Spatial analysis improves species distribution modelling during range expansion

Departamento de Biologia Geral, ICB, Universidade Federal de Goiás, 74001-970 Goiânia, GO, Brazil.
Biology letters (Impact Factor: 3.25). 11/2008; 4(5):577-80. DOI: 10.1098/rsbl.2008.0210
Source: PubMed


Species distribution models (SDMs) assume equilibrium between species' distribution and the environment. However, this assumption can be violated under restricted dispersal and spatially autocorrelated environmental conditions. Here we used a model to simulate species' ranges expansion under two non-equilibrium scenarios, evaluating the performance of SDM coupled with spatial eigenvector mapping. The highest fit is for the models that include space, although the relative importance of spatial variables during the range expansion differs in the two scenarios. Incorporating space to the models was important only under colonization-lag non-equilibrium, under the expected scenario. Thus, mechanisms that generate range cohesion and determine species' distribution under climate changes can be captured by spatial modelling, with advantages compared with other techniques and in line with recent claims that SDMs have to account for more complex dynamic scenarios.

Download full-text


Available from: Jose Alexandre Felizola Diniz-Filho,
  • Source
    • "Both models attempt to predict the spatial distribution of plant species diversity using a set of soil and topographic variables, but only the latter incorporates the effect of SAC. As in the previous studies, we expect that the spatial model would show better performances than its non-spatial counterpart with regard to both predictive power and independence among residuals (e.g., Augustin et al., 1996; Bini et al., 2009; Borcard and Legendre, 2002; de Marco et al., 2008; Dormann, 2007; Dray et al., 2006; Lichstein et al., 2002). Second, we will examine the effect of incorporating SAC on partial regression coefficients. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In the literature of ecological indicators, more attention has yet to be paid to the potential effects of spatial autocorrelation (SAC) on the prediction of plant community structure. At a selected foredune ridge in a temperate coast of South Korea, this research developed two regression models: (1) a non-spatial ordinary least squares (OLS) in which the fine-scale (ca. 10 m) variability of diversity was predicted by soil and topographic parameters and (2) a spatial model in which spatial filters extracted by spatial eigenvector mapping were incorporated as additional predictors into the original OLS. After such incorporation, a reduction apparently occurred in the predictive power of the environmental variables, especially those with an inherently high amount of SAC. For example, Mg2+ was the most significant predictor for species diversity in OLS, but it became insignificant in spatial regression. This indicates that, during the incorporation of SAC, the predictive importance of Mg2+ was replaced by that of spatial filters. In other words, because the SAC of Mg2+ was inherently strong (global Moran's I = 0.68, p < 0.001), this soil attribute became redundant when the spatial filters were added to the non-spatial OLS. These discussions corroborate the general idea of this paper that SAC potentially indicates the degree of shifts in the predictive power of environmental factors for plant diversity. In sum, we suggest that environmental variables, which are highly structured over space, should be the target of special attention and care in future modeling attempts aiming to predict the spatial patterns of plant species diversity in coastal dunes. This fine-scale approach can also be applied to macroecological studies along a variety of ecological systems, spanning latitudinal or disturbance gradients.
    Ecological Indicators 01/2016; 60:1130-1141. DOI:10.1016/j.ecolind.2015.09.021 · 3.44 Impact Factor
  • Source
    • "For each validation, we analyzed (i) the area under the ROC curves (AUC; Fawcett 2004; Ko et al. 2011), (ii) the correct classification rate (CCR; Ahmadi et al. 2013), (iii) the Cohen's kappa (k) (Manel et al. 2001), and (iv) the Boyce's index (B) (Boyce et al. 2002; Jones-Farrand et al. 2011). We tested for residual spatial autocorrelation with Moran's I correlogram (1—predicted values for each location; De Marco et al. 2008). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Non-invasive genetic sampling has been used to reconstruct spatial patterns of carnivore distributions, identify regions where conflicts with human activities could threaten the survival of a species, and assess the effectiveness of conservation strategies. In this study, we used detailed information on wolf (Canis lupus) and livestock distributions to infer depredation risks in a wide area of the Italian Apennines. We carried out a General Niche Environment System Factor Analysis (GNESFA) to define the potential distribution of wolves genotyped from 8565 samples collected during 12 years of non-invasive genetic monitoring in 3622 locations. Habitat suitability models indicated that the proportion of meadows, altitude, slope, roughness, and distance from human settlements were the main factors positively related to the potential wolf distribution, in contrast with the extension of cultivated fields and human settlements. Results of GNESFA were used to infer the local depredation risk, which was high in 46.9 % of the pastures, and to rank the areas where prevention tools should be used with priority. In this way, the use of often-limited financial resources for prevention could be promoted in pastures with the highest depredation risk and conflicts between husbandry and wolf presence might be mitigated.
    European Journal of Wildlife Research 07/2015; DOI:10.1007/s10344-015-0942-4 · 1.63 Impact Factor
  • Source
    • "(Wikle 2003). On the other hand, classical niche modeling outcomes are limited by the non-temporal, static structure of the modeling (Araújo and Pearson 2005; Marco et al. 2008; Elith et al. 2010). The approach of dynamic occupancy models used here is more mechanistic than niche modeling, accounts for the temporal component of the diffusion process and at the same time integrates the heterogeneity of this process induced by external factors, either environmental or anthropogenic. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Species distribution models (SDM) have often been used to predict the potential ranges of introduced species and prioritize management strategies. However, this approach assumes equilibrium between occurrences and environmental gradients, an assumption which is violated during the invasion process, where many suitable sites are empty because the species has not yet reached them. Here we considered the invasive ladybird Harmonia axyridis as a case study to show the benefits of using a dynamic colonization-extinction model that does not assume equilibrium. We used a multi-year occupancy model incorporating environmental, anthropogenic and neighborhood effects, to identify factors that explained spreading variation of this species in France from 2004, when only a few occupied sites were detected, to 2011. We found that anthropogenic factors (urbanization, agriculture, vineyards, and presence/absence of highways) explained more variation in the diffusion process than environmental factors (winter and summer temperatures, wind-speed, and rainfall). The surface of urbanization was the major anthropogenic factor increasing the probability of colonization. The average summer temperature was the main environmental factor affecting colonization, with a negative effect when high or low. The neighborhood effect revealed that colonization was mostly influenced by contributions coming from a radius of 24 km around the focal cell. The contribution of neighborhood decreases over time, suggesting that H. axyridis is reaching its equilibrium in France. This is confirmed by the small discrepancy observed between the performance of our approach and a SDM approach when predicting a single year occupancy pattern at the end of the study period. Our approach has the advantage of explicitly modelling the state of the biological system during the spatial expansion and identifying colonization constraints. This allows managers to explore the effect of different actions on the system at key moments of the invasion process, hence providing a powerful approach to prioritize management strategies.
    Ecography 07/2015; DOI:10.1111/ecog.01389 · 4.77 Impact Factor
Show more