A retrospective approach for evaluating ecological niche modeling transferability over time: the case of Mexican endemic rodents DOI Creative Commons
Claudia N. Moreno-Arzate, Enrique Martínez‐Meyer

PeerJ, Journal Year: 2024, Volume and Issue: 12, P. e18414 - e18414

Published: Nov. 29, 2024

Ecological niche modeling (ENM) is a valuable tool for inferring suitable environmental conditions and estimating species’ geographic distributions. ENM widely used to assess the potential effects of climate change on species distributions; however, choice algorithm introduces substantial uncertainty, especially since future projections cannot be properly validated. In this study, we evaluated performance seven popular algorithms—Bioclim, generalized additive models (GAM), linear (GLM), boosted regression trees (BRT), Maxent, random forest (RF), support vector machine (SVM)—in transferring across time, using Mexican endemic rodents as model system. We retrospective approach, from near past (1950–1979) more recent (1980–2009) vice versa, evaluate their in both forecasting hindcasting. Consistent with previous studies, our results highlight that input data quality significantly impact accuracy, but most importantly, found varied between While no single outperformed others temporal directions, RF generally showed better forecasting, while Maxent performed hindcasting, though it was sensitive small sample sizes. Bioclim consistently lowest performance. These findings underscore not all or algorithms are suited projections. Therefore, strongly recommend conducting thorough evaluation quality—in terms quantity biases—of interest. Based assessment, appropriate algorithm(s) should carefully selected rigorously tested before proceeding transfers.

Language: Английский

The METRIC-framework for assessing data quality for trustworthy AI in medicine: a systematic review DOI Creative Commons
Daniel Schwabe, Katinka Becker, Martin Seyferth

et al.

npj Digital Medicine, Journal Year: 2024, Volume and Issue: 7(1)

Published: Aug. 3, 2024

The adoption of machine learning (ML) and, more specifically, deep (DL) applications into all major areas our lives is underway. development trustworthy AI especially important in medicine due to the large implications for patients' lives. While trustworthiness concerns various aspects including ethical, transparency and safety requirements, we focus on importance data quality (training/test) DL. Since dictates behaviour ML products, evaluating will play a key part regulatory approval medical products. We perform systematic review following PRISMA guidelines using databases Web Science, PubMed ACM Digital Library. identify 5408 studies, out which 120 records fulfil eligibility criteria. From this literature, synthesise existing knowledge frameworks combine it with perspective medicine. As result, propose METRIC-framework, specialised framework training comprising 15 awareness dimensions, along developers should investigate content dataset. This helps reduce biases as source unfairness, increase robustness, facilitate interpretability thus lays foundation METRIC-framework may serve base systematically assessing datasets, establishing reference designing test datasets has potential accelerate

Language: Английский

Citations

13
Optimising occurrence data in species distribution models: sample size, positional uncertainty, and sampling bias matter DOI Creative Commons
Vítězslav Moudrý, Manuele Bazzichetto, Ruben Remelgado

et al.

Ecography, Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 2, 2024

Species distribution models (SDMs) have proven valuable in filling gaps our knowledge of species occurrences. However, despite their broad applicability, SDMs exhibit critical shortcomings due to limitations occurrence data. These include, particular, issues related sample size, positional uncertainty, and sampling bias. In addition, it is widely recognised that the quality as well approaches used mitigate impact aforementioned data depend on ecology. While numerous studies evaluated effects these SDM performance, a synthesis results lacking. without comprehensive understanding individual combined effects, ability predict influence modelled species–environment associations remains largely uncertain, limiting value model outputs. this paper, we review bias, ecology We build upon findings provide recommendations for assessment intended use SDMs.

Language: Английский

Citations

12
Assessing the applicability of binary land-cover variables to species distribution models across multiple grains DOI Creative Commons
Lukáš Gábor, Jeremy M. Cohen, Vítězslav Moudrý

et al.

Landscape Ecology, Journal Year: 2024, Volume and Issue: 39(3)

Published: March 4, 2024

Abstract Context Species distribution models are widely used in ecology. The selection of environmental variables is a critical step SDMs, nowadays compounded by the increasing availability data. Objectives To evaluate interaction between grain size and binary (presence or absence water) proportional (proportion water within cell) representation cover variable when modeling bird species distribution. Methods eBird occurrence data with an average number records 880,270 per across North American continent were for analysis. Models (via Random Forest) fitted 57 species, two seasons (breeding vs. non-breeding), at four grains (1 km 2 to 2500 ) using as variable. Results models’ performances not affected type adopted (proportional binary) but significant decrease was observed importance form. This especially pronounced coarser during breeding season. Binary useful finer sizes (i.e., 1 ). Conclusions At more detailed ), simple presence certain land-cover can be realistic descriptor occurrence. particularly advantageous collecting habitat field simply recording significantly less time-consuming than its total area. For grains, we recommend variables.

