Influence of model complexity, training collinearity, collinearity shift, predictor novelty and their interactions on ecological forecasting

Global Ecology and Biogeography, Journal Year: 2023, Volume and Issue: 33(3), P. 371 - 384

Published: Nov. 29, 2023

Abstract Aim Ecological forecasting is critical in understanding of ecological responses to climate change and increasingly used mitigation plans. The forecasts from correlative models can be challenged by model complexity, training collinearity, collinearity shift novel conditions predictors that are common during extrapolation. individual effect these four factors has been investigated, but it still unclear how interactively affect forecasting. To fill this gap, we conducted a comprehensive simulation experiment quantify the influence Location Simulated regions. Time Period scenarios. Methods We modelled three response variables commonly following normal, Poisson binomial distributions as function functional relationships represented complexity under levels using generalized linear models. By calculating prediction error 3,780,000 testing scenarios, partitioned its variance shift, predictor novelty their interactions. Results found increased degraded performance, leading up double errors when predictor's range ~22% or correlation r between two changed >~0.8 for combination high interaction relationship. Predictor reduced on suggesting negative them. This pattern was more pronounced collinearity. Main Conclusions accuracy depends Besides consideration parsimonious 0.7 training, our study further recommends threshold <22%–50% depending and/or <0.8 making reliable

Language: Английский

---

Conservation Responsibility for Priority Habitats under Future Climate Conditions: A Case Study on Juniperus drupacea Forests in Greece

Land, Journal Year: 2023, Volume and Issue: 12(11), P. 1976 - 1976

Published: Oct. 26, 2023

Juniperus drupacea is a highly morphologically and genetically differentiated Tertiary relict, displaying disjunct geographical range in the eastern Mediterranean. Being thermophilous, light-demanding, moderately drought-resistant tree, it survived past climatic oscillations via altitudinal migration. The species has its westernmost limit, only populations EU, Mts Parnon Taygetos (Greece). These are isolated distinct compared to their Asian counterparts. For Europe, categorized as an endangered by IUCN. forests constitute priority habitat for conservation EU. However, species’ status never been assessed Greece same applies climate land-use change assessment. As already facing short-term impacts of climate- human-induced change, studies dealing with potential long-term effects on rare plant distribution urgently needed implement efficient management plans. Our research employs models, considering multiple scenarios abiotic factors across different timeframes (2020s, 2050s, 2080s), factoring threat forest fires. Additionally, we assess extinction risk at European level, according IUCN Criteria A B. Study findings indicate significant changes elevated Greece. To safeguard this habitat, informed strategies, plans, policy making recommended, based our scientific insights.

Language: Английский

---

Key factors for species distribution modeling in benthic marine environments

Frontiers in Marine Science, Journal Year: 2023, Volume and Issue: 10

Published: Dec. 4, 2023

Species distribution modeling is a widely used technique for estimating the potential habitats of target organisms based on their environmental preferences. These methods serve as valuable tools resource managers and conservationists, utilization increasing, particularly in marine environments where data limitations persist challenge. In this study, we employed global predictions six cold-water coral species case study to investigate various factors influencing predictions, including algorithms, background points sampling strategies sizes, collinearity datasets, using both discriminative functional performance metrics. The choice method exhibits stronger influence model compared effects point size, dataset. Predictions that utilize kernel density backgrounds, maintain an equal number presences algorithms BRT, RF, MARS, employ substantial MAXENT, coupled with collinearity-filtered dataset modeling, yield higher levels performance. Overall, BRT RF outperformed conclusion further substantiated by analysis smoothed residuals uncertainty associated predicted habitat suitability Madrepora oculata . This offers insights enhancing benthic environments, thereby benefiting management conservation species.

Language: Английский

---

High‐Resolution Ensemble Modelling of Coral Distributions in the Northern Gulf of Mexico Based on Geomorphometry: Coral Diversity and Benthic Habitat Fragmentation From Oil and Gas Infrastructure to Inform Spatial Planning

Aquatic Conservation Marine and Freshwater Ecosystems, Journal Year: 2024, Volume and Issue: 34(9)

Published: Sept. 1, 2024

ABSTRACT The northern Gulf of Mexico is home to several species corals that provide a wide range ecosystem services other organisms. Oil and gas infrastructure, such as platforms pipelines, form an extensive network throughout the Mexico. Detrimental impacts associated with oil exploration extraction have been recorded in this area at depths where are found. Due these ecosystems' vulnerability long‐term impacts, it necessary determine areas interest would benefit from further informed spatial planning. This study aimed identify potential for coral studies Ensemble distribution models 13 including scleractinians, black corals, octocorals were produced based on seafloor characteristics combined relatively higher diversity than others. ensemble modelling approach robust outputs, evaluated by under curve, Cohen's kappa coefficient, sensitivity, specificity proportion correct predictions. proximity suitable habitat active proposed infrastructure was evaluated; analysis showed infrastructures potentially impact 23.5% all predicted contribute benthic fragmentation. Twelve greater 100 km 2 located outside 4‐km zone influence delineated deemed planning, hypothetical prioritization scenarios planning presented. maps can inform discussions among stakeholders reach best outcomes while considering ecological, social, economic governance factors.

Language: Английский

---

Modeling the Current and Future Distribution of Indianthus virgatus (Roxb.) Suksathan & Borchs.: A Monotypic Plant Endemic to the Western Ghats‐Sri Lanka Biodiversity Hotspot

Ecology and Evolution, Journal Year: 2024, Volume and Issue: 14(10)

Published: Oct. 1, 2024

Species distribution modeling (SDM) is an essential tool in ecology and conservation for predicting species distributions based on presence/absence data environmental variables. The present study aimed to understand the pattern habitat suitability of

Language: Английский

---