Abstract: [Objective]Temporal and spatial heterogeneity of vegetation cover in Yuanmou dry-hot valley from 2008 to 2016 was studied to analyze the causes for VFC change, in order to provide basic data and a theoretical basis for ecological protection of regional vegetation.[Method]Based on remote sensing images of Landsat ETM and OLI in five periods (2008, 2010, 2012, 2014 and 2016), we obtained VFC data of the five periods by using pixel dichotomy method with ENVI as technical platform. Base on determining the classification criteria for the vegetation coverage levels, some geospatial analysis methods were used to study characteristics of VFC degree, and to analyze composition of VFC degree in each elevation belt. Area of different VFC degree in each year was calculated by ArcGIS, as well as a VFC transfer matrix of different VFC degree was obtained by GIS overlay analysis between 2008 and 2016. Spatial samplings of vegetation coverage in different years were carried out with spatial grids with an area equal to the study area, then standard deviation and regression slope of vegetation coverage were calculated by multivariate statistical method to analyze time evolutionary characteristics of the vegetation coverage.[Result]Spatial pattern of vegetation coverage in Yuanmou dry-hot valley was characterized by middle-low, low, middle and high from the valley dam area to the middle-high mountains, higher vegetation coverage in east and south direction rather than in west and north direction, bounded by Longchuan river valley and Jinshajiang river valley. The values of vegetation coverage of the whole study area were respectively 0.562 in 2008, 0.586 in 2010, 0.494 in 2012, 0.578 in 2014, and 0.566 in 2016. The areas of VFC at level Ⅰ and level Ⅱ in the middle-high mountains respectively accounted for 60% and 50% of the total area of VFC at level Ⅰ and level Ⅱ in the study area. The total area of VFC at level Ⅲ and level Ⅳ in the low-mountain and low-middle mountains around the dam accounted for 70% to 80% of the total area of VFC at level Ⅲ and level Ⅳ in the study area. The area of VFC at level Ⅴ in the valley area accounted for more than 60% of the total area of VFC at level Ⅴ in the study area. The transfer area of different VFC degree accounted for 61.03% of the total area in the study area in the past eight years. There were 95.19 km² transferring from VFC level Ⅰ to Level Ⅱ. The standard deviation (SD) of vegetation coverage was 0-0.541. The ratio of increased area of vegetation coverage to reduced area was 10:9, while the significantly reduced area and the significantly increased area respectively accounted for 9.132% and 6.794% of the study area.[Conculsion] Spatial differences of vegetation coverage in dry-hot valley were significant. VFC value of the whole study area was relatively low, and VFC transformation was relatively frequent. Annual variation of vegetation coverage was not significant. The increased area of vegetation coverage was slightly larger than the reduced area of vegetation coverage, but the significantly reduced area of vegetation coverage was greater than the significantly increased area of vegetation coverage. Vegetation coverage in the middle-high mountainous areas in the east and south was deteriorated. We should continue to strengthen measures of land conversion from farming to forests and grasses, strengthening natural forests protection, reducing intensity of human disturbance in the middle-high mountains and the low mountains. Artificial vegetation restoration should actively be carried out to promote continuous development of regional vegetation ecology.

Key words: dry-hot valley, vegetation coverage, temporal and spatial heterogeneity, GIS, RS, spatial pattern, interannual fluctuation, human disturbance