Dataset Information

The effects of sampling bias and model complexity on the predictive performance of MaxEnt species distribution models.

ABSTRACT: Species distribution models (SDMs) trained on presence-only data are frequently used in ecological research and conservation planning. However, users of SDM software are faced with a variety of options, and it is not always obvious how selecting one option over another will affect model performance. Working with MaxEnt software and with tree fern presence data from New Zealand, we assessed whether (a) choosing to correct for geographical sampling bias and (b) using complex environmental response curves have strong effects on goodness of fit. SDMs were trained on tree fern data, obtained from an online biodiversity data portal, with two sources that differed in size and geographical sampling bias: a small, widely-distributed set of herbarium specimens and a large, spatially clustered set of ecological survey records. We attempted to correct for geographical sampling bias by incorporating sampling bias grids in the SDMs, created from all georeferenced vascular plants in the datasets, and explored model complexity issues by fitting a wide variety of environmental response curves (known as "feature types" in MaxEnt). In each case, goodness of fit was assessed by comparing predicted range maps with tree fern presences and absences using an independent national dataset to validate the SDMs. We found that correcting for geographical sampling bias led to major improvements in goodness of fit, but did not entirely resolve the problem: predictions made with clustered ecological data were inferior to those made with the herbarium dataset, even after sampling bias correction. We also found that the choice of feature type had negligible effects on predictive performance, indicating that simple feature types may be sufficient once sampling bias is accounted for. Our study emphasizes the importance of reducing geographical sampling bias, where possible, in datasets used to train SDMs, and the effectiveness and essentialness of sampling bias correction within MaxEnt.

SUBMITTER: Syfert MM

PROVIDER: S-EPMC3573023 | biostudies-literature | 2013

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

The effects of sampling bias and model complexity on the predictive performance of MaxEnt species distribution models.

Syfert Mindy M MM Smith Matthew J MJ Coomes David A DA

PloS one 20130214 2

Species distribution models (SDMs) trained on presence-only data are frequently used in ecological research and conservation planning. However, users of SDM software are faced with a variety of options, and it is not always obvious how selecting one option over another will affect model performance. Working with MaxEnt software and with tree fern presence data from New Zealand, we assessed whether (a) choosing to correct for geographical sampling bias and (b) using complex environmental response ...[more]

PMID: 23457462

Similar Datasets

Project description:BackgroundPredicting species' potential geographical range by species distribution models (SDMs) is central to understand their ecological requirements. However, the effects of using different modeling techniques need further investigation. In order to improve the prediction effect, we need to assess the predictive performance and stability of different SDMs.MethodologyWe collected the distribution data of five common tree species (Pinus massoniana, Betula platyphylla, Quercus wutaishanica, Quercus mongolica and Quercus variabilis) and simulated their potential distribution area using 13 environmental variables and six widely used SDMs: BIOCLIM, DOMAIN, MAHAL, RF, MAXENT, and SVM. Each model run was repeated 100 times (trials). We compared the predictive performance by testing the consistency between observations and simulated distributions and assessed the stability by the standard deviation, coefficient of variation, and the 99% confidence interval of Kappa and AUC values.ResultsThe mean values of AUC and Kappa from MAHAL, RF, MAXENT, and SVM trials were similar and significantly higher than those from BIOCLIM and DOMAIN trials (p<0.05), while the associated standard deviations and coefficients of variation were larger for BIOCLIM and DOMAIN trials (p<0.05), and the 99% confidence intervals for AUC and Kappa values were narrower for MAHAL, RF, MAXENT, and SVM. Compared to BIOCLIM and DOMAIN, other SDMs (MAHAL, RF, MAXENT, and SVM) had higher prediction accuracy, smaller confidence intervals, and were more stable and less affected by the random variable (randomly selected pseudo-absence points).ConclusionsAccording to the prediction performance and stability of SDMs, we can divide these six SDMs into two categories: a high performance and stability group including MAHAL, RF, MAXENT, and SVM, and a low performance and stability group consisting of BIOCLIM, and DOMAIN. We highlight that choosing appropriate SDMs to address a specific problem is an important part of the modeling process.

