Species-environment Relationships Research Articles

A Novel Web-Based Approach for Monitoring Biodiversity.

Understanding complexities in biodiversity is one of the fundamental goals of ecology and its monitoring is significant for ecosystem sustainability, maintenance, and conservation. However, biodiversity monitoring needs improvement to handle complex datasets and their analyses. This study attempts to understand these ecological complexities quickly, efficiently, and easily. The aim is to provide an alternative to ecologists, researchers, instructors, and stakeholders for biodiversity monitoring with the flexibility to visualize and customize outputs without software knowledge. A novel web-based technique is applied to monitor the biodiversity of a complex mountain ecosystem using a national database. The species-environment relationships of different vegetation types across a mountain ecosystem's elevation gradient are investigated using open-source climatic, physiographic, and socioeconomic variables. The proposed interactive tool to monitor biodiversity and understand its complexities is designed to visualize the data structure, summary, correlations, and sampling effectiveness quickly and easily. Plant species richness patterns and life forms (herb, shrub, and tree) across elevational gradients are investigated. We highlight the preliminary investigation of the data structure and their spatial distribution and apply the multicollinearity test to select variables for modeling. The drop-down menu helps users browse different datasets and select those datasets for instant visualization. Preliminary investigations on interactions between variables and species richness of vegetation types along elevation gradient interactively displayed with options to select variables, plant richness, and an elevational range. Species-environment relationships are investigated using multiple modeling protocols, and results are interactively displayed with options to download in different file formats and colors at the click of a button. This visualization tool helps to understand ecosystem structure, species richness patterns and species-environment relationships easily and efficiently. The R-codes used in this tool are reproducible and can be implemented with multiple datasets to monitor ecosystems.

Simulating multi-scale optimization and variable selection in species distribution modeling

Species distribution modeling (SDM) is a fundamental tool in theoretical and applied ecology. However, relatively little is known about the performance of different approaches for scale optimization, model selection, and algorithmic prediction in the context of nonlinear, multiscale and interactive relationships between environmental variables and species occurrence. Modelers often struggle to optimize a tradeoff between ecological relevance, model robustness, complexity, and overfitting. In this paper, we investigated several methods designed to optimize spatial scale and variable selection in SDMs, in each case evaluating model fitness, parsimony and predictive performance. We used a simulation approach to produce a large pool of alternative underlying habitat relationships that reflect a broad range of realistic habitat associations. We also compared several different modeling algorithms, including logistic regression with a generalized linear model (GLM), Lasso and Elastic-Net Regularized GLMs (GLMNet), and random forest (RF), as well as alternative variable and scale selection methods. We found that GLM methods employing all-subsets dredge routines for variable selection were consistently the best predictors based on all criteria of our model performance assessment and across all attributes of the simulated underlying relationship, including nonlinearity and interaction. We had expected machine learning approaches, such as random forest, to perform better in these more complex forms of species-environment relationships. GLM using dredge variable selection was also the method that included the fewest spurious covariates and included the most correct predictors as a proportion of all predictors. We found that univariate scaling was the most robust method of variable and scale selection, along with Minimal Redundancy Maximal Relevancy (MRMR) which performed equivalently. The simulation experiment presented here provides a robust assessment of simulated multi-species distribution model performance, complexity and fidelity. By simulating a large range of potential habitat relationships with varying spatial scale, effect sizes, linearity, and interactions, we comprehensively evaluated model performance across gradients of complexity of the underlying relationships and violations of classical statistical assumptions. This study provides a valuable assessment and a broader example of the power and utility of controlled simulation experiments in habitat relationships and other ecological spatial predictive modeling.

Assessing abundance-suitability models to prioritize conservation areas for the dwarf caimans in South America.

