元谋干热河谷景观生态安全时空变化

摘要

为了系统地定量评价干热河谷景观生态安全状态及其时空变化,明确生态安全的主要影响因素及作用机制,以2008年Landsat7-ETM和2016年Landsat8-OLI遥感数据为数据源,基于GIS和RS技术,建立景观生态安全度(LESD)模型,结合空间自相关和地统计学理论,对景观生态安全时空变化特征和原因进行分析.结果表明:1)研究区景观生态安全格局指数、景观生态质量指数和景观生态安全度值都在0.40 ～0.60之间,均属于临界安全等级.2)景观生态安全度全局空间自相关值,从2008年的0.600上升至2016年的0.633,全局空间自相关性加强;景观生态安全度局部自相关格局与研究区地形分异格局较为一致,与2008年相比,2016年高值-高值区在东部的中高山区域,出现了较为连续的空间分布,低值-低值区在金沙江沿岸坝周低山和中低山区分割明显.3)2008和2016年块金方差与及基台值之比值(C0/ C0+ C)分别为28.5％和32.6％,变程值(A0)分别为764.82和2 031.65 m.研究区景观生态安全较为敏感,景观生态安全度在整体和局部空间均具有空间自相关性.以地形、气候为主的结构性自然因素是研究区景观生态安全度空间分布的决定性因素,以人为干扰为主的非结构性因素,强化了对景观生态安全性空间分布的影响力.该研究为干热河谷景观生态调控提供了基础数据和理论依据.%[Background] Dry-hot valley is typical of ecological vulnerable environment.Taking Yuanmou Dry-hot Valley as a case area,systematically and quantitatively evaluating the status,temporal and spatial changes of landscape ecological security,are important for clarifying main influencing factors and their action mechanism of landscape ecological security for providing basic data and theoretical basis.[Methods] Based on remote sensing images of Landsat7-ETM in 2008 and Landsat8-OLI in 2016,we established a landscape ecological security degree (LESD) model to analyze spatial and temporal change characteristics and causes of landscape ecological security by using spatial autocorrelation analysis and geostatistics methods.[Results] 1) The values of landscape ecological security pattern (LESP),landscape ecological quality (LEQ) and LESD ranged from 0.40 to 0.60,belonging to the critical safety level.2) The values of global spatial auto correlation of LESD rose from 0.600 in 2008 to 0.633 in 2016,showing that global spatial autocorrelation of landscape ecological security increased.The spatial pattern of the local autocorrelation of LESD was consistent with terrain differentiation pattern in the study area.Significant distribution areas of local autocorrelation mainly included high-high value areas and low-low value areas.Different from 2008,high-high value areas of local autocorrelation obtained a continuous spatial distribution in eastern middle-high mountain areas in 2016,while low-low value residential areas in 2016 were obviously split in the low hills around the river dam and the middle-low mountain areas along Jinsha river.3) The values of the ratio of nugget variance to sill (C0/C0 + C) of LESD were respectively 28.5％ in 2008 and 32.6％ in 2016,as well as the values of the change range (A0) of LESD were respectively 764.82 m in 2008 and 2031.65 m in 2016.[Conclusions] Landscape ecological security of the study area was sensitive.Landscape ecological security of the study area had obvious spatial autocorrelation overall and locally.Structural natural factors dominated by terrain and climate were still determinants of the spatial distribution of regional landscape ecological security,while unstructured factors dominated by human disturbances reinforced their influence on regional landscape ecological security.