气候变化对黄土高原浅层滑坡影响的模拟研究——以延安宝塔区为例
徐增辉1, 金继明1,2, 蔡耀辉3,4, 杨 涛1

(1.西北农林科技大学 水利与建筑工程学院, 陕西 杨凌 712100; 2.中国旱区节水研究院, 陕西 杨凌 712100; 3.中国科学院 水利部 水土保持研究所, 陕西 杨凌712100; 4.西北农林科技大学 水土保持研究所, 陕西 杨凌 712100)

气候变化; 浅层滑坡; 降雨降尺度; TRIGRS模型; Rosenblueth点估法

Impact of Climate Change on Shallow Landslides in the Loess Plateau-A Case Study in Baota Region, Yan'an City
XU Zenghui1, JIN Jiming1,2, CAI Yaohui3,4, YANG Tao1

(1.Institute of Water Conservation and Architectural Engineering, Northwest A&F University, Yangling, Shaanxi 712100, China; 2.Institute of Water-Saving Agriculture in Arid Areas in China, Yangling, Shaanxi 712100, China; 3.Institute of Soil and Water Conservation, CAS & MWR, YangLing, Shaanxi 712100,China; 4.Institute of Soil and Water Conservation, Northwest A&F University, Yangling, Shaanxi 712100, China)

climate change; shallow landslide; precipitation downscale; TRIGRS; Rosenblueth point estimation method

备注

黄土高原为浅层滑坡地质灾害易发区。在全球气候变化的背景下,极端降水增多,是否会导致浅层滑坡地质灾害增多亟待研究。以延安宝塔区为例,在3种全球气候模式(GCM)2种典型浓度情景RCP(Representative Concentration Pathway)下,利用研究区降水降尺度数据驱动TRIGRS(Transient Rainfall Infiltration and Grid based Regional Slope-stability Model)滑坡预报模型,结合Rosenblueth点估法解决土壤参数不确定性问题,对延安宝塔区1979—2100年滑坡分布情况进行模拟,研究了不同气候模式和情景下降雨事件变化趋势,进而对浅层滑坡的影响。通过对13 188次滑坡模拟结果的分析表明:不同的气候模式降水驱动的滑坡模拟预报表现出不同的趋势,其中GFDL-ESM2G和MIROC5两种气候模式未来滑坡增多趋势明显,在RCP4.5的情景下,与历史时期(1979—2018年)相比,未来(2059—2098年)研究区内浅层滑坡面积大约增多23.10%和43.16%,在RCP8.5的情景下大约增多31.14%和47.17%; IPSL-CM5A-MR气候模式未来滑坡预报呈略微下降趋势。未来研究区内浅层滑坡增多主要是由于诱发滑坡的降雨事件次数增多导致。本研究可为黄土高原地区未来灾害预警等提供参考。

The Loess Plateau is the prone area of shallow landslide geological disaster in China. Under the background of climate change, the global extreme precipitation weather is increasing, and the corresponding geological disasters are also increasing. Taking Baota Region of Yan'an as an example, the rainfall forecast results of 1979—2098 under three climate models and two kinds of RCP scenarios in CMIP5 were processed by statistical downscaling, and the change of shallow landslide in 1979—2098 in the study area was obtained by using the rainfall downscaling data driven TRIGRS landslide forecast model and Rosenblueth point estimation method. Based on the analysis of 13 188 landslide simulation results, it is shown that: different climate models driven by precipitation show different trends in landslide simulation prediction, among which GFDL-ESM2G and MIROC5 two climate models have obvious upward trend in future landslide prediction; under the scenario of RCP4.5, compared with the historical period(1979—2018), the area of shallow landslide in the future(2059—2098)study area will increase by about 23.10% and 43.16%, increased by 31.14% and 47.17% under the scenario of RCP8.5; the future landslide prediction of IPSL-CM5A-MR climate model shows a slight downward trend. In the future, the increase of shallow landslides in the study area is mainly due to the increase of rainfall events. This study can provide reference for future disaster warning in the Loess Plateau.