Species-environment relationships have been extensively explored through species distribution models (SDM) and species abundance models (SAM), which have become key components to understand the spatial ecology and population dynamics directed at biodiversity conservation. Nonetheless, within the internal structure of species' ranges, habitat suitability and species abundance do not always show similar patterns, and using information derived from either SDM or SAM could be incomplete and mislead conservation efforts. We gauged support for the abundance-suitability relationship and used the combined information to prioritize the conservation of South American dwarf caimans (Paleosuchus palpebrosus and P. trigonatus). We used 7 environmental predictor sets (surface water, human impact, topography, precipitation, temperature, dynamic habitat indices, soil temperature), 2 regressions methods (Generalized Linear Models-GLM, Generalized Additive Models-GAM), and 4 parametric distributions (Binomial, Poisson, Negative binomial, Gamma) to develop distribution and abundance models. We used the best predictive models to define four categories (low, medium, high, very high) to plan species conservation. The best distribution and abundance models for both Paleosuchus species included a combination of all predictor sets, except for the best abundance model for P. trigonatus which incorporated only temperature, precipitation, surface water, human impact, and topography. We found non-consistent and low explanatory power of environmental suitability to predict abundance which aligns with previous studies relating SDM-SAM. We extracted the most relevant information from each optimal SDM and SAM and created a consensus model (2,790,583 km2) that we categorized as low (39.6%), medium (42.7%), high (14.9%), and very high (2.8%) conservation priorities. We identified 279,338 km2 where conservation must be critically prioritized and only 29% of these areas are under protection. We concluded that optimal models from correlative methods can be used to provide a systematic prioritization scheme to promote conservation and as surrogates to generate insights for quantifying ecological patterns.

Predicted future range expansion of a small carnivore: swift fox in North America

ContextSmall carnivores are declining globally due to a complex suite of threats. Conservation of these species requires an understanding of their distributions and potential responses to future land-use and climate change.ObjectivesWe modelled species-environment relationships of swift fox (Vulpes velox), a species of concern across their range. We developed spatial projections of current and future distribution to aid in conservation planning.MethodsWe assembled swift fox occurrence data from managers and community science sources to develop ensemble distribution models. In addition to landscape and climatic predictors, we developed a model of red fox distribution to represent effects of competition. We forecasted spatial predictions into the year 2070 under two climate change scenarios representing high (SSP 5–8.5) and low (SSP 1–2.6) emissions scenarios.ResultsPercent cover by grassland, mean annual precipitation, and minimum temperature of the coldest quarter were the three most important variables for swift fox distribution. Current suitable habitat for swift fox extends across 16 North American states and provinces. Future projections of swift fox distribution suggest an overall increase in area of swift fox suitable habitat under both emissions scenarios of > 56.9%, though patterns of gain and loss vary spatially.ConclusionsThe expansion of suitable habitat in future scenarios reflects swift fox adaptability to multiple land uses in a period following multi-organizational conservation efforts. Our spatial projections can be used in conservation planning and can serve as a case study of a small carnivore species likely to recover under future change scenarios provided that threats are addressed and landscape-scale conservation efforts continue.

Habitat suitability modelling to predict the distribution of deep coral ecosystems: The case of Linosa Island (southern Mediterranean Sea, Italy)

In areas with limited field data, predictive habitat mapping is a valuable method for elucidating species-environment relationships and enhancing our knowledge of the spatial distribution and complexity of benthic habitats. Species distribution models (SDMs) can be an important tool to support in science-based ecosystem management. The availability of direct observations of mesophotic species, including gorgonians and black corals, during costly surveys is generally limited. Therefore, predicting the distribution of mesophotic species in relation to key physical parameters of the seafloor would help improving conservation strategies in existing and new Marine Protected Areas (MPAs). This study aims to assess the distribution of gorgonians and black corals off Linosa Island, in the Strait of Sicily, a biogeographic boundary area between the western and eastern Mediterranean. The volcanic island of Linosa represents a small, naturally preserved area, with very limited human pressure, hosting rich marine benthic biodiversity on its wide submarine portions. Distribution of the most common coral species off Linosa Island was modelled under an SDM framework, relying on direct observations collected during two research cruises in 2016 and 2017 and a series of terrain parameters acquired through geophysical techniques. We used the so-called “ensemble of small models” approach to calibrate SDMs, which achieved fair-to-excellent results (AUC >0.7). In addition to identifying depth as the primary factor influencing coral distribution, our study also highlighted ruggedness as a significant terrain variable. Specifically, the depth range of 110–230 m emerged as the critical parameter determining habitat suitability for all modelled species, also highlighting peculiar and specie–specific habitat requirements.

