Improving the validation of ecological niche models with remote sensing analysis

José-Silva, Leandro; Santos, Reginaldo Carvalho dos; Lima, Bruna Martins de; Lima, Mendelson; Oliveira‐Júnior, José Francisco de; Teodoro, Paulo Eduardo; Eisenlohr, Pedro V.; Silva, Carlos Antônio da

doi:10.1016/j.ecolmodel.2018.04.013

Cited by 5 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The total number of occurrence records was divided into 70% for training and 30% for validation [45,46], with five replications for calibration of the adjusted models. For the RF, CTA, and GBM algorithms, the processing considered the number of pseudo-absences (PAs) equal to the number of occurrence points of the species.…”

Section: Algorithms Used Processing Parameters and Model Evaluationmentioning

confidence: 99%

Geomorphological and Bioclimatic Relationships in the Occurrence of Species of Agro-Extractivist Interest in the Cerrado Biome

Matias,

Rezende,

Sevilha

et al. 2024

Sustainability

View full text Add to dashboard Cite

The distribution of species of agro-extractivist interest and their ecological relationship with the physical environment geomorphological and bioclimatic allow supporting strategies aimed at socioeconomic and environmental development. We evaluated the contribution of high spatial resolution topographic variables in ecological niche models and the relationship of the distribution of five tree species with the geomorphological units and bioclimatic variables. The variables related to temperature variation and water availability proved to be important in predicting the areas of occurrence of the target species, with increased suitability of occurrence in regions with higher isothermality, located in the plateau and table geomorphological units. The predictions showed a significant difference when high spatial resolution variables were used, generating a more conservative scenario in the indication of suitable regions for the occurrence of species, important for local scale studies. The geomorphological units of plateau and tableland showed high suitability of occurrence, while the fluvial plains and dissected depressions did not present suitability for the occurrence of the species. The results allow us to strategically define areas with the greatest productive potential and prioritize areas for conservation, management, ecological restoration of forests, and targeting areas for the implementation of community agro-industries, essential for territorial planning within traditional communities.

show abstract

Section: Algorithms Used Processing Parameters and Model Evaluationmentioning

confidence: 99%

Geomorphological and Bioclimatic Relationships in the Occurrence of Species of Agro-Extractivist Interest in the Cerrado Biome

Matias,

Rezende,

Sevilha

et al. 2024

Sustainability

View full text Add to dashboard Cite

show abstract

“…For examples, photosynthetic activity derived from a SPOT-NDVI time series at a 1 km spatial resolution was used to classify the potential extent of tree species [30], and LiDAR-derived light availability was used to classify the potential extent of understory species [31]. Interestingly, several studies classified the potential area of vegetation using topo-climatic variables and then classified the actual vegetation area using vegetation variables [5,32,33].…”

Section: Identifying Potential Restoration or Invasion Areasmentioning

confidence: 99%

“…To decrease the bias in statistical accuracy indices, several studies based on "presenceonly" classifiers included absence reference data extracted from RS images and/or vegetation maps to estimate classification accuracy using Kappa, overall accuracy, or F1-score indices [5,32,51,73,137]. Besides, the literature review highlights that spatial autocorrelation in reference data was usually ignored (Table 9).…”

Section: Assessing Classification Accuracymentioning

confidence: 99%

One-Class Classification of Natural Vegetation Using Remote Sensing: A Review

Rapinel

Hubert‐Moy

2021

Remote Sensing

View full text Add to dashboard Cite

Advances in remote sensing (RS) technology in recent years have increased the interest in including RS data into one-class classifiers (OCCs). However, this integration is complex given the interdisciplinary issues involved. In this context, this review highlights the advances and current challenges in integrating RS data into OCCs to map vegetation classes. A systematic review was performed for the period 2013–2020. A total of 136 articles were analyzed based on 11 topics and 30 attributes that address the ecological issues, properties of RS data, and the tools and parameters used to classify natural vegetation. The results highlight several advances in the use of RS data in OCCs: (i) mapping of potential and actual vegetation areas, (ii) long-term monitoring of vegetation classes, (iii) generation of multiple ecological variables, (iv) availability of open-source data, (v) reduction in plotting effort, and (vi) quantification of over-detection. Recommendations related to interdisciplinary issues were also suggested: (i) increasing the visibility and use of available RS variables, (ii) following good classification practices, (iii) bridging the gap between spatial resolution and site extent, and (iv) classifying plant communities.

show abstract

“…The integration of remote sensing (RS) products into the development of ENMs has gained signi cant attention in recent years (see Sillero et al, 2009Sillero et al, , 2012Campos et al, 2023;Regos et al, 2022). This upward trend continues to increase due to the potential to enhance the performance of ENMs, improving multi-scale accuracy and detailed estimations of species' conservation status (Arenas-Castro & Sillero, 2021; Campos et al, 2023;José-Silva et al, 2018;Randin et al, 2020). It also provides knowledge of habitat quality and dynamics through in-depth analysis (Regos et al, 2022).…”

Section: Introductionmentioning

confidence: 99%

Assessing spatial and temporal trends over time in potential species richness using satellite time-series and ecological niche models

Garcia,

Campos,

Alírio

et al. 2024

Preprint

View full text Add to dashboard Cite

Assessing biodiversity in protected areas is crucial. However, conventional methods for estimating species richness can lead to inaccurate assessments of biodiversity in these regions. To address this, we present a complementary approach to assess potential species richness at a finer resolution (1 km) in the special conservation area of Montesinho/Nogueira (SAC-MN; Portugal), based on a spatial and temporal estimates of species richness trends over time (2001–2021). We compiled 10,190 species occurrence records from four taxonomic groups: amphibians, reptiles, birds, and mammals. We created correlative ecological niche models (MaxEnt) with a time series of satellite-remote sensing variables (SRS-ENMs) using solely Google Earth Engine (GEE). We calculated annual potential species richness (PSR) maps by stacking the SRS-ENMs and identified positive and negative trends with the Mann-Kendall test. Individual species’ SRS-ENMs indicated distinct habitat suitability within SAC-MN over time. The eastern and western boundaries of SAC-MN revealed an increase in PSR, while the central area indicated PSR declines. Each group displayed different trends in PSR: birds and mammals were more similar between them than amphibians and reptiles. This approach serves as a valuable instrument for ecologists and conservationists, assisting in evaluating PSR and trends over time using remote sensing. Furthermore, it provides meaningful insights for subsequent analysis in order to inform decision-makers and other stakeholders to enhance local biodiversity conservation efforts.

show abstract

Improving the validation of ecological niche models with remote sensing analysis

Cited by 5 publications

References 45 publications

Geomorphological and Bioclimatic Relationships in the Occurrence of Species of Agro-Extractivist Interest in the Cerrado Biome

Geomorphological and Bioclimatic Relationships in the Occurrence of Species of Agro-Extractivist Interest in the Cerrado Biome

One-Class Classification of Natural Vegetation Using Remote Sensing: A Review

Assessing spatial and temporal trends over time in potential species richness using satellite time-series and ecological niche models

Contact Info

Product

Resources

About