Does a correlation exist between environmental suitability models and plant population parameters? An experimental approach to measure the influence of disturbances and environmental changes

Due to increasing human pressures, there is a need to understand how environmental and anthropogenic disturbances could affect the conservation status of endangered plant species. When information on distribution or population parameters is poor, Species Distribution Models (SDMs) may offer a valuable additional source to assess threats and to evaluate conservation options. In this sense, if the output of SDMs represents the relationships between habitat features and species occurrence, SDM results can also be related to other key parameters of population. For the endangered yellow gentian, we tested the relationship between six field population parameters and the suitabilities obtained by SDMs with natural and limiting parameters (i.e. proxies of disturbances). Specifically, the six population parameters were the surface area covered by each population, the number of vegetative and reproductive individuals per population, the density of reproductive and all individuals per plot and the proportion of reproductive individuals. Thus, threats were evaluated by testing if relationships between population parameters and suitabilities increased when proxies of disturbances were included in models. Best-fitting models resulted when all natural and human disturbance variables were included. In addition, results show relationships between suitability and population parameters only when disturbance parameters were used for SDMs. When the index related to the sensitivity to climate change was included in SDMs, the density of all individuals and number of reproductive plants were lower than in low suitability sites, suggesting that climate change is likely already challenging the ability of yellow gentian to bloom and germinate. Otherwise, we observed a decrease of the extent of localities in areas with higher suitability obtained through the index related to trail intensity. This confirmed the positive effect of conservation activities, which were mainly implemented in the proximity of trails. Using a thoroughly studied plant species as a straightforward example, we showed the potentiality of SDMs to inform on population parameters instead of only discriminate species presence or absence. This information can suggest the use of disturbance parameters when specific SDMs aim to support strategic decisions in management and conservation.