Ocean warming and acidification adjust inter- and intra-specific variability in the functional trait expression of polar invertebrates

Climate change is known to affect the distribution and composition of species, but concomitant alterations to functionally important aspects of behaviour and species-environment relations are poorly constrained. Here, we examine the ecosystem ramifications of changes in sediment-dwelling invertebrate bioturbation behaviour—a key process mediating nutrient cycling—associated with near-future environmental conditions (+ 1.5 °C, 550 ppm [pCO2]) for species from polar regions experiencing rapid rates of climate change. We find that responses to warming and acidification vary between species and lead to a reduction in intra-specific variability in behavioural trait expression that adjusts the magnitude and direction of nutrient concentrations. Our analyses also indicate that species behaviour is not predetermined, but can be dependent on local variations in environmental history that set population capacities for phenotypic plasticity. We provide evidence that certain, but subtle, aspects of inter- and intra-specific variation in behavioural trait expression, rather than the presence or proportional representation of species per se, is an important and under-appreciated determinant of benthic biogeochemical responses to climate change. Such changes in species behaviour may act as an early warning for impending ecological transitions associated with progressive climate forcing.

The niche through time: Considering phenology and demographic stages in plant distribution models

Abstract Species distribution models (SDMs) are widely used to infer species–environment relationships, predict spatial distributions and characterise species' environmental niches. While the importance of space and spatial scales is widely acknowledged in SDM applications, temporal components of the niche are rarely addressed. We discuss how phenology and demographic stages affect model inference in plant SDMs. Ignoring conspicuousness and timing of phenological stages may bias niche estimates through increased observer bias, while ignoring stand age may bias niche estimates through temporal mismatches with environmental variables, especially during times of rapid global warming. We present different methods to consider phenology and demographic stages in plant SDMs, including the selection of causal, spatiotemporally explicit predictors and the calibration of stage‐specific SDMs. Based on a case study with citizen science data, we illustrate how spatiotemporal SDMs provide deeper insights on the coincidence of range and phenological shifts under climate change. The proliferation of digitally available biodiversity and citizen science data increasingly allows considering time explicitly in SDMs. This offers a more mechanistic understanding of plant distributions, and more robust predictions under global change, especially if the reporting of phenological stages and age is facilitated and promoted by relevant data portals.

Female proportion has a stronger influence on dispersal than body size in nematodes of mountain lakes.

Nematodes disperse passively and are amongst the smallest invertebrates on Earth. Free-living nematodes in mountain lakes are highly tolerant of environmental variations and are thus excellent model organisms in dispersal studies, since species-environment relationships are unlikely to interfere. In this study, we investigated how population or organism traits influence the stochastic physical nature of passive dispersal in a topologically complex environment. Specifically, we analyzed the influence of female proportion and body size on the geographical distribution of nematode species in the mountain lakes of the Pyrenees. We hypothesized that dispersal is facilitated by (i) a smaller body size, which would increase the rate of wind transport, and (ii) a higher female proportion within a population, which could increase colonization success because many nematode species are capable of parthenogenetic reproduction. The results showed that nematode species with a low proportion of females tend to have clustered spatial distributions that are not associated with patchy environmental conditions, suggesting greater barriers to dispersal. When all species were pooled, the overall proportion of females tended to increase at the highest elevations, where dispersal between lakes is arguably more difficult. The influence of body size was barely relevant for nematode distributions. Our study highlights the relevance of female proportion as a mechanism that enhances the dispersal success of parthenogenetic species, and that female sex is a determining factor in metacommunity connectivity.

Refining Ixodes scapularis (Acari: Ixodidae) distribution models: a comparison of current methods to an established protocol.

Blacklegged ticks (Ixodes scapularis Say) pose an enormous public health risk in eastern North America as the vector responsible for transmitting 7 human pathogens, including those causing the most common vector-borne disease in the United States, Lyme disease. Species distribution modeling is an increasingly popular method for predicting the potential distribution and subsequent risk of blacklegged ticks, however, the development of such models thus far is highly variable and would benefit from the use of standardized protocols. To identify where standardized protocols would most benefit current distribution models, we completed the "Overview, Data, Model, Assessment, and Prediction" (ODMAP) distribution modeling protocol for 21 publications reporting 22 blacklegged tick distribution models. We calculated an average adherence of 73.4% (SD ± 29%). Most prominently, we found that authors could better justify and connect their selection of variables and associated spatial scales to blacklegged tick ecology. In addition, the authors could provide clearer descriptions of model development, including checks for multicollinearity, spatial autocorrelation, and plausibility. Finally, authors could improve their reporting of variable effects to avoid undermining the models' utility in informing species-environment relationships. To enhance future model rigor and reproducibility, we recommend utilizing several resources including the ODMAP protocol, and suggest that journals make protocol compliance a publication prerequisite.

