近40年来抚仙湖面积时空分异遥感动态监测

THE SPATIAL-TEMPORAL VARIATION OF FUXIAN LAKE AREA DURING 1974—2014 BASED ON REMOTE SENSING

湖泊面积的变化能反映区域环境和气候的变化,对研究区域的气候和可持续发展意义重大。在RS、GIS技术的支撑下,以MSS、TM、ETM+、OLI遥感影像,彩红外航片,Quick Bird、World View-2数据为主要数据源,结合基础地理信息数据、专题资料和其他相关研究文献资料,利用膨胀型和腐蚀型算子的边缘检测算法提取抚仙湖1974、1977、1987、1993、1996、2000、2001、2002、2004、2005、2006、2009、2012、2014、2015年共15个时段的湖体边界,计算其面积,并分析抚仙湖岸线和面积的时空变化特征;最后,结合流域的气象、土地利用数据,采用灰色关联分析方法对岸线和面积变化的驱动力进行分析,得出其主要驱动因子并构建其多元线性回归模型。研究结果表明:(1)时间尺度上,1974年以来,抚仙湖岸线和面积的变化趋势都是增大-减小-再增大,2009年以后都在减小,且幅度较大,变化强度明显增强。(2)空间尺度上,40年来,抚仙湖在各个方向上都有萎缩和扩张,变化趋势复杂,总体重心往南偏移。(3)灰色关联分析显示人为和自然因素与岸线和面积变化的关联度都在0.6以上,表明抚仙湖湖面变化是耕地、建筑用地、植被覆盖以及气象因子共同作用的结果,而年平均气温与岸线和面积的关联度分别达到0.935和0.993,表明年平均气温是引起湖面变化的主要因子。

Using GIS and RS technologies and combining MSS, TM, ETM＋, OLI remote sensing images with geographic data and available project outcomes, we used an edge detection algorithm to extract the boundary change of the Fuxian Lake, Yunnan, from 1974 to 2015. The temporal-spatial variations in area and shoreline length of the lake were analyzed. Meanwhile, the meteorological and land-use data of the lake basin were combined using gray correlation, from which the driving forces of the variations and the main factors were determined in multiple linear regression. The results show that, in 1974—2009, the shoreline length and the area of the lake increased and then followed by a large decrease after 2009. The lake area could shrink or expand in all directions in a complicated pattern, and the position of the lake center moved southward. Grey correlation analysis showed that the coefficient of correlation between the shoreline length and the area changes was 0.7. These changes in Fuxian Lake reflect interactions among agricultural, housing, and construction activities, and meteorological factors as well. The correlation of average annual temperature to the shoreline length and the lake area reached 0.935 and 0.993, respectively. Therefore, the annual temperature was the main factor causing the changes.