In this study, we used three genetic parameters (Ne, He, and I) as assessment criteria for genetic diversity (Figs. 5, 6). Ne was used to reflect the size of the genetic variation within the population, He as a surrogate of genetic diversity within the population, and I as a measure of biodiversity in the ecological system (Yan et al. 2008; Van Zonneveld et al. 2012). Because one parameter alone is insufficient to fully describe genetic diversity, we used all three parameters together to identify those populations with low genetic variation that are therefore in urgent need of protection. The group with low genetic diversity includes Pop 2, 4, 8, and 10, and the other populations belong to the group with high genetic diversity.
Moreover, in this study, the value of Nm is 2.089, indicating that gene flow exists between populations (Li et al. 2013). A value of F close to 0 is suitable for in situ conservation because of the inbreeding between populations (an F value close to 1 meaning there is not one genetic exchange between populations), and values much lower than 0 are suitable for ex situ measures due to an excess of heterozygosity, which could determine the methods of protection from the perspective of gene flow (Van Zonneveld et al. 2012). Hence, we think a negative F value (clearly below 0) requires special attention. As shown in Figure 4A and Table S4, populations 2, 4, 8, and 10 (with clearly negative F values) are suitable for ex situ conservation, while populations 7, 12, and 16 (moderate F values) are most suited to in situ conservation.
Habitat suitability was directly associated with the occurrence probability of the species and would be expected to change with climate change (Dubey et al. 2013); Figures 2, 3 illustrate these changes, and this type of map is useful for the visualization of those sites most in need of study and protection.
Moreover, the genetic variance parameters Ne, He, and I have a significantly positive geospatial correlation with habitat suitability, as measured using binominal regression analysis (Fig. 5). Hence, the maintenance and conservation of habitat suitability is important for the maintenance of genetic diversity (Razgour et al. 2011). Consideration of genetic diversity alone is insufficient, however, because habitat suitability is also vital in the evaluation criteria; therefore, we need to assess the habitat suitability of species. Previous studies that have assessed the responses of species to climate change have shown that individuals move to habitats that are suitable for their maintenance (Kramer et al. 2010; Sork et al. 2010; Collevatti et al. 2011; Alsos et al. 2012; Brown and Knowles 2012). Hence, we regarded the presence of the species (i.e., Maxent values ≥0.5) as a precondition of protection of genetic diversity because the plant populations are carriers of a genotype adapted to climate change (Collevatti et al. 2011). We found that when the Maxent value was below 0.5, genetic diversity dropped precipitously (Fig. 5). We believe that when genetic diversity is too high or too low and is unstable, genetic diversity can impede population viability. As genetic diversity responds with a lag to changes in habitat suitability, an irregular trend in habitat suitability may not result in well-adapted genetic diversity. A population with a high and stable genetic diversity might be more tolerant to rapid climate change. If the change in climate is drastic, while the genotype might change until adapted to the new habitat, this would take a long time and need the creation of a stable environment in which the species can evolve to prevent extinction (in this study, we set this condition as the threshold of Maxent, that is, ≥0.5; Chevin et al. 2010; Dawson et al. 2011; Hoffmann and Sgrò 2011). Although our data showed that habitat suitability is likely to decrease, it would only require small changes to meet the requirements to protect, for example, Pop 6's genetic diversity. The habitat suitability of Pops 2, 4, 7, 8, 11, 12, 13, and 16 are still below 0.5, so ex situ conservation that we transfer these populations away from less suitable habitats to highly suitable habitats is required by default (Fig. 4). In some cases, where habitat suitability is likely to be above 0.5 in some scenarios and the population has low genetic diversity, a more adaptive approach is required. For example, for Pop 10, in situ conservation in suitable habitats could be adopted as the primary plan with ex situ conservation in unsuitable habitats as a subsidiary plan in response to environmental deterioration due to climate change. Populations with less certain future habitat suitability would be protected through in situ conservation with habitat monitoring, for instance, Pop 1, 3, 5, 6, 9, and 10. Hence, we should use ex situ conservation to transfer these populations away from less suitable habitats to highly suitable habitats. Habitat suitability needs to be considered along with different future emissions scenarios to plan in situ or ex situ conservation approaches. Clearly, populations with low genetic diversity would persistently suffer in areas of low habitat suitability (where the Maxent values are much lower than 0.5); hence, these should be subjected to in situ conservation to prevent species extinction. In the current study, we found that special attention needs to be paid to the populations with unstable habitat suitability, such as Pop 14 and 15, which need to be actively monitored by conservationists (Fig. 4; Minteer and Collins 2010; Sgro et al. 2011).