Importance of data selection and filtering in species distribution models: A case study on the Cantabrian brown bear

Zarzo‐Arias, Alejandra; Penteriani, Vincenzo; Gábor, Lukáš; Šímová, Petra; Grattarola, Florencia; Moudrý, Vítězslav

doi:10.1002/ecs2.4284

Cited by 12 publications

(11 citation statements)

References 118 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Selecting the appropriate environmental characteristics is a critical step in building these models. Climatic variables indirectly influence animal distribution through their effects on food and shelter, particularly for large‐medium sized species (Zarzo‐Arias et al., 2022). The collinearity among these variables can significantly compromise the model's accuracy (Yoon & Lee, 2021).…”

Section: Methodsmentioning

confidence: 99%

Evaluating dynamics in human‐bear interactions within social‐ecological system

Ji,

Wei,

2024

People and Nature

View full text Add to dashboard Cite

Over the past few decades, social‐ecological transformations have contributed to the restoration of numerous carnivores, presenting new challenges for human‐wildlife coexistence. Evaluating and understanding human‐wildlife interaction dynamics, particularly for conflict‐prone species, is crucial for conservation and management strategies. Our study focuses on Asiatic Black Bear (Ursus thibetanus) in Western Yunnan, China, a region with favourable bear conditions and significant environmental changes due to conservation policies. This study employed a multifaceted approach to gather data on bear occurrences and human‐bear conflicts. Utilising species distribution models, we predicted the potential bear habitat in 2000 and 2020. We proposed a methodology integrating bear habitat with human footprint to delineate the human‐bear interface. Additionally, we evaluated the spatial–temporal variations in bear habitat and human‐bear interface between 2000 and 2020. Semi‐structured interviews were conducted to capture the perceptions of local communities regarding the bear population, human‐bear interactions and conservation policies. Our findings indicated that the potential habitats of bears were strongly influenced by forest cover. We noticed improvements in bear habitat from 2000 to 2020, notwithstanding localised losses in specific areas. Most respondents attributed the increasing bear population to forest restoration and reduced human impact during this period. Notably, a significant proportion of respondents supported the conservation policies, previously participating in their formulation and implementation. Our analysis further revealed increases in the human‐bear interface and human‐bear interactions from 2000 to 2020. Remarkably, these changes did not significantly diminish local communities' endorsement of conservation policies. This resilience can be attributed to the net positive impacts of conservation policies on the livelihoods of these communities. Our study highlights the complexity and dynamic nature of human‐wildlife interactions within the social‐ecological system and provides new insights to facilitate sustainable conservation and management measures. Read the free Plain Language Summary for this article on the Journal blog.

show abstract

Section: Methodsmentioning

confidence: 99%

Evaluating dynamics in human‐bear interactions within social‐ecological system

Ji,

Wei,

2024

People and Nature

View full text Add to dashboard Cite

show abstract

“…We built SDMs in the statistical software R (package ‘sdm' ver. 1.0‐98, http://www.r-project.org; Naimi and Araújo 2016) using the MaxEnt modeling method (Phillips et al 2006), a presence‐background method often adopted in ecological studies (Rodríguez et al 2019, Santamarina et al 2019, Ancillotto et al 2020, El‐Gabbas et al 2021, Boral and Moktan 2021, Ellis‐Soto et al 2021, Venne and Currie 2021, Zarzo‐Arias et al 2022). We used 10 000 randomly sampled background points and default model settings (Phillips and Dudík 2008), except that we set the beta parameter to 0.5 and restricted used features.…”

Section: Methodsmentioning

confidence: 99%

“…Species distribution model (SDMs) are a widely used class of ecological models that use occurrence data to estimate species–environment relationships (Ferrier et al 2017) and allow researchers to predict the relative probability of occurrence across unsampled areas of a study region. SDMs have broad utility in ecology (Elith and Leathwick 2009, Franklin 2010, Guisan et al 2014, Zurell et al 2022) and have been successfully used to identify critical habitats (Volis and Tojibaev 2021), delineate suitable locations for relocations (Segal et al 2021), or assess the potential impacts of climate change (Santini et al 2021). SDMs are also frequently used to infer the importance of environmental variables defining the species niche (Moudrý and Šímová 2013, Bradie and Leung 2017, Lecours et al 2020, Li and Kou 2021, Smith and Santos 2020) and to determine the shapes of species responses to the environment (Austin et al 2006, Hargreaves et al 2014, Lee‐Yaw et al 2016, Dvorský et al 2017, Bazzichetto et al 2018).…”

