Using multitemporal Landsat imagery to monitor and model the influences of landscape pattern on urban expansion in a metropolitan region
Studying the interaction between landscape patterns and temporal land-use changes in a metropolitan area can improve understanding of the urbanization process. Multitemporal remote sensing imagery is widely used to map the urbanization-caused temporal land-use dynamics, which mainly appear as built-...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/101756 http://hdl.handle.net/10220/19768 |
| Summary: | Studying the interaction between landscape patterns and temporal land-use changes in a metropolitan area can improve understanding of the urbanization process. Multitemporal remote sensing imagery is widely used to map the urbanization-caused temporal land-use dynamics, which mainly appear as built-up growth. Remote sensing integrated with landscape metrics is also used to quantitatively describe the landscape pattern of the urban area in recent literature. However, few studies have focused on the interaction between the pattern and the process of urbanization in a metropolitan area. We propose a grid-based framework to analyze the influence of the landscape pattern on the built-up growth by using the multitemporal Landsat imagery. Remote sensing classification method is used to obtain thematic land-use maps. Built-up growth is then extracted from the multitemporal classification results by a postclassification change detection. Landscape pattern, which is quantitatively described by landscape metrics, is derived from the thematic land-use maps. A grid-based method is used to analyze the spatial variation of landscape pattern and its related built-up growth. Finally, the spatial relationship between the landscape pattern and the built-up growth characters is assessed and modeled by using the mathematical regression method. The present study shows that an apparent correlation between landscape pattern and built-up growth exists. The correlation reflects the inherent influences of landscape pattern on urban expansion. The landscape pattern indicates the land development stage, while the urbanization stage determines the speed and style of the following built-up growth. Scales, including temporal scale and spatial scale, are important to modeling the landscape pattern effects on the built-up growth. The proposed analysis framework is efficient in detecting and modeling the landscape pattern effects on the built-up growth. |
|---|