Revealing the effects of environmental and spatio-temporal variables on changes in Japanese sardine (Sardinops melanostictus) high abundance fishing grounds based on interpretable machine learning approach

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 11

Published: Jan. 13, 2025

The construction of accurate and interpretable predictive model for high abundance fishing ground is conducive to better sustainable fisheries production carbon reduction. This article used refined statistical maps visualize the spatial temporal patterns catch changes based on 2014-2021 fishery statistics Japanese sardine Sardinops melanostictus in Northwest Pacific Ocean. Three models (XGBoost, LightGBM, CatBoost) two variable importance visualization methods (model built-in (split) SHAP methods) were comparative analysis determine optimal modeling strategies. Results: 1) From 2014 2021, annual showed an overall increasing trend peaked at 220,009.063 tons 2021; total monthly increased then decreased, with a peak 76, 033.4944 (July), was mainly concentrated regions 39.5°-43°N 146.75°-155.75°E; 2) Catboost predicted than LightGBM XGBoost models, highest values accuracy F1-score, 73.8% 75.31%, respectively; 3) ranking model’s method differed significantly from that method, variables increased. Compared informs magnitude direction influence each global local levels. results research help us select construct prediction grounds Ocean, which will provide scientific basis achieve environmental economically development.

Language: Английский

---

Evaluating species richness, turnover, and range shifts under climate change for fluvial fishes in Northeastern and Midwestern USA

Ecological Processes, Journal Year: 2025, Volume and Issue: 14(1)

Published: April 23, 2025

Abstract Background Fluvial fish habitat in the Northeastern and Midwestern U.S. is substantially affected by natural landscape factors anthropogenic stressors, with climate change expected to alter influences exacerbate stressor effects. To conserve fluvial species future, it crucial understand which habitats will be most strongly influenced changing climate, are sensitive change, how changes individual affect entire assemblages. answer these questions, we modeled distributions under projected could suitability of for 55 widely distributed fishes differing thermal preferences region. Using boosted regression tree models, predicted at a stream reach scale using four contemporary variables including annual mean air temperature, precipitation, variation monthly temperature precipitation along seven variables. We then used values from eight general circulation models (GCMs) during 2041–2080 evaluate potential patterns richness, turnover, range shifts across study Results Most cold-water cool-water were lose habitat; however, loss also occurred certain small-bodied warm-water species. The percentage richness all reaches ranged − 40.4 33.93%, regions major losses occurring southern portions coast Midwest regions. Species turnover 0 43.5% substantial upper Midwest. Conclusions Temperature influence distribution substantially. Our findings provide multiple measures describing community aid management conservation future.

Language: Английский

---

Regional Research on Ecological Environment in China: A Literature Review

Regional science and environmental economics, Journal Year: 2025, Volume and Issue: 2(2), P. 13 - 13

Published: May 21, 2025

With the rapid development of China’s economy, resource consumption and environmental pollution have become challenges faced by China in its process. In order to effectively achieve a balance between economic ecological protection, Chinese government has successively introduced strategies for environment construction. However, how scientifically evaluate quality regional environments, analyze related impacts, promote national governance always been difficult reach consensus continues receive attention from academic community. This paper sorts through research recent years about assessments summarize current assessment methods dimensions China, as well impact terms evaluation methods, this analyzes applicability limitations mainstream methods. dimensions, summarizes results different dimensions. construction, elaborates on three aspects influencing factors, effects method analysis. Based above analysis, finally proposes that focus future should be digital analysis quality, so provide more scientific accurate support governance.

Language: Английский

---

Modelling the Spatial Distribution of Dosidicus gigas in the Southeast Pacific Ocean at Multiple Temporal Scales Based on Deep Learning

Fishes, Journal Year: 2025, Volume and Issue: 10(6), P. 273 - 273

Published: June 5, 2025

With the advent of big data era in ocean remote sensing and fisheries, there is a growing demand for finer temporal scales to predict spatial distribution jumbo flying squid (Dosidicus gigas). This can help reduce fuel costs provide higher quality faster decision-making. Therefore, this study employed deep neural network (DNN) model, using sea surface temperature, height, salinity, photosynthetically active radiation as input factors, with catch per unit effort output factor. We construct five cases 3, 6, 10, 15, 30 days spanning 10 years (2012–2021). Additionally, performance DNN was compared those traditional methods such generalized additive model (GAM), extreme gradient boosting (XGBoost), artificial (ANN). The results demonstrated that had best performance. As scale decreased, mean squared error absolute increased, whereas area under precision−recall curve indicating decline interpretability analysis indicated factors significantly contributed longitude exhibiting highest contribution. To improve accuracy scales, future research should focus on reducing noise address presence-only nature fishery data, particularly by cleaning unsampled portions.

Language: Английский

---

Nonstationary response of commercially pelagic species in the Humboldt current ecosystem to climate variability

Reviews in Fish Biology and Fisheries, Journal Year: 2024, Volume and Issue: 34(4), P. 1599 - 1617

Published: Aug. 27, 2024

Language: Английский

---

Habitat Distributions and Abundance of Four Wild Herbivores on the Qinghai–Tibetan Plateau: A Review

Land, Journal Year: 2024, Volume and Issue: 14(1), P. 23 - 23

Published: Dec. 26, 2024

Understanding the change in habitat distributions and abundance of wildlife space time is critical for conservation biodiversity mitigate human–wildlife conflicts (HWCs). Tibetan antelope or chiru (Pantholops hodgsonii), gazelle goa (Procapra picticaudata), wild ass kiang (Equus kiang), Wild yak (Bos mutus) have been sympatric on Qinghai–Tibetan plateau (QTP) numerous generations. However, reviews these four herbivores (WHs), as well methods examining changes aspects, are still lacking. Here, we firstly review major WHs QTP across different periods, underlying causes HWCs. Furthermore, critically compare three aspects methods: transect surveys, machine learning (ML), deep (DL) studying WHs. The results show that since 1990s, exhibited a trend initial decline followed by recovery, largely attributed to global climate warming decrease illegal hunting. recent years, primary challenge has shifted from protection balancing human interests within constraints limited resources. In future, should focus enhancing ecological functions habitats achieve harmonious coexistence between humans nature, establishing scientific compensation mechanism conflicts. order accurately calculate changes, select appropriate models analyze based their specific characteristics environmental conditions. Additionally, with advancement large models, AI (artificial intelligence) be utilized precise rapid conservation. findings this study also provide guidance reference addressing issues related other regions globally.

Language: Английский

---