Section: Introductionmentioning

confidence: 99%

Species distribution models affected by positional uncertainty in species occurrences can still be ecologically interpretable

et al. 2023

Self Cite

View full text Add to dashboard Cite

Species distribution models (SDMs) have become a common tool in studies of species–environment relationships but can be negatively affected by positional uncertainty of underlying species occurrence data. Previous work has documented the effect of positional uncertainty on model predictive performance, but its consequences for inference about species–environment relationships remain largely unknown. Here we use over 12 000 combinations of virtual and real environmental variables and virtual species, as well as a real case study, to investigate how accurately SDMs can recover species–environment relationships after applying known positional errors to species occurrence data. We explored a range of environmental predictors with various spatial heterogeneity, species' niche widths, sample sizes and magnitudes of positional error. Positional uncertainty decreased predictive model performance for all modeled scenarios. The absolute and relative importance of environmental predictors and the shape of species–environmental relationships co‐varied with a level of positional uncertainty. These differences were much weaker than those observed for overall model performance, especially for homogenous predictor variables. This suggests that, at least for the example species and conditions analyzed, the negative consequences of positional uncertainty on model performance did not extend as strongly to the ecological interpretability of the models. Although the findings are encouraging for practitioners using SDMs to reveal generative mechanisms based on spatially uncertain data, they suggest greater consequences for applications utilizing distributions predicted from SDMs using positionally uncertain data, such as conservation prioritization and biodiversity monitoring.

show abstract

“…With coarser response grains, studies often include (bio)climatic variables (Chauvier et al, 2022; Kaliontzopoulou et al, 2008; Seo et al, 2009). Typically, such studies report declining model performance with the coarsening of the resolution of the response variable (Chauvier et al, 2022; Gábor et al, 2022a; Heikkinen et al, 2007; Kaliontzopoulou et al, 2008; Seo et al, 2009; Zarzo-Arias et al, 2022), suggesting that modelling species at coarser resolutions is not optimal. However, these studies typically focus on the general performance of the models and do not report the effect of changing the response grain on the variables’ importance, which may provide valuable insights into which variables shape species distributions at individual grain sizes (but see Chauvier et al, 2022; Hanberry, 2013).…”

Section: Effects Of Changing the Resolution Of Predictor And Response...mentioning

confidence: 99%

Scale mismatches between predictor and response variables in species distribution modelling: A review of practices for appropriate grain selection

Moudrý

Keil

Cord

et al. 2023

Progress in Physical Geography: Earth and Environment

Self Cite

View full text Add to dashboard Cite

There is a lack of guidance on the choice of the spatial grain of predictor and response variables in species distribution models (SDM). This review summarizes the current state of the art with regard to the following points: (i) the effects of changing the resolution of predictor and response variables on model performance; (ii) the effect of conducting multi-grain versus single-grain analysis on model performance; and (iii) the role of land cover type and spatial autocorrelation in selecting the appropriate grain size. In the reviewed literature, we found that coarsening the resolution of the response variable typically leads to declining model performance. Therefore, we recommend aiming for finer resolutions unless there is a reason to do otherwise (e.g. expert knowledge of the ecological scale). We also found that so far, the improvements in model performance reported for multi-grain models have been relatively low and that useful predictions can be generated even from single-scale models. In addition, the use of high-resolution predictors improves model performance; however, there is only limited evidence on whether this applies to models with coarser-resolution response variables (e.g. 100 km2 and coarser). Low-resolution predictors are usually sufficient for species associated with fairly common environmental conditions but not for species associated with less common ones (e.g. common vs rare land cover category). This is because coarsening the resolution reduces variability within heterogeneous predictors and leads to underrepresentation of rare environments, which can lead to a decrease in model performance. Thus, assessing the spatial autocorrelation of the predictors at multiple grains can provide insights into the impacts of coarsening their resolution on model performance. Overall, we observed a lack of studies examining the simultaneous manipulation of the resolution of predictor and response variables. We stress the need to explicitly report the resolution of all predictor and response variables.

show abstract

Importance of data selection and filtering in species distribution models: A case study on the Cantabrian brown bear

Cited by 12 publications

References 118 publications

Evaluating dynamics in human‐bear interactions within social‐ecological system

Evaluating dynamics in human‐bear interactions within social‐ecological system

Species distribution models affected by positional uncertainty in species occurrences can still be ecologically interpretable

Scale mismatches between predictor and response variables in species distribution modelling: A review of practices for appropriate grain selection

Contact Info

Product

Resources

About