Arctic puzzle: Pioneering a northern shrimp (Pandalus borealis) habitat model in Disko Bay, West Greenland

Recent advancements in spatial modelling leverage remote sensing data and statistical species-environment relationships to forecast the distribution of a specific species. Our study focuses on Disko Bay in West Greenland, recognized as a significant marine biodiversity hotspot in the region. We conducted comprehensive analyses using multiple datasets spanning from 2010 to 2019, incorporating shrimp and fish surveys, commercial shrimp fishery catches, high-resolution (25 × 25 m) multibeam bathymetry and backscatter data along with a medium-resolution (200 × 200 m) bathymetric model, measured and modelled oceanographic data, and satellite chlorophyll data. Through multivariate regression analysis, we tested the significance of various physical factors (seafloor depth, sediment class, bottom water temperature, bottom water salinity, bottom current velocity, space, and time), biological factors (chlorophyll a, Greenland halibut (Reinhardtius hippoglossoides)), and anthropogenic impact (shrimp fishery; standardized catch per unit effort) on the density of northern shrimp in the area. Our results indicate a significant association between northern shrimp density, seafloor depth, and sediment class, explaining 36 % of the variation in shrimp density. Subsequently, we developed a high-resolution (optimized) spatial linear mixed–effect model to map the distribution of northern shrimp across Disko Bay, representing the first model of its kind developed for an Arctic area. The optimal habitat for northern shrimp is characterized by medium-deep waters (approximately 150–350 m), turbulent conditions, and mixed sediments, predominantly located in the northern and southern regions of Disko Bay. Notably, the northern region hosts a relatively diverse benthic community, with northern shrimp and sponges as the primary contributors of epibenthic biomass. This novel high-resolution model significantly enhances our understanding of the physical drivers and detailed spatial patterns influencing the distribution of northern shrimp in the Arctic.

Grasslands are more vulnerable to plant invasions than forests in south-central Nepal

Understanding the patterns of plant invasions across vegetation types and identifying the underlying drivers for such patterns is essential to prioritize sites for the effective management of invasive species. We sampled three forest types (Shorea robusta, Dalbergia sissoo, and Tropical deciduous riverine forest) and two grasslands in south-central Nepal to evaluate the richness of naturalized and invasive alien plant species (IAPS). Species richness of naturalized species and the IAPS among vegetation types within each physiographic region was compared using analysis of variance and t-test, and the ordination was used to analyze species-environment relations. Of the total 327 species of vascular plants recorded, 14 % were naturalized, and over a third of the naturalized species were invasive. Grasslands had the highest richness of naturalized species and the IAPS. In contrast, S. robusta forest had the lowest naturalized and IAPS richness. Naturalized species richness increased with native species richness. Ordination revealed that a high number of IAPS were associated with areas exposed to frequent fire and livestock grazing and close to human settlements. The results suggest that the grasslands closer to the human settlements and exposed to frequent disturbance are vulnerable to the establishment of IAPS. Early detection and control of IAPS in such frequently disturbed grasslands would prevent their spread into the surrounding landscapes.

Predicting the current fishable habitat distribution of Antarctic toothfish (Dissostichus mawsoni) and its shift in the future under climate change in the Southern Ocean.

Global warming continues to exert unprecedented impacts on marine habitats. Species distribution models (SDMs) are proven powerful in predicting habitat distribution for marine demersal species under climate change impacts. The Antarctic toothfish, Dissostichus mawsoni (Norman 1937), an ecologically and commercially significant species, is endemic to the Southern Ocean. Utilizing occurrence records and environmental data, we developed an ensemble model that integrates various modelling techniques. This model characterizes species-environment relationships and predicts current and future fishable habitats of D. mawsoni under four climate change scenarios. Ice thickness, depth and mean water temperature were the top three important factors in affecting the distribution of D. mawsoni. The ensemble prediction suggests an overall expansion of fishable habitats, potentially due to the limited occurrence records from fishery-dependent surveys. Future projections indicate varying degrees of fishable habitat loss in large areas of the Amery Ice Shelf's eastern and western portions. Suitable fishable habitats, including the spawning grounds in the seamounts around the northern Ross Sea and the coastal waters of the Bellingshausen Sea and Amundsen Sea, were persistent under present and future environmental conditions, highlighting the importance to protect these climate refugia from anthropogenic disturbance. Though data deficiency existed in this study, our predictions can provide valuable information for designing climate-adaptive development and conservation strategies in maintaining the sustainability of this species.

