基于CA-Markov模型的滨海土壤盐渍化动态变化研究

为了研究滨海土壤盐渍化的动态变化规律，本文以河北省黄骅市为研究区，通过解译1990年、2000年、2010年三个相同季节的遥感影像，得到三期土壤盐渍化程度分布图。使用Markov模型和动态度模型分别求出了盐渍化土壤转移面积矩阵和变化速度，使用CROSSTAB模块得到土壤盐渍化程度空间变化图，并分析了影响研究区盐渍化的4个主要因素，通过MCE模块得到土壤盐渍化影响因素图集；将1990年、2000年两期解译影像输入CAMarkov模型得到2010年仿真图，同2010年实际解译影像相比，数量精度和空间精度均能接近或达到85%；进而模拟2020年黄骅市土壤盐渍化分布状况，CA-Markov模型预测结果表明：2010-2020年间，轻度盐渍化土壤面积持续增加，中度、重度盐渍化土壤面积均有所减少，到2020年轻度、中度、重度盐渍化土壤面积分别占研究区的26.50%、19.89%和14.79%。此研究可为当地土壤盐渍化的治理及相关规划的编制提供依据。

Dynamic Change in Coastal Soil Salinization Based on the CA-Markov Model

In order to study dynamic change in coastal soil salinization and predict future trends we examined Huanghua in Hebei province and derived three soil salinization degree distribution maps based on the interpretation of same season remote sensing images for 1990,2000 and 2010. Using Markov and dynamic models we obtained the transferred area matrix,spatial changes and change rate of salinized soil based on the CROSSTAB module. Analyzing four factors which affects the salinization of this area,and obtained atlas of influence factors of soil salinization by MCE module. Using two remote sensing interpretation images for 1990 and 2000 in the CA-Markov model,the simulation map of 2010 was deduced. We compared the simulation map of 2010 with the actual interpretation image in 2010 and found that number precision and spatial precision is close to 85%. The simulation results show that the area of mild salinization soil is increasing while the areas of moderate salinization and severe salinization are declining. By 2020,the area of mild salinization, moderate salinization and severe salinization will account for 26.50%,19.89% and 14.79% of the total area respectively. This research is of vital significance in that it provides a basis for the management of local soil salinization and the establishment of related environment planning.