云南省2000-2020年石漠化时空演变分析

Spatio-temporal evolution process of rocky desertification in Yunnan Province from 2000 to 2020

【目的】在宏观尺度利用遥感手段识别和提取石漠化斑块信息,定量分析石漠化的空间分布特征和时序演变过程,可为中国西南地区石漠化的发生、发展过程研究提供技术手段和数据支撑。【方法】以云南省为研究区,基于谷歌地球引擎(Google Earth Engine)平台,以2000、2010和2020年为研究时段,采用决策树分类法定量提取石漠化斑块信息;通过时空演变模型分析云南省石漠化的时空演变过程和特征。【结果】(1)研究时段内,云南省石漠化状况整体得到改善,石漠化占国土面积比例从9.65%下降至6.48%,但局部地区仍在恶化。(2)云南省石漠化空间分布呈东多西少的特征,其中昭通、曲靖、文山、红河和昆明等地区的石漠化分布面积较大,占全省石漠化发生面积的65.42%~72.14%。(3)石漠化演变流向复杂,既转移至石漠化程度高的等级又向石漠化程度低的等级转移,边治理边破坏的现象未被遏制。(4)极重度石漠化修复所需的周期较长,近20 a间云南省极重度石漠化大部分转移至重度石漠化。(5)石漠化内部演变过程剧烈,轻度石漠化和中度石漠化的综合变化速率较快,易于改善的同时亦存在较高的恶化风险,需加以重视。【结论】依据相关规程中的石漠化等级判定标准,采用决策树分类法,可准确提取宏观尺度的石漠化斑块信息。云南省石漠化土地面积呈减少趋势,空间分布不均匀,不同石漠化等级之间存在相互转化的现象。

[Objective]This study aims to identify and extract information of rocky desertification patches at macroscopic scale by remote sensing tools and quantitatively analyze the spatial distribution characteristics and temporal evolution process of rocky desertification,so as to provide technical means and data support for the study of the occurrence process and development of rocky desertification in southwest China.[Method]Information of rocky desertification patches in Yunnan Province in 2000,2010 and 2020 was extracted quantitatively by decision tree classification method based on Google Earth Engine platform.The spatio-temporal evolution model was used to analyze the spatial-temporal evolution process and characteristics of rocky desertification.[Result](1)The overall rocky desertification situation in Yunnan Province was improved during the study period but still deteriorated in some areas.The ratio of rocky desertification to land area decreased from 9.65%to 6.48%.(2)The spatial distribution of rocky desertification in Yunnan Province was characterized by more in the east and less in the west,and the major distribution areas were Zhaotong,Qujing,Wenshan,Honghe and Kunming,accounting for 65.42%−72.14%of the occurrence area of rocky desertification in the whole province.(3)The evolution of rocky desertification tended to be complex,which shifted to both high grade and low grade of rocky desertification,and the phenomenon of destruction while treatment was not curbed.(4)Most of the extremely severe rocky desertification has shifted to severe rocky desertification in the past 20 years,indicating that it would take a long time for the extremely severe rocky desertification to be restored.(5)The internal evolution process of rocky desertification was intense,and the comprehensive change rate of light rocky desertification and moderate rocky desertification was fast.It was easy to improve but also had a higher risk of deterioration.[Conclusion]The relevant protocols are used as criteria for determining the grade of rocky desertification,and the macro-scale information of rocky desertification patches can be accurately extracted by decision tree classification method.The area of rocky desertification in Yunnan Province is decreasing,with uneven spatial distribution and mutual transformation among different grades of rocky desertification.