Using kernel density estimation to assess the spatial pattern of road density and its impact on landscape fragmentation |
| |
| Authors: | Xuejiao Cai Jiong Cheng |
| |
| Institution: | 1. Guangzhou Institute of Geochemistry, Chinese Academy of Sciences , Guangzhou , China;2. Graduate University of the Chinese Academy of Sciences , Beijing , China;3. Guangdong Institute of Eco-Environmental and Soil Sciences , Guangzhou , China;4. Guangdong Institute of Eco-Environmental and Soil Sciences , Guangzhou , China |
| |
| Abstract: | Road density (i.e., km/km2) is a useful broad index of the road network in a landscape and has been linked to several ecological effects of roads. However, previous studies have shown that road density, estimated by grid computing, has weak correlation with landscape fragmentation. In this article, we propose a new measure of road density, namely, kernel density estimation function (KDE) and quantify the relation between road density and landscape fragmentation. The results show that road density estimated by KDE (km/km2) elucidates the spatial pattern of the road network in the region. Areas with higher road density are dominated by a larger proportion of built-up landscape and less possession of forest and vice versa. Road networks segregated the landscape into smaller pieces and a greater number of patches. Furthermore, Spearman rank correlation model indicates that road density (km/km2) is positively related to landscape fragmentation. Our results suggest that road density, estimated by KDE, may be a better correlate with effects of the road on landscape fragmentation. Through KDE, the regional spatial pattern of road density and the prediction of the impact of the road on landscape fragmentation could be effectively acquired. |
| |
| Keywords: | road density road network KDE landscape fragmentation core area of the Pearl River Delta |
|
|
