Abstract
The fundamental niche is the set of conditions within which a species would be successful in the absence of biotic factors that might limit success. When a species or taxa is not well studied, as in the case with many lichens, understanding the environmental conditions that form its fundamental niche is a useful starting to point to predict species distribution. In recent years, ecological niche models have been used to successfully estimate fundamental niches and to predict species distribution. Ecological niche models are based on species presence and absence data and combinations of environmental variables. Once the environmental requirements for a species have been identified, predictive models can be used to extrapolate where the species would be expected to occur elsewhere in the landscape, allowing the fundamental niche to be mapped. The fundamental niche is reflected on a map based on the areas where abiotic conditions are right for a given species to occur. This lays the foundation to explore larger questions such as the role of biotic interactions, predicting species responses to climate change, or developing conservation strategies. In this study, I have applied ecological niche modeling to Niebla homalea, a fog lichen endemic to the coastal zones of California and Baja California. The distribution and habitat requirements of N. homalea are largely unknown and the goal of this study is to expand our understanding of these attributes. Presence/absence data of N. homalea were collected at Bodega Bay, Point Reyes National Seashore, Half Moon Bay/San Francisco, and Monterey; in California, United States of America. Sampling sites were restricted to public lands with rock outcrops and stratified by average summer day-time fog density. Data on nine independent variables were also collected. Non-parametric multiplicative regression (NPMR) was used to find which independent variables are the best predictor variables for the presence of N. homalea. NPMR uses a local multiplicative smoothing function with leave-one-out cross-validation to estimate the response variable. The best predictor variables were found to be precipitation, habitat type, temperature, and fog density. The model was then used to extrapolate the probability that N. homalea exists in similar environments to those where the data were collected, based on independent variable characteristics. A distribution map representing the relative likelihood (high to low) of suitable habitat, based on the inter-relationships of the predictor variables was created using ArcGIS. The mapping allowed a comparison between the predicted habitat suitability map and the known distribution of N. homalea. A comparison of predicted habitat suitability and known locations of N. homalea found that N. homalea tended to be in high probability of occurrence areas on the habitat suitability map.