Dealing with area‐to‐point spatial misalignment in species distribution models

Species distribution models (SDMs) are extensively used to estimate species–environment relationships (SERs) and predict species distribution across space and time. For this purpose, it is key to choose relevant spatial grains for predictor and response variables at the onset of the modelling process. However, environmental variables are often derived from large‐scale climate models at a grain that can be coarser than the one of the response variable. Such area‐to‐point spatial misalignment can bias estimates of SER and jeopardise the robustness of predictions. We used a virtual species approach, running simulations across different levels of area‐to‐point spatial misalignment to seek statistical solutions to this problem. We specifically compared accuracy of SER estimates and predictive performances, assessed across different degrees of spatial heterogeneity in environmental conditions, of three SDMs: a GLM, a spatial GLM and a Berkson error model (BEM) that accounts for fine‐grain environmental heterogeneity within coarse‐grain cells. Only the BEM accurately estimates SER from relatively coarse‐grain environmental data (up to 50 times coarser than the response grain), while the two GLMs provide flattened SER. However, all three models perform poorly when predicting from coarse‐grain data, particularly in environments that are more heterogeneous than the training conditions. Conversely, decreasing environmental heterogeneity relative to the training dataset reduces the predictive biases. Because predictions are made from covariate‐grain data, the BEM displays lower predictive performance than the two GLMs. Thus, standard model selection methods would fail to select the model that best estimates SERs (here, the BEM), which could lead to false interpretations about the environmental drivers of species distributions. Overall, we conclude that the BEM, because it can robustly estimate SER at the response grain, holds great promise to overcome area‐to‐point misalignment.

Distinct differences of vertical phytoplankton community structure in mainstream and a tributary bay of the Three Gorges Reservoir, China.

The vertical distribution of phytoplankton plays a crucial role in shaping the dynamics and structure of aquatic communities. In highly dynamic reservoir systems, water level fluctuations significantly affect the physiochemical conditions and the phytoplankton community. However, the specific effects on the vertical characteristics of phytoplankton between the mainstream and the tributary bay of the reservoir remain unstudied. This study investigated the vertical aspects of phytoplankton density, biomass, α and β diversity through monthly sampling over two years in the mainstream (Chang Jiang, CJ) and a tributary bay (Xiang Xi, XX) of the Three Gorges Reservoir in China. Phytoplankton density and biomass were significantly higher in XX, indicating an increased risk of algal blooms in the tributary. The phytoplankton community in CJ showed more stable species-environment relationships, a lower Shannon index and a higher evenness index, suggesting a relatively simple structure and a more uniform distribution of phytoplankton among different water layers. Conversely, XX showed greater differences between water layers (higher β diversity), with significant negative correlations with water level and positive correlations with DO difference, dissolved silica (DSi) difference, and stratification. Peak phytoplankton density and biomass, as well as high β diversity in XX, occurred during periods of decreased water levels with strong stratification in spring and summer. A structural equation model complemented by path analysis revealed that a decrease in water level could increase β diversity either directly through internal processes with extended residence time or indirectly by modifying stratification and the vertical distribution of DSi in XX. Therefore, a proposed water quality management strategy for XX was to increase the water level or reduce β diversity by implementing artificial mixing during stratification periods. Overall, this study lies in its comprehensive investigation of the vertical characteristics of the phytoplankton community in both the mainstream and the tributary bay of the Three Gorges Reservoir, elucidating the significant impact of water level fluctuations and providing insights for targeted water quality management strategies in the tributary bay to mitigate potential ecological impacts.

