Using Terrain-related Variables to Assess the Negative Effects of Topographic Obstacles on Taiwan Red Cypress Distribution

From a mechanistic viewpoint, it is desirable to predict the distribution of species based on ecological parameters (e.g. rainfall or temperature) that are causal factors for their distribution. Data for such ecological factors are difficult or expensive to collect and are usually insufficient. Terrain-related variables are used as proxies of ecological factors in the study since data for them can be easily obtained by remote sensing. Taiwan red cypress (TRC) grow in fog-forest belt (FFB) with rich moisture brought by northeastern (NE) season wind but with lower solar radiation due to high moisture in fogs and clouds. TRCs grow on the north part of the study area, but almost do not grow in the Huisun Experimental Forest Station (HEFS), except Shou-Cheng Mountain, in the south part of the study area. The study attempted to determine if the NE wind blocked by topographic obstacles is a key to above-mentioned distribution by reversely thinking their positive effect with another case of Taiwan fir. Topographic sheltering index (TSI) thus was developed. The accuracy of ecological niche model (ENM) using MAXENT with elevation plus TSI was better than that of the ENM using MAXENT with elevation plus slope, NDVI, and PCT bands. Consequently, TSI was useful for explaining the above-mentioned distribution. Moreover, the greater the TSI, the fewer TRCs grow in the study area, especially in the HEFS due to the negative effect of topographic obstacles opposite to the case of Taiwan fir. This proxy therefore may be a critical predictor variable for ENMs. The rise in accuracy is relatively limited despite TSI improving the predictive ability of these ENMs. Hence, the more refined TSI, the proxy of solar radiation (SR), and high-resolution DEM will be included into ENMS so that their predictive ability can be greatly improved.