Modelling habitat suitability of Pinguicula crystallina sibth. & sm. subsp. hirtiflora (Ten.) Strid (Lentibulariaceae) in Campania (Southern Italy) through an expert based approach.

In the last few decades, modelling species’ distribution and habitat suitability has greatly improved as a result of new powerful statistics and GIS tools which can use a wide number of ecological variables. Species distribution models represent an important technique for the conservation of biodiversity and for town-and-country planning (Guisan & Zimmermann, 2000). Rare and threatened plant species, particularly the ones with known scattered distribution, can be of highest interest in modelling to assess conservation priorities and evaluate habitat suitability. Pinguicula crystallina Sibth. & Sm. subsp. hirtiflora (Ten.) Strid (Lentibulariaceae) is a carnivorous plant which has sticky leaves used as flypaper-type traps and grows on dripping carbonate or serpentine rocks primarily in mountain ecosystems such as wet cliffs inside ravines which can be mainly ascribed to Natura 2000 priority habitat 7220, which is petrifying springs with tufa formation (Cratoneurion W. Koch). Current distribution of this taxon is in South-Eastern Europe (mainly Albania and Greece), with only few populations living in Southern Italy (Campania and Calabria) (Innangi et al. 2011). Many studies about karyology (Casper & Stimper, 2009) and molecular data (Degtjareva et al., 2006), suggest that P. crystallina subsp. hirtiflora along with P. crystallina could be the sister group of all the other species in the genus Pinguicula, which comprises 74 species distributed in the Northern hemisphere, South America and South-Eastern Asia. In Campania, where the vast majority of Italian populations are found, according to IUCN the plant has been assessed as Vulnerable (VU). Within this region, the plant is known to grow only in ravines and river gorges of two mountain systems, namely Monti Lattari and Monti Picentini (Innangi et al., 2011). For this region, an expert-based habitat suitability model has been developed in order to look for unknown locations and/or plan possible repopulation strategies. Expert-based models, although introducing a subjective component in their development, can take into account a wide range of ecological needs for a taxon and their results are extremely synthetic (Boitani et al., 2002). According to the ecological features of known presence points of in Campania (Innangi et al., 2011), a set of five variables has been chosen, namely altitude, slope, rock type, CORINE land cover, and distance from rivers. For each variable, four scores have been assigned, that is zero for no suitability to three for best suitability. The results highlight the strict connection of this plant to water streams and courses, thus, in order to correlate chemical and biological features of the water to the habitat suitability for P. crystallina subsp. hirtiflora, water has been sampled in both presence and not-presence locations for chemical analysis, namely chromatography and toxicity tests, using bioassay organisms such as Daphnia magna Straus, Sorghum saccharatum Pers., and Lepidium sativum L. (APAT, 2002). Acknowledgements. The authors are grateful to Prof. M. Guida and M. Trifuoggi for the toxicity tests and chemical analyses.