High-resolution species distribution models for two ecologically contrasting, closely related plant species

Domenico Amantea ¹, Francesco Cristiano ², Alessandro Petraglia ², Domenico Gargano ¹

Department of Biology, Ecology and Earth Sciences, University of Calabria, Via Pietro Bucci, Rende (CS), 87033, Italy ¹, Department of Chemistry, Life Sciences and Environmental Sustainability, University of Parma, Str. dell'Università, Parma (PR), 43121, Italy ²

Open habitats, such as grasslands, often form complex mosaics, shaped by factors including highly localized ecological conditions, variations in substrates and topography. These settings often express remarkable biodiversity even on very narrow scales; however, these systems are easily threatened by global change components causing fragmentation and disruption of natural systems. High-resolution ecological niche modelling can be crucial in identifying the environmental factors responsible for small-scale patterns of species distribution. Furthermore, these models can be integrated in conservation frameworks for assessing species vulnerability to global change. In this study, we wanted to investigate the ecological niche of two closely related woodrush species, Luzula campestris subsp. campestris and L. calabra (a Calabrian endemism). More specifically, the purpose was to verify the power of high-resolution remote-sensed data in discriminating the spatial envelope of species associated with different environmental conditions (i.e., dry pastures vs. wetlands). To achieve this, we carried out a sampling campaign in Macchia Sacra, a Special Area of Conservation within the Sila National Park (Calabria, Italy) characterized by a complex mosaic of marshlands, wetlands and dry pastures. The sampling process involved the collection of presence records for the two species, as well as vegetation data. For the species distribution models, we used a set of spectral indices derived from Sentinel-2 raw data and UAV-based topographic variables. The analyses on vegetation data confirmed that the two species act as ecological vicariants within the study area. The species distribution models demonstrated high accuracy in distinguishing the niches of the two species, highlighting the potential of RS data for assessing species distribution at high resolutions.

Main author career stage: PhD student

Contribution type: Talk

First choice session: 3. Biodiversity and global change

Second choice session: 2. Ecology