Explaining inter-individual differences in habitat relationships among wildcat hybrids in Scotland

Little is known about the factors that drive nonstationarity and inter-individual differences in realized habitat niches and species-environment relationships. We explored this topic by developing individual habitat selection models for 14 wildcat hybrids distributed across Scotland, and assessed how differences in their predicted probabilities of occurrence were related to factors including (1) geographic distance, (2) multivariate ecological distance, (3) difference in degree of hybridization and (4) difference in sex (male vs female). We found that the individual models were exceptionally effective in predicting the habitat use and occurrence of the particular individuals on whose data they were trained, but were generally highly divergent and not transferable among individuals. We conducted a reciprocal validation approach where we calculated the AUC for each individual model, predicting the occurrence patterns of the 13 other individuals. We then fit regression and nonparametric splines to evaluate the impacts of geographical distance, ecological distance, hybridization distance and difference in the sex of individuals in the ability of individual wildcat hybrid habitat models to predict the occurrences of other individuals. We found that, of the four factors assessed, ecological distance was supported as being inversely related to ability of a model from one individual to predict occurrence of another individual. The other three factors were not strongly related to differences in reciprocal model predictive ability. This suggests that ecological differences where individual wildcat hybrids reside drive differences in their habitat selection, but that geographical distance, degree of genetic hybridization and difference in the sex of individuals are not consistently associated with differences in model prediction or reciprocal validation performance. These results highlight the effect of ecological limiting factors, and the importance of nonstationary limiting factors in determining the habitat they select, their expressed species-environment relationship and the description of their realized habitat niches.

Going back for the future: Incorporating Pleistocene fossil records of saiga antelope into habitat suitability models

AbstractAimMany species have suffered anthropogenic range contraction and no longer occupy all available suitable environmental conditions. This is particularly problematic for the construction of habitat suitability models (HSMs), which assume that a species' contemporary range reflects its full species–environment relationship. HSMs therefore risk underestimating suitable environment areas, and misinforming conservation decisions. Incorporating historic (centuries‐old) records partly reduces this bias, but even these records are also subject to human disturbance. We incorporated fossil records of the critically endangered saiga antelope (Saiga tatarica, L., 1776), alongside historic and current records, into current and future habitat suitability models. Saiga has experienced drastic range contraction and may have a truncated species–environment relationship. The results allowed us to test whether its current habitat provides optimal environmental conditions, or whether saiga should be considered a refugee species.LocationNorthern Hemisphere.TaxonSaiga tatarica (Bovidae, Artiodactyla).MethodsWe collated historic and fossil saiga occurrence records from published literature, museum archives and global databases. Modern occurrence records were obtained from the International Union for Conservation of Nature Red List assessment. Four bioclimatic variables were downloaded from Worldclim.org. HSMs were generated through Maxent, using the maxnet package in R. Three HSMs were developed: present only, present historic and present fossil. Each of these models was projected onto current and two future (2070) climate change scenarios.ResultsSaiga fossil records increased the predicted suitable environment area by 783% and 1416% for current and future climate projections respectively. Our results suggest the saiga is not a refugee species but occupies only a portion of its potential environmental niche. The saiga's contemporary range is predicted environmentally suitable throughout all models and projections, and therefore in situ conservation management is recommended.Main ConclusionsThis study highlights the importance of incorporating fossil records into HSMs to better understand species–environment relationships and develop more robust conservation strategies for appropriate endangered species.

Modelling tools for including climate change in pest risk assessments

AbstractThis paper provides a comprehensive overview of the modelling tools available for integrating climate change impacts into pest risk assessments (PRA), elucidating the existing methodologies and models employed to understand the potential distributions of pests based on historical data and under future climate change scenarios. We highlight the strengths and weaknesses of these models and provide commentary on their ability to identify emerging threats due to climate change accurately and adequately, considering pest establishment likelihood, host crop exposure and the distribution of impacts. The simplest methods are based on climate‐matching models, degree‐day development models and Köppen–Geiger climate classification. Correlative species distribution models derive species–environment relationships and have been applied to PRA with mixed success. When fitted models are applied to different continents they are usually challenged to extrapolate climate suitability patterns outside the climate space used to train them. Global climate change is creating novel climates, exacerbating this ‘transferability’ problem. Some tools have been developed to reveal when these models are extrapolating. Process‐oriented models, which focus on mechanisms and processes rather than distribution patterns, are inherently more reliable for extrapolation to novel climates such as new continents and future climate scenarios. These models, however, require more skill and generally more knowledge of the species to craft robust models.

