http://www.bio-nica.info/Biblioteca/Anderson2003PredictiveModels.pdf

Evaluating predictive models of species’ distributions: criteria for selecting optimal models

Summary 1. Few examples of habitat-modelling studies of rare and endangered species exist in the literature, although from a conservation perspective predicting their distribution would prove particularly useful. Paucity of data and lack of valid absences are the probable reasons for this shortcoming. Analytic solutions to accommodate the lack of absence include the ecological niche factor analysis (ENFA) and the use of generalized linear models (GLM) with simulated pseudo-absences. 2. In this study we tested a new approach to generating pseudo-absences, based on a preliminary ENFA habitat suitability (HS) map, for the endangered species Eryngium alpinum . This method of generating pseudo-absences was compared with two others: (i) use of a GLM with pseudo-absences generated totally at random, and (ii) use of an ENFA only. 3. The influence of two different spatial resolutions (i.e. grain) was also assessed for tackling the dilemma of quality (grain) vs. quantity (number of occurrences). Each combination of the three above-mentioned methods with the two grains generated a distinct HS map. 4. Four evaluation measures were used for comparing these HS maps: total deviance explained, best kappa, Gini coefficient and minimal predicted area (MPA). The last is a new evaluation criterion proposed in this study. 5. Results showed that (i) GLM models using ENFA-weighted pseudo-absence provide better results, except for the MPA value, and that (ii) quality (spatial resolution and locational accuracy) of the data appears to be more important than quantity (number of occurrences). Furthermore, the proposed MPA value is suggested as a useful measure of model evaluation when used to complement classical statistical measures. 6. Synthesis and applications. We suggest that the use of ENFA-weighted pseudoabsence is a possible way to enhance the quality of GLM-based potential distribution maps and that data quality (i.e. spatial resolution) prevails over quantity (i.e. number of data). Increased accuracy of potential distribution maps could help to define better suitable areas for species protection and reintroduction.

/pdf/an-improved-approach-for-predicting-the-distribution-of-rare-43colkkrya.pdf

An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo-absence data

For many applications the continuous prediction afforded by species distribution modeling must be converted to a map of presence or absence, so a threshold probability indicative of species presence must be fixed. Because of the bias in probability outputs due to frequency of presences (prevalence), a fixed threshold value, such as 0.5, does not usually correspond to the threshold above which the species is more likely to be present. In this paper four threshold criteria are compared for a wide range of sample sizes and prevalences, modeling a virtual species in order to avoid the omnipresent error sources that the use of real species data implies. In general, sensitivity–specificity difference minimizer and sensitivity–specificity sum maximizer criteria produced the most accurate predictions. The widely-used 0.5 fixed threshold and Kappa-maximizer criteria are the worst ones in almost all situations. Nevertheless, whatever the criteria used, the threshold value chosen and the research goals that determined its choice must be stated.

Threshold criteria for conversion of probability of species presence to either–or presence–absence

Aim To assess the geographical transferability of niche-based species distribution models fitted with two modelling techniques.

Location  Two distinct geographical study areas in Switzerland and Austria, in the subalpine and alpine belts.

Methods  Generalized linear and generalized additive models (GLM and GAM) with a binomial probability distribution and a logit link were fitted for 54 plant species, based on topoclimatic predictor variables. These models were then evaluated quantitatively and used for spatially explicit predictions within (internal evaluation and prediction) and between (external evaluation and prediction) the two regions. Comparisons of evaluations and spatial predictions between regions and models were conducted in order to test if species and methods meet the criteria of full transferability. By full transferability, we mean that: (1) the internal evaluation of models fitted in region A and B must be similar; (2) a model fitted in region A must at least retain a comparable external evaluation when projected into region B, and vice-versa; and (3) internal and external spatial predictions have to match within both regions.

Results  The measures of model fit are, on average, 24% higher for GAMs than for GLMs in both regions. However, the differences between internal and external evaluations (AUC coefficient) are also higher for GAMs than for GLMs (a difference of 30% for models fitted in Switzerland and 54% for models fitted in Austria). Transferability, as measured with the AUC evaluation, fails for 68% of the species in Switzerland and 55% in Austria for GLMs (respectively for 67% and 53% of the species for GAMs). For both GAMs and GLMs, the agreement between internal and external predictions is rather weak on average (Kulczynski's coefficient in the range 0.3–0.4), but varies widely among individual species. The dominant pattern is an asymmetrical transferability between the two study regions (a mean decrease of 20% for the AUC coefficient when the models are transferred from Switzerland and 13% when they are transferred from Austria).

Main conclusions  The large inter-specific variability observed among the 54 study species underlines the need to consider more than a few species to test properly the transferability of species distribution models. The pronounced asymmetry in transferability between the two study regions may be due to peculiarities of these regions, such as differences in the ranges of environmental predictors or the varied impact of land-use history, or to species-specific reasons like differential phenotypic plasticity, existence of ecotypes or varied dependence on biotic interactions that are not properly incorporated into niche-based models. The lower variation between internal and external evaluation of GLMs compared to GAMs further suggests that overfitting may reduce transferability. Overall, a limited geographical transferability calls for caution when projecting niche-based models for assessing the fate of species in future environments.

/pdf/are-niche-based-species-distribution-models-transferable-in-4dvwvg3i54.pdf

Are niche-based species distribution models transferable in space?

http://users.clas.ufl.edu/mbinford/GEO5159_GIS_Environmental/References_for_links/Beaumont.etal_2005_BIOCLIM.Species.Dist.pdf

Predicting species distributions: use of climatic parameters in BIOCLIM and its impact on predictions of species’ current and future distributions

A first exploration of applications of ecological niche modeling and geographic distributional prediction to endangered species protection is developed. Foci of richness of endangered bird species are identified in coastal California and along the southern fringe of the United States. Species included on the Endangered Species List on the basis of peripheral populations inflate these concentrations considerably. Species without protection in the US National Park System are focused particularly in peninsular Florida. Application of this methodology to additional taxa and regions holds promise for diverse conservation applications.

Mandaline E. Godown

Papers

Preliminary distributional analysis of US endangered bird species