Analysis on urban lake change during rapid urbanization using a synergistic approach: A case study of Wuhan, China

Abstract

To grasp the evolution of urban lakes accurately is quite necessary for studying on the mechanism of city ecological development. The study about extraction with different types of water by remote sensing technology has developed for decades. Many water indexes as the main methods are used to extract water information. Each method has advantage and disadvantage in different situation. A synergistic approach in this study can reduce the uncertainty of urban lake extraction by using four methods: NDWI, MNDWI, RNDWI and SPM. The basic idea behind the synergistic approach is to give each pixel a score based upon the agreement among the different products of four water extraction methods. According to the score of each pixel, the synergy map, which has been created by the products of four methods, is decomposed into sixteen sub-synergy maps. We use Bayesian Decision Theory to screen out the sub-synergy maps with low confidence level. The remaining ones are recombined a refined map. The overall accuracy of refined map reaches 96.44%, higher than any one of the four methods. Wuhan, known as the City of Hundred Lakes, is selected as the study area. We use the synergistic method to keep track of twenty lakes in Wuhan City changing from 1990 to 2013. The total area of twenty lakes has reduced from 130.2478 km2 to 102.2971 km2 during twenty-three years. The area of Shaihu Lake, which is the most serious of all observed lakes, has shrunk by 77.27%. And Nanhu Lake has lost 8.5 km2 of its area that is the most among all lakes. We also find 1990–2000 is the high tide of urban lake shrinking. After the year of 2000, the situation of lake shrinking has been controlled gradually.