Drivers of Spatial Distributions of Basking Shark (Cetorhinus maximus) in the Southwest Pacific

Frontiers in Marine Science, Год журнала: 2021, Номер 8

Опубликована: Апрель 26, 2021

Basking sharks ( Cetorhinus maximus ) were widely reported throughout New Zealand waters. Once commonly observed, and sometimes in large numbers, basking are now infrequently reported. shark observations known to be highly variable across years, their distribution occurrence have been shown influenced by environmental predictors such as thermal fronts, chl- a concentration, the abundance of prey (zooplankton). Little is South Pacific more information on distribution, habitat use, migratory patterns required better understand species’ regional ecology. Here, we used bootstrapped Habitat Suitability Models [HSM, ensembled from Boosted Regression Tree (BRT) Random Forest (RF) models] determine drivers predict suitability estimated uncertainty for first time. High−resolution (1 km 2 grid resolution) biotic data, including inferred species, all available records Zealand’s Exclusive Economic Zone (EEZ) included ensemble HSMs. The most influential driver modeled was vertical flux particulate organic matter at seabed, which may indicate higher levels primary production surface ocean density mesopelagic zone seafloor. BRT RF models had good predictive power (AUC TSS > 0.7) both performed similarly with low variability model fit metrics. Areas high east west coasts Island, Puysegur Ridge, Auckland Island slope. outputs produced here could incorporated into future management framework assessing threat conservation needs (e.g., spatially explicit risk assessment) this regionally protected well providing guidance research efforts areas interest sampling).

Язык: Английский

---

Evaluation of the full set of habitat suitability models for vulnerable marine ecosystem indicator taxa in the South Pacific high seas

Fisheries Management and Ecology, Год журнала: 2024, Номер 31(4)

Опубликована: Март 16, 2024

Abstract In the high seas, regional fishery management organisations are required to implement measures prevent significant adverse impacts on vulnerable marine ecosystems (VMEs). Our objectives were develop habitat suitability models for use in spatial of bottom fisheries South Pacific and evaluate these existing using independent data from high‐quality seafloor imagery. Presence‐only seven VME indictor taxa developed complement previous modelling. Evaluation withheld indicated mean True Skill Statistic scores 0.44–0.64. Most performed adequately when assessed with taxon presence absence but poor surrogates abundance. We therefore advocate caution presence‐only call more systematically collected abundance models.

Язык: Английский

---

Accounting for uncertainty in marine ecosystem service predictions for spatial prioritisation

Diversity and Distributions, Год журнала: 2024, Номер 30(6)

Опубликована: Март 29, 2024

Abstract Aim Spatial assessments of Ecosystem Services (ES) are increasingly used in environmental management, but rarely provide information on the prediction accuracy. Uncertainty estimates essential to confidence quality and credibility ES for informed decision making. In marine environments, need uncertainty is unparalleled as they data scarce, poorly (spatially) defined, with complex interconnectivity seascapes. This study illustrates associated a principle‐based method modelling by accounting model variability, coverage thresholds parameters. Location Tauranga, New Zealand. Methods A sensitivity analysis was applied models bivalves ( Austrovenus stutchburyi Paphies australis ) their contribution Food provision , Water regulation Nitrogen removal Sediment stabilisation . from were compared against baseline predictions. patterns analysed individual through bi‐plots multiple spatial prioritisation using Zonation. Results Our showed spatially explicit differences between species. had highest maximum (>5 points) also largest area high certainty conditions. Zonation conducted conservative values overall robust outcomes top 30% area, important nuances shifts 5% 10% areas that allowed consistently better representation when uncertainty. Main Conclusions The combination biplots tools planning focus value can thereby help reduce risk aid decision‐making at acceptable levels. type urgently needed likewise extends other environments improve transparency.

Язык: Английский

---

Testing spatial transferability of species distribution models reveals differing habitat preferences for an endangered delphinid (Cephalorhynchus hectori) in Aotearoa, New Zealand

Ecology and Evolution, Год журнала: 2024, Номер 14(7)

Опубликована: Июль 1, 2024

Abstract Species distribution models (SDMs) can be used to predict distributions in novel times or space (termed transferability) and fill knowledge gaps for areas that are data poor. In conservation, this determine the extent of spatial protection required. To understand how well a model transfers spatially, it needs independently tested, using from habitats. Here, we test transferability SDMs Hector's dolphin ( Cephalorhynchus hectori ), culturally important (taonga) endangered, coastal delphinid, endemic Aotearoa New Zealand. We collected summer three populations 2021 2023. Using Generalised Additive Models, built presence/absence each population validated predictive ability top (with TSS AUC). Then, tested by predicting remaining two populations. showed useful performance within their respective (Banks Peninsula Otago), but when outside models' source data, declined markedly. third area (Timaru) performed poorly, both prediction transferred spatially. When building were combined areas, results mixed. Model interpolation was better Otago, an low density, with higher otherwise The overall poor suggests habitat preferences dolphins vary between areas. For these dolphins, population‐specific should conservation planning. More generally, demonstrate one fits all approach is not always suitable. data‐poor assessment new required, interpreted caution.

Язык: Английский

---

Modelling habitat suitability of the Indo-Pacific humpback dolphin using artificial neural network: The influence of shipping

Ecological Informatics, Год журнала: 2021, Номер 62, С. 101274 - 101274

Опубликована: Март 12, 2021

Язык: Английский

---