Landscape factors affecting the distribution of the great tit in fragmented urban forests of Seoul, South Korea

Abstract

Seoul City, as the capital and largest metropolis of South Korea, has high population density, and its remaining forests are extremely fragmented as a result of intensive development. The purpose of this study is to analyze the distribution of the great tit (Parus major) and develop a habitat model. Tits have been known as forest birds that inhabit urban forests in Seoul City. A Seoul biotope map and land cover map were combined, and the urban forest areas were delineated on these maps. As factors affecting the distribution of the great tit, landscape indices were calculated by separating them into intrapatch indices, namely logged patch area (PA), area-weighted mean patch shape index (PSI), tree rate (TR), and tree area (TA), and interpatch indices, namely patch degree (PD), patch betweenness (PB), patch proximity (PP), and difference probability of connectivity (DPC), to analyze the internal properties of the patches and their connectivity by tit occurrence data. The results of a t test analyzing the differences of habitat variables by great tit presence/absence and distribution showed differences in PA (p < 0.05), PSI (p < 0.001), TR (p < 0.01), TA (p < 0.001), PD (p < 0.05), PB (p < 0.01), PP (p < 0.05), and DPC (p < 0.1). In the results of logistic regression modeling, in most cases, more than one model had ΔAICc less than 2.0, suggesting that multiple models may be the best model. In the comprehensive models, the TR, PB, and DPC metrics were the best predictors. Among all variables in the intrahabitat model, PSI was the best predictor of species occurrence. Within the interhabitat model, DPC was the best predictor and had a positive relationship with species occurrence, which is the most important of all the factors. In other words, the great tit is more likely to be found in a large circular forest patch than a small and long-shaped forest patch, and connectivity in the threshold distance is the most important factor in predicting great tit occurrence. These results suggest that ecological potential, not only “location potential” but also “supply potential,” should be considered in fragmented urban forests.