Project description:BackgroundTsetse-transmitted African trypanosomiasis is a debilitating and fatal disease of humans and livestock if left untreated. While knowledge of the spatial distribution patterns of tsetse is essential for the development of risk-based vector control strategies, existing distribution maps in Zambia are more than 40 years old and were based on coarse spatial resolution data. The recently developed vehicle-mounted sticky trap (VST) provides an alternative sampling device to aid in updating existing distribution maps but has not been applied outside an experimental setting and is limited to motorable tracks. Therefore, the objective of the present study was to demonstrate the effectiveness of utilizing the VST for area-wide surveys of Glossina morsitans and to use the occurrence records to predict its spatial distribution in Zambia under current environmental conditions using Maxent.Methodology/principal findingsTwo-sided all-blue VST baited with butanone and 1-octen-3-ol was used to survey 692 and 1020 km of transect routes in G. m. centralis Machado and G. m. morsitans Westwood previously published distribution in Zambia. Maxent species distribution technique was used to predict the potential distribution of the two subspecies using current climatic and environmental data which was then compared to the historical distribution. A total of 15,602 tsetse were captured with G. m. morsitans (58%) being the most abundant. G. m. centralis and G. pallidipes Austin represented 39 and 2% of the catch respectively, and G. brevipalpis Newstead was also detected. The predicted potential distribution for G. m. centralis was 80,863 km2 while that of G. m. morsitans was 70,490 km2 representing a 47 and 29% reduction compared to their historical distributions, respectively.Conclusion/significanceThe VST is effective for sampling G. morsitans outside experimental settings and is recommended for use as an additional tsetse survey tool. The spatial distribution of G. morsitans in Zambia has reduced by 101,051 km2 due to temperature and land cover changes.

Project description:In the context of global warming, the habitats of Ephedra, including Ephedra sinica Stapf, Ephedra intermedia Schrenk ex Mey, and Ephedra equisetina Bunge, have been substantially threatened and deteriorated in recent years. Little is known about the potential geographic dynamics of economically renowned species, including those used in sand fixation and traditional Chinese medicine, under climate change. Therefore, evaluating their potential habitat and determining the crucial environmental variables affecting E. sinica, E. intermedia and E. equisetina under the driving force of global warming are extremely important. In this study, an optimized MaxEnt model in the kuenm package on the basis of occurrence records (a total of 103, 101 and 97 points for E. sinica, E. intermedia and E. equisetina, respectively) and 37 environmental factors were utilized to simulate the distribution of the three species. Two representative concentration pathways (SSP2.6 and SSP8.5) at 2041-2060 and 2061-2080, respectively, were used to establish a future distribution model of the three species. The results indicated that approximately 6.92 × 105 km2, 2.95 × 105 km2, and 11.5 × 105 km2 of suitable regions for E. sinica, E. intermedia and E. equisetina were obtained, which were mostly distributed in central and eastern Inner Mongolia, eastern and southern Gansu, and northern Xinjiang, respectively. Critical environmental variables, such as land cover and annual precipitation, were regarded as critical parameters for the three species. Future assessment revealed that over 60 % of the potential distribution area was affected, and the stability of E. sinica under the SSP8.5 scenario was the greatest. The spatial dynamic changes in suitable areas for E. intermedia were smaller than those for E. equisetina and E. sinica in the future. The comprehensive analysis revealed that the fluctuations in the distributions of the three Ephedra species under climate change are small and provide useful information for future conservation. Therefore, target conservation and management measures should be implemented in combination with the suitability thresholds of different environmental parameters. Our results provide useful recommendations for the current and future protection of Ephedra populations.

