (BRTs), but comparative studies have rarely been conducted; most rely presence‐only data; few explored how features such as characteristics affect model performance. Since the majority of BRTs predict habitat suitability, we first compared GAMs that presence/absence response variable. We then results from these suitability density (animals per km 2 ) built with a subset data here previously received extensive validation. both explanatory power (i.e., goodness fit) predictive performance novel dataset) taxonomically diverse suite cetacean using robust set systematic survey (1991–2014) within California Current Ecosystem. Both were successful at describing overall patterns throughout study area considered, when predicting data, exhibited substantially greater than BRTs, likely due different variables fitting algorithms. Our an improved understanding some strengths limitations developed two methods. These can be by modelers developing resource managers tasked spatial determine best technique their question interest.

Language: Английский

---

Where did they not go? Considerations for generating pseudo-absences for telemetry-based habitat models

Movement Ecology, Journal Year: 2021, Volume and Issue: 9(1)

Published: Feb. 17, 2021

Abstract Background Habitat suitability models give insight into the ecological drivers of species distributions and are increasingly common in management conservation planning. Telemetry data can be used habitat to describe where animals were present, however this requires use presence-only modeling approaches or generation ‘pseudo-absences’ simulate locations did not go. To highlight considerations for generating pseudo-absences telemetry-based models, we explored how different methods pseudo-absence affect model performance across species’ movement strategies, types, environments. Methods We built marine terrestrial case studies, Northeast Pacific blue whales ( Balaenoptera musculus ) African elephants Loxodonta africana ). tested four commonly models: (1) background sampling; (2) sampling within a buffer zone around presence locations; (3) correlated random walks beginning at tag release location; (4) reverse last location. using generalised linear mixed additive boosted regression trees. Results found that separation environmental niche space between presences was single most important driver explanatory power predictive skill. This result consistent habitats, two with vastly syndromes, three types. The best-performing method depended on which created greatest separation: elephants. However, despite fact greater performed better according traditional skill metrics, they always produce biologically realistic spatial predictions relative known distributions. Conclusions may positively biased cases sampled from environments dissimilar presences. emphasizes need carefully consider extent domain heterogeneity samples when developing highlights importance scrutinizing ensure fit objectives.

Language: Английский

---

Patterns and drivers of macroalgal ‘blue carbon’ transport and deposition in near-shore coastal environments

The Science of The Total Environment, Journal Year: 2023, Volume and Issue: 890, P. 164430 - 164430

Published: May 27, 2023

The role of macroalgae (seaweed) as a global contributor to carbon drawdown within marine sediments - termed 'blue carbon' remains uncertain and controversial. While studies are needed validate the potential for macroalgal‑carbon sequestration in coastal sediments, fundamental questions regarding fate dislodged macroalgal biomass need be addressed. Evidence suggests may advected deposited other vegetated ecosystems down deep ocean; however, contributions near-shore waters remain uncertain. In this study combination eDNA metabarcoding surficial sediment sampling informed by seabed mapping from different physical environments was used test presence south-eastern Australia, factors influencing patterns transport deposition. DNA products total 68 taxa, representing all major groups (Phaeophyceae, Rhodophyta, Chlorophyta) were successfully detected at 112 locations. These findings confirm exported into suggest donors could both speciose diverse. Modelling suggested that deposition, organic (TOC), influenced complex interactions between several environmental including water depth, grain size, wave orbital velocity, current speed, direction, extent infralittoral zone around depositional areas. Extrapolation optimised model predict spatial deposition TOC across coastline identify potentially important sinks. This builds on recent providing empirical evidence deposits framework predicting distribution sinks informing future surveys aimed determining long-term sediments.

Language: Английский

---

Marine heatwaves redistribute pelagic fishing fleets

Fish and Fisheries, Journal Year: 2024, Volume and Issue: 25(4), P. 602 - 618

Published: April 4, 2024

Abstract Marine heatwaves (MHWs) have measurable impacts on marine ecosystems and reliant fisheries associated communities. However, how MHWs translate to changes in fishing opportunities the displacement of fleets remains poorly understood. Using vessel tracking data from automatic identification system (AIS), we developed distribution models for two pelagic targeting highly migratory species, U.S. Atlantic longline Pacific troll fleets, understand MHW properties (intensity, size, duration) influence core grounds fleet displacement. For both size had largest ground area with northern gaining southern decreasing area. response varied between coasts, as displaced farther regions whereas most shifted farther. Characterizing responses these anomalous conditions can help identify regional vulnerabilities under future extreme events aid supporting climate‐readiness resilience fisheries.

Language: Английский

---

Environmental variables and machine learning models to predict cetacean abundance in the Central-eastern Mediterranean Sea

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: Feb. 14, 2023

Although the Mediterranean Sea is a crucial hotspot in marine biodiversity, it has been threatened by numerous anthropogenic pressures. As flagship species, Cetaceans are exposed to those impacts and global changes. Assessing their conservation status becomes strategic set effective management plans. The aim of this paper understand habitat requirements cetaceans, exploiting advantages machine-learning framework. To end, 28 physical biogeochemical variables were identified as environmental predictors related abundance three odontocete species Northern Ionian (Central-eastern Sea). In fact, models built using sighting data collected for striped dolphins Stenella coeruleoalba, common bottlenose Tursiops truncatus, Risso's Grampus griseus between July 2009 October 2021. Random Forest was suitable machine learning algorithm cetacean estimation. Nitrate, phytoplankton carbon biomass, temperature, salinity most influential predictors, followed latitude, 3D-chlorophyll density. proposed here validated acquired during 2022 study area, confirming good performance strategy. This provides valuable information support decisions measures EU spatial planning context.

