Human settlement evaluation in mountain areas based on remote sensing, GIS and ecological niche modeling |
| |
| Authors: | Jian Zhao Min Xu Shi-lei Lu Chun-xiang Cao |
| |
| Institution: | 1. State Key Laboratory of Remote Sensing Science, Jointly Sponsored by the Institute of Remote Sensing Applications of Chinese Academy of Sciences and Beijing Normal University, Beijing, 100101, China
2. Graduate School of Chinese Academy of Sciences, Beijing, 100049, China
|
| |
| Abstract: | The Qinghai-Tibet Plateau is the world’s highest and largest plateau. Due to increasing demands for environment exploration and tourism, a large transitional area is required for altitude adaptation. Hehuang valley, which locates in the transition zone between the Loess Plateau and the Qinghai-Tibet Plateau, has convenient transportation and relatively low elevation. Our question is whether the geographic conditions here are appropriate for adapted stay before going into the Qinghai-Tibet Plateau. Therefore, in this study, we examined the potential use of ecological niche modeling (ENM) for mapping current and potential distribution patterns of human settlements. We chose the Maximum Entropy Method (Maxent), an ENM which integrates climate, remote sensing and geographical data, to model distributions and assess land suitability for transition areas. After preprocessing and selection, the correlation between variables and spatial autocorrelation input data were removed and 106 occurrence points and 9 environmental layers were determined as the model inputs. The threshold-independent model performance was reasonable according to 10 times model running, with the area under the curve (AUC) values being 0.917 ± 0.01, and 0.923 ± 0.002 for test data. Cohen’s kappa coefficient of model performance was 0.848. Results showed that 82.22% of the study extent was not suitable for human settlement. Of the remaining areas, highly suitable areas accounted for 1.19%, moderately for 5.3% and marginally for 11.28%. These suitable areas totaled 418.79 km2, and 86.25% of the sample data was identified in the different gradient of suitable area. The decisive environmental factors were slope and two climate variables: mean diurnal temperature range and temperature seasonality. Our model showed a good performance in mapping and assessing human settlements. This study provides the first predicted potential habitat distribution map for human settlement in Ledu County, which could also help in land use management. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
