未来气候变化情景下横断山北部灾害易发区极端降水时空特征

Changes of Extreme Precipitation Patterns for Disaster Prone Areas in the North of the Hengduan Mountains under Climate Change

极端降水是导致气候变化下滑坡、泥石流等山地灾害变化的重要因素。极端降水的时空特征作为气候变化下山地灾害风险的研究热点之一,为滑坡等山地灾害危险性评价和构建山地社会安全空间奠定了重要基础。尤其对于地形复杂、灾害频发的横断山地区,高精度的极端降水时空特征研究有利于优化国土空间,促进当地减轻灾害风险与加强气候变化适应相结合。本研究以横断山北部岷山、邛崃山和大雪山为例,充分考虑区域复杂地形的影响,在全球气候模式降水数据的基础上,采用统计降尺度的方法,获取该区域2010—2060时段的逐日降水数据(空间分辨率1×1km);并充分考虑极端降水可能对物理事件造成的影响,构建包括绝对量指数、频度指数和强度指数的极端降水评价指标体系。研究结果表明,未来气候变化下,研究区极端降水总体呈现增加—减少—增加的趋势,短时极端降水和持续性极端降水的空间分布相对一致。就短时极端降水而言,大部分区域发生50mm以上极端降水事件的次数较多,而研究区中部邛崃山区甚至会发生超过100mm的极端降水事件。持续性事件的分布受地形阻隔作用影响,主要发生在大雪山高山区域和邛崃山区。结合研究区地理环境条件,地质灾害风险在未来气候变化情景下可能增加。

Extreme precipitation may lead to various mountain hazards. Hence mountainous regions are faced with more uncertainties posed by a changing climate. In order to reduce the adverse impact of extreme precipitation and relevant mountain hazards, it is of vital importance to characterize the spatial-temporal distribution of extreme precipitation for the future. Many studies have analyzed the holistic distribution for future climate scenarios. However, there is still a lack of detailed information to depict daily extreme precipitation, which establishes the foundation for hazard dynamic analysis in specific locations.In this study, the recently released climate projection data were statistically downscaled by historical data from 50 local meteorology stations in and around the study area over the period 1981-2010. Based on the downscaled NEX-GDDP daily precipitation over the period 2010-2060, spatiotemporal patterns of extreme precipitation from three perspectives including intensity, frequency and duration had been analyzed for the disaster-prone area in the north Hengduan Mountains with spatial resolution 1×1 km . Results showed that the total precipitation tended to fluctuate greatly and regional variation existed in the study area. It's worth noting that both the frequency and spatial range of extreme precipitation presented a significant increasing trend in the RCP 4.5 scenario. Regarding duration index, the spatial patterns of short-term extreme precipitation were consistent with that of long-duration extreme precipitation except for the Min Mountains. For short-term extreme precipitation, rainstorms(≥50 mm) occurred more frequently in most of the study area while heavy rainstorms(≥100 mm) were projected to occur mainly in the Qionglai Mountains. In terms of long-duration extreme precipitation, the occurrence frequency was higher in alpine regions of the Qionglai Mountains and the Daxue Mountains, as well as the transition zones from mountain to plain, which may be caused by the passing - obstructing effect of terrain complexity. Moreover,extreme precipitations on the link with the distribution of local slope factors showed higher frequency in steep convex slope than concave slope, which was also confirming the similar local topography effect.