Quantifying agricultural land-use intensity for spatial biodiversity modelling: implications of different metrics and spatial aggregation methods

ContextAgricultural intensification is a major driver of farmland biodiversity declines. However, the relationship between land-use intensity (LUI) and biodiversity is complex and difficult to characterise, not least because of the difficulties in accurately quantifying LUI across heterogeneous agricultural regions.ObjectivesWe investigated how the use of different LUI metrics and spatial aggregation methods can lead to large variations in LUI estimation across space and thus affect biodiversity modelling.MethodsWe used three spatial aggregation methods (square, hexagonal, and voronoi grids) to calculate ten commonly used LUI metrics describing three LUI dimensions: land use, land management and landscape structure. Using a virtual species approach, we compared how LUI values sampled at biodiversity monitoring sites vary across different metrics and grids. We modelled the distribution of three virtual species using Generalised Additive Models to test how omitting certain LUI dimensions from the models affected the model results.ResultsThe density distributions of LUI values at the presence points of the virtual species were significantly different across metrics and grids. The predefined species-environment relationships characterising the environmental niches of two out of three virtual species remained undetected in models that omitted certain LUI dimensions.ConclusionsWe encourage researchers to consider the implications of using alternative grid types in biodiversity models, and to account for multiple LUI dimensions, for a more complete representation of LUI. Advances in remote sensing-derived products and increased accessibility to datasets on farm structure, land-use and management can greatly advance our understanding of LUI effects on biodiversity.

Guidelines for the use of spatially varying coefficients in species distribution models

AbstractAimSpecies distribution models (SDMs) are increasingly applied across macroscales using detection‐nondetection data. These models typically assume that a single set of regression coefficients can adequately describe species–environment relationships and/or population trends. However, such relationships often show nonlinear and/or spatially varying patterns that arise from complex interactions with abiotic and biotic processes that operate at different scales. Spatially varying coefficient (SVC) models can readily account for variability in the effects of environmental covariates. Yet, their use in ecology is relatively scarce due to gaps in understanding the inferential benefits that SVC models can provide compared to simpler frameworks.InnovationHere we demonstrate the inferential benefits of SVC SDMs, with a particular focus on how this approach can be used to generate and test ecological hypotheses regarding the drivers of spatial variability in population trends and species–environment relationships. We illustrate the inferential benefits of SVC SDMs with simulations and two case studies: one that assesses spatially varying trends of 51 forest bird species in the eastern United States over two decades and a second that evaluates spatial variability in the effects of five decades of land cover change on grasshopper sparrow (Ammodramus savannarum) occurrence across the continental United States.Main conclusionsWe found strong support for SVC SDMs compared to simpler alternatives in both empirical case studies. Factors operating at fine spatial scales, accounted for by the SVCs, were the primary divers of spatial variability in forest bird occurrence trends. Additionally, SVCs revealed complex species–habitat relationships with grassland and cropland area for grasshopper sparrow, providing nuanced insights into how future land use change may shape its distribution. These applications display the utility of SVC SDMs to help reveal the environmental factors that drive species distributions across both local and broad scales. We conclude by discussing the potential applications of SVC SDMs in ecology and conservation.

The influence of scale-dependent geodiversity on species distribution models in a biodiversity hotspot.

Improving models of species' distributions is essential for conservation, especially in light of global change. Species distribution models (SDMs) often rely on mean environmental conditions, yet species distributions are also a function of environmental heterogeneity and filtering acting at multiple spatial scales. Geodiversity, which we define as the variation of abiotic features and processes of Earth's entire geosphere (inclusive of climate), has potential to improve SDMs and conservation assessments, as they capture multiple abiotic dimensions of species niches, however they have not been sufficiently tested in SDMs. We tested a range of geodiversity variables computed at varying scales using climate and elevation data. We compared predictive performance of MaxEnt SDMs generated using CHELSA bioclimatic variables to those also including geodiversity variables for 31 mammalian species in Colombia. Results show the spatial grain of geodiversity variables affects SDM performance. Some variables consistently exhibited an increasing or decreasing trend in variable importance with spatial grain, showing slight scale-dependence and indicating that some geodiversity variables are more relevant at particular scales for some species. Incorporating geodiversity variables into SDMs, and doing so at the appropriate spatial scales, enhances the ability to model species-environment relationships, thereby contributing to the conservation and management of biodiversity. This article is part of the Theo Murphy meeting issue 'Geodiversity for science and society'.