Language: Английский

Citations

5
Risk Assessment of Carbon Stock Loss in Chinese Forests Due to Pine Wood Nematode Invasion DOI Open Access

Shaoxiong Xu,

Wenjiang Huang, Dacheng Wang

et al.

Forests, Journal Year: 2025, Volume and Issue: 16(2), P. 315 - 315

Published: Feb. 11, 2025

Chinese forests, particularly the coniferous forest ecosystems represented by pines, play a crucial role in global carbon cycle, significantly contributing to mitigating climate change, regulating regional climates, and maintaining ecological balance. However, pine wilt disease (PWD), caused wood nematode (PWN), has become major threat stocks China. This study evaluates impact of PWN invasion on China using multi-source data an optimized MaxEnt model, analyzes this invasion’s spread trends potential risk areas. The results show that high-suitability area for expanded from 68,000 km2 2002 184,000 2021, with accelerating, especially under warm humid conditions due human activities. China’s increased 111.34 billion tons (tC) 168.05 tC, but also 87 million tC 99 highlighting ongoing storage capacity. further reveals significant differences tree species’ sensitivity PWN, highly sensitive species such as Masson’s black mainly concentrated southeastern coastal regions, while less white larch stronger resistance northern southwestern finding highlights vulnerability high-sensitivity high-risk areas Guangdong, Guangxi, Guizhou, where urgent effective control measures are needed reduce stock losses. To address challenge, recommends strengthening monitoring proposes specific improve management policy interventions, including promoting cross-regional joint control, enhancing early warning systems, utilizing biological measures, encouraging local governments communities actively participate. By collaboration implementing health sustainable development can be ensured, safeguarding forests’ important regulation sequestration change mitigation.

Language: Английский

Citations

0
Assessing the Impact of Climate Change on Hippophae neurocarpa in China Using Biomod2 Modeling DOI Creative Commons

Tingjiang Gan,

Quanwei Liu, Danping Xu

et al.

Agriculture, Journal Year: 2025, Volume and Issue: 15(7), P. 722 - 722

Published: March 27, 2025

Hippophae neurocarpa is a relatively new member of the Rhamnus genus that has various potential edible and medicinal values, but still needs to be further developed. To better develop H. neurocarpa, it crucial determine its current future population distribution. This study utilized “Biomod2” package in R integrate five individual models investigate effects climate change on distribution as well key climatic factors influencing The results indicated that, under scenario, mainly concentrated eastern parts Loess Plateau Qinghai–Tibet Plateau. In future, suitable habitats will undergo varying degrees change: area medium/low suitability decrease, while high shift westward increase. analysis changes, was found some Sichuan Shaanxi directly transition from highly unsuitable areas. Key environmental variable showed temperature, particularly low factor affecting neurocarpa. Additionally, altitude also significant impact predicted which aid development provide reference for selecting regions cultivation.

Language: Английский

Citations

0
Predicting the Future Geographic Distribution of the Traditional Chinese Medicinal Plant Epimedium acuminatum Franch. in China Using Ensemble Models Based on Biomod2 DOI Creative Commons
Zhiling Wang, Zhihang Zhuo, Biyu Liu

et al.

Plants, Journal Year: 2025, Volume and Issue: 14(7), P. 1065 - 1065

Published: March 30, 2025

This study employs the Biomod2 model, along with 22 bioclimatic variables, to predict geographic distribution of medicinal plant Epimedium acuminatum Franch. for current period and three future timeframes (2050s, 2070s, 2090s). Ultimately, 11 key environmental variables were identified as critical assessing habitat suitability plant. These include annual mean temperature (Bio 1), isothermally 3), seasonality 4), maximum warmest month 5), minimum coldest 6), driest quarter 9), 11), precipitation 17), elevation (Elev), aspect, slope. The results indicate that high areas are primarily distributed across Yunnan, Chongqing, Sichuan, Hunan, Guangxi, Hubei provinces. In future, extent is expected increase. aims provide a theoretical reference conservation E. genetic resources from perspective.

Language: Английский

Citations

0
The Potential Distribution Prediction of the Forestry Pest Cyrtotrachelus buqueti (Guer) Based on the MaxEnt Model across China DOI Open Access
Chun Fu, Zhiling Wang, Yaqin Peng

et al.