Language: Английский

---

Identifying priority habitat for conservation and management of Australian humpback dolphins within a marine protected area

Scientific Reports, Journal Year: 2020, Volume and Issue: 10(1)

Published: Sept. 1, 2020

Abstract Increasing human activity along the coast has amplified extinction risk of inshore delphinids. Informed selection and prioritisation areas for conservation delphinids requires a comprehensive understanding their distribution habitat use. In this study, we applied an ensemble species modelling approach, combining results six algorithms to identify high probability occurrence globally Vulnerable Australian humpback dolphin in northern Ningaloo Marine Park (NMP), north-western Australia. Model outputs were based on sighting data collected during systematic, boat-based surveys between 2013 2015, relation various ecogeographic variables. Water depth distance identified as most important variables influencing presence, with dolphins showing preference shallow waters (5–15 m) less than 2 km from coast. Areas (> 0.6) primarily (90%) multiple use where extractive activities are permitted, poorly represented sanctuary (no-take) zones. This spatial mismatch emphasises need reassess future planning marine park management plan reviews NMP. Shallow, coastal here should be considered priority species.

Language: Английский

---

Combining Regional Habitat Selection Models for Large-Scale Prediction: Circumpolar Habitat Selection of Southern Ocean Humpback Whales

Remote Sensing, Journal Year: 2021, Volume and Issue: 13(11), P. 2074 - 2074

Published: May 25, 2021

Machine learning algorithms are often used to model and predict animal habitat selection—the relationships between occurrences characteristics. For broadly distributed species, selection varies among populations regions; thus, it would seem preferable fit region- or population-specific models of for more accurate inference prediction, rather than fitting large-scale using pooled data. However, where the aim is make range-wide predictions, including areas which there no existing data selection, how can regional best be combined? We propose that ensemble approaches commonly combine different a single region reframed, treating as candidate models. By doing so, we incorporate variation when predictive across large ranges. test this approach satellite telemetry from 168 humpback whales five geographic regions in Southern Ocean. Using random forests, fitted relating whale locations, versus background 10 environmental covariates, made circumpolar prediction selection. also models, predictions input features four approaches: an unweighted ensemble, weighted by similarity each cell, stacked generalization, hybrid wherein covariates were new model. tested performance these on independent validation dataset sightings whaling catches. These multiregional resulted with higher naive machine algorithms. This yield animals may show

Language: Английский

---

The Role of Environmental Drivers in Humpback Whale Distribution, Movement and Behavior: A Review

Frontiers in Marine Science, Journal Year: 2021, Volume and Issue: 8

Published: Nov. 10, 2021

Humpback whales, Megaptera novaeangliae , are a highly migratory species exposed to wide range of environmental factors during their lifetime. The spatial and temporal characteristics such play significant role in determining suitable habitats for breeding, feeding resting. existing studies the relationship between oceanic conditions humpback whale ecology provide basis understanding impacts on this species. Here we have determined most relevant drivers identified peer-reviewed literature published over last four decades, assessed methods used identify relationships. A total 148 were extracted through an online search. These combined estimated 105,000 observations 1,216 accumulated study years investigating whales both Northern Southern Hemispheres. Studies focusing areas found preferences upwelling, high chlorophyll-a concentration frontal with changes temperature, depth currents, where prey can be concentration. Preferred calving grounds as shallow, warm slow water movement aid survival calves. few migration routes shallow waters close shorelines moderate temperature Extracting information influence key behavioral modes important conservation, particularly regard expected under climate change.

Language: Английский

---

Divergent responses of highly migratory species to climate change in the California Current

Diversity and Distributions, Journal Year: 2023, Volume and Issue: 30(2)

Published: Dec. 8, 2023

Abstract Aim Marine biodiversity faces unprecedented threats from anthropogenic climate change. Ecosystem responses to change have exhibited substantial variability in the direction and magnitude of redistribution, posing challenges for developing effective climate‐adaptive marine management strategies. Location The California Current (CCE), USA. Methods We project suitable habitat 10 highly migratory species System using an ensemble three high‐resolution (~10 km) downscaled ocean projections under Representative Concentration Pathway 8.5 (RCP8.5). Spanning period 1980 2100, our analysis focuses on assessing distance distributional shifts, as well changes core area each species. Results Our findings reveal a divergent response among impacts. Specifically, four were projected undergo significant poleward shifts exceeding 100 km, gain (~7%–60%) Conversely, six shift towards coast, resulting loss ranging 10% 66% by end century. These could typically be characterized mode thermoregulation (i.e. ectotherm vs. endotherm) species' affiliations with cool productive upwelled waters that are characteristic region. Furthermore, study highlights increase niche overlap between protected those targeted fisheries, which may lead increased human interaction events Main Conclusions By providing valuable distribution projections, research contributes understanding effects offers critical insight support climate‐ready fished

Language: Английский

---

Predicting plant distribution on the River Nile islands in Egypt using machine learning algorithms

International Journal of Environmental Science and Technology, Journal Year: 2025, Volume and Issue: unknown

Published: March 2, 2025

Language: Английский

---