Project description:Loss of habitat, specifically deforestation, is a major driver of biodiversity loss. Species-area relationship (SAR) models traditionally have been used for estimating species richness, species loss as a function of habitat loss, and extrapolation of richness for given areas. Sampling-species relationships (SSRs) are interrelated yet separate drivers for species richness estimates. Traditionally, however, SAR and SSR models have been used independently and not incorporated into a single approach. We developed and compared predictive models that incorporate sampling effort species-area relationships (SESARS) along the entire Atlantic Forest of South America, and then applied the best-fit model to estimate richness in forest remnants of Interior Atlantic Forest of eastern Paraguay. This framework was applied to non-volant small mammal assemblages that reflect different tolerances to forest loss and fragmentation. In order to account for differences in functionality we estimated small mammal richness of 1) the entire non-volant small mammal assemblage, including introduced species; 2) the native species forest assemblage; and 3) the forest-specialist assemblage, with the latter two assemblages being subsets of the entire assemblage. Finally, we geospatially modeled species richness for each of the three assemblages throughout eastern Paraguay to identify remnants with high species richness. We found that multiple regression power-law interaction-term models that only included area and the interactions of area and sampling as predictors, worked best for predicting species richness for the entire assemblage and the native species forest assemblage, while several traditional SAR models (logistic, power, exponential, and ratio) best described forest-specialist richness. Species richness was significantly different between assemblages. We identified obvious remnants with high species richness in eastern Paraguay, and these remnants often were geographically isolated. We also found relatively high predicted species richness (in relation to the entire range of predicted richness values) in several geographically-isolated, medium-size forest remnants that likely have not been considered as possible priority areas for conservation. These findings highlight the importance of using an empirical dataset, created using sources representing diverse sampling efforts, to develop robust predictive models. This approach is particularly important in geographic locations where field sampling is limited yet the geographic area is experiencing rapid and dramatic land cover changes. When combined, area and sampling are powerful modeling predictors for questions of biogeography, ecology, and conservation, especially when addressing habitat loss and fragmentation.

Project description:Experts use knowledge to infer the distribution of species based on fuzzy logical assumptions about the relationship between species and the environment. Thus, expert knowledge is amenable to fuzzy logic modelling, which give to propositions a continuous truth value between 0 and 1. In species distribution modelling, fuzzy logic may also be used to model, from a number of records, the degree to which conditions are favourable to the occurrence of a species. Therefore, fuzzy logic operations can be used to compare and combine models based on expert knowledge and species records. Here, we applied fuzzy logic modelling to the distribution of amphibians in Uruguay as inferred from expert knowledge and from observed records to infer favourable locations, with favourability being the commensurable unit for both kinds of data sources. We compared the results for threatened species, species considered by experts to be ubiquitous, and non-threatened, non-ubiquitous species. We calculated the fuzzy intersection of models based on both knowledge sources to obtain a unified prediction of favourable locations. Models based on expert knowledge involved a larger number of variables and were less affected by sampling bias. Models based on experts had the same overprediction rate for the three types of species, whereas models based on species records had a lower prediction rate for ubiquitous species. Models based on expert knowledge performed equally as well or better than corresponding models based on species records for threatened species, even when they had to discriminate and classify the same set of records used to build the models based on species records. For threatened species, expert models predicted more restrictive favourable territories than those predicted based on records. Observed records generated the best-fitted models for non-threatened non-ubiquitous species, and ubiquitous species. Fuzzy modelling permitted the objective comparison of the potential of expert knowledge and incomplete distribution records to infer the territories favourable for different species. Distribution of threatened species was able to be better explained by subjective expert knowledge, while for generalist species models based on observed data were more accurate. These results have implications for the correct use of expert knowledge in conservation planning.

Dataset Information

The effects of sampling bias and model complexity on the predictive performance of MaxEnt species distribution models.

Publications

The effects of sampling bias and model complexity on the predictive performance of MaxEnt species distribution models.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets