Detecting Spatiotemporal Dynamic Landscape Patterns Using Remote Sensing and the Lacunarity Index: A Case Study of Haikou City, China

Abstract

Quantifying the landscape pattern and its dynamics is essential for the monitoring and assessment of the ecological consequences of urbanization. As one of the Special Economic Zones, Haikou is one of the fastest growing regions among all Chinese cities, owing to rapid real estate development. Using a GIS-based land-use dataset from 1986, 1996, and 2000, in combination with a lacunarity index, we attempt to quantify the spatial pattern in the Haikou metropolitan area. After the landscape structure changes over the periods 1986–96 and 1996–2000 are analyzed, a Markov conversion matrix is applied in order to study the sources and destinations of landscape dynamic changes. The lacunarity index is calculated in order to measure the landscape dynamics, with respect to several major land-use types, at a range of spatial scales. The findings indicate that the leapfrog development of real estate and the rapid economic growth of Haikou City have had a great impact on the dynamic landscape patterns. From 1986 to 1996 urban land expanded dramatically and clustered, while cropland was encroached upon and fragmented. From 1996 to 2000, after the government had implemented the strict cropland protection measures, urban expansion and cropland misuse were controlled to a large degree, and a lot of cropland was reclaimed in certain areas. We investigate the dynamic landscape pattern and process, and their implications in policy and economic development.