Forests, Journal Year: 2024, Volume and Issue: 15(6), P. 1049 - 1049

Published: June 18, 2024

Exploring the geographical distribution of forestry pests is crucial for formulating pest management strategies. Cyrtotrachelus buqueti (Guer) stands out as one primary among China’s hazards. This study employs MaxEnt model, along with 19 bioclimatic variables and habitat characteristics, to predict current future C. under three typical emission scenarios 2050 2070 (2.6 W/m2 (SSP1-2.6), 7.0 (SSP3-7.0), 8.5 (SSP5-8.5)). Among variables, BIO 14 (precipitation driest month), 8 (mean temperature wettest quarter), Elev, slope, aspect were identified significant contributors. These five are critical environmental factors affecting suitability habitats representative its potential habitat. The results indicate that predominantly inhabits southern regions such Chongqing, Guizhou, Yunnan, Sichuan, Guangxi, Shaanxi, Hubei, Hainan, Taiwan. them, Yunnan areas high suitability. In future, centroid’s movement direction will generally shift southward, an expansion trend observed in each province. enhances researchers’ understanding dynamics promotes proactive strategies mitigate their impact on forest ecosystems agricultural productivity.

Language: Английский

Citations

3
Spatiotemporal range dynamics and conservation optimization for endangered medicinal plants in the Himalaya DOI Creative Commons

F Liu,

Winnie W. Mambo,

Jie Liu

et al.

Global Ecology and Conservation, Journal Year: 2024, Volume and Issue: unknown, P. e03390 - e03390

Published: Dec. 1, 2024

Language: Английский

Citations

3
A perspective on the need for integrated frameworks linking species distribution and dynamic forest landscape models across spatial scales DOI Creative Commons
Anouschka R. Hof, Marco Mina, Paola Mairota

et al.

Frontiers in Ecology and Evolution, Journal Year: 2024, Volume and Issue: 12

Published: Sept. 19, 2024

Climate change significantly alters species distributions. Numerous studies project the future distribution of using Species Distribution models (SDMs), most often coarse resolutions. Working at resolutions in forest ecosystems fails to capture landscape-level dynamics, spatially explicit processes, and temporally defined events that act finer can disproportionately affect outcomes. Dynamic Forest Landscape Models (FLMs) simulate survival, growth, mortality (stands of) trees over long time periods small However, as they are able fine resolutions, study landscapes remain relatively due computational constraints. The large amount feedbacks between biodiversity, forest, ecosystem processes cannot completely be captured by FLMs or SDMs alone. Integrating with enables a more detailed understanding impact perturbations on their biodiversity. Several have used this approach landscape scales, Yet, many scientific questions fields biogeography, macroecology, conservation management, among others, require focus both scales Here, drawn from literature experience, we provide our perspective important challenges need overcome use integrated frameworks spatial larger than Future research should prioritize these better understand drivers distributions effectively design strategies under influence changing climates processes. We further discuss possibilities address challenges.

Language: Английский

Citations

2
Modeling the seasonal wildfire cycle and its possible effects on the distribution of focal species in Kermanshah Province, western Iran DOI Creative Commons
Maryam Morovati, Peyman Karami

PLoS ONE, Journal Year: 2024, Volume and Issue: 19(10), P. e0312552 - e0312552

Published: Oct. 28, 2024

Predicting environmental disturbances and evaluating their potential impacts on the habitats of various plant animal species is a suitable strategy for guiding conservation efforts. Wildfires are type disturbance that can affect many aspects an ecosystem its species. Therefore, through integration spatial models distribution (SDMs), we make informed predictions occurrence such phenomena impacts. This study focused five focal species, namely, brown bear ( Ursus arctos ), wild goat Capra aegagrus sheep Ovis orientalis wildcat Felis silvestris striped hyena Hyaena hyaena ). used MODIS active fire data ensemble machine learning methods to model risk wildfire in 2023 spring, summer, autumn separately. also investigated suitability via SDMs. The predicted probability maps habitat were converted binary values true skill statistic (TSS) threshold. overlap map areas was analyzed GAP analysis. area prone summer winter equal 9077.32; 10,199.83 13,723.49 KM 2 calculated, which indicates increase risk. Proximity roads one most important factors affecting possible effects wildfires all seasons. Most occurrences concentrated agricultural lands, which, when integrated with other land use types, have destroy residues critical factor wildfires. range each considered component susceptibility. Hence, autumn, 5.257, 5.856, 6.889 km respectively, affected by possibility fire. In contrast, these lowest , 162, 127, 396 respectively. dependent human-based ecosystems highest vulnerability wildfire. Conservation efforts should focus familiarizing farmers destroying as well consequences intentional fires. findings this be mitigate negative protect

Language: Английский

Citations

1