基于格网数据和博弈论组合赋权的中巴经济走廊洪灾风险评价

Evaluation of flood disaster risk in China-Pakistan Economic Corridor by combination weighting based on improved game theory and grid data

中巴经济走廊是中国“一带一路”倡议的先试区和重点区,如何实现该区域内的精确可靠的洪灾风险评价是中国“一带一路”推进中所面临的关键问题。该研究依据中巴经济走廊0.25°×0.25°格点数据集(V1)中的日降水数据集,综合考虑致灾因子、孕灾环境和承灾体等构建了由多变量组成的评价指标体系,提出运用改进的博弈论耦合直觉模糊层次分析主观权重和基于指标相关性的客观权重(Criteria Importance Though Intercrieria Correlation,CRITIC)并计算得到最优组合权重。以中巴经济走廊为典型研究区开展洪灾风险评价的实例分析并对其年代际演变特征进行了研究,结果表明:1)基于改进的博弈论组合赋权实现了单位约束下组合权重寻优计算,在此基础上得到的风险分区更加精细;2)计算所得2010年洪水淹没面积为14.9万km^(2),计算结果与实际调查洪水淹没面积间相关系数为0.66,验证洪灾风险评价方法合理、可靠,评价结果基本符合历史实际洪灾情况;3)中巴经济走廊内洪灾危险性受降水、地形影响较大,其风险度的空间分布表现为东南部大于西北部;4)中高风险区和高风险区占研究区面积的28.5%,基本为地形变化小、河网密集、人口分布集中的区域,由此指示需关注中巴经济走廊内洪灾对社会经济发展的影响与危害;5)区域内洪灾高风险区随1990—2010年代际变化而扩大,引起风险上升的主要因素为极端降水的发生。研究结果可为“一带一路”等重点地区的自然灾害防范和评估分析提供理论依据和决策支持,并为洪涝灾害预报提供有效信息。

The China-Pakistan Economic Corridor(CPEC)is a pilot and key area of China's“One Belt and One Road”Initiative.Achieving accurate and reliable flood risk assessment in this region was a key issue.Based on the daily precipitation data set in the 0.25°×0.25°grid data set(V1)of the China-Pakistan Economic Corridor,an evaluation index system composed of multiple variables was constructed by comprehensively considering the disaster causing factors,disaster environment,and disaster bearing body,etc.This study proposed an improved game theory coupled with the subjective weight of intuitionistic fuzzy hierarchy analysis and the objective weight based on the index Correlation(Criteria Importance Though Intercrieria Correlation,CRITIC),which can be used to calculate the optimal combination weight.Taking the CPEC as a typical research area to conduct a case study of flood risk assessment and research on its interdecadal evolution characteristics.A case study of flood risk assessment was carried out,and its interdecadal evolution characteristics were studied.At the same time,it has strengthened the research on the index system of risk assessment in the CPEC and the construction of the weight model.In order to obtain theoretical reference and decision support for sustainable development and infrastructure construction in key areas and provided effective information for flood disaster forecasting.The results showed that:1)By the improved model,the extreme zones of floods were continuously enlarged,and the description of their distribution was more refined and perfect.The flood area calculated in 2010 was 14.9×104 km^(2),and the correlated coefficient between the calculated flood area and the actual flood area was 0.66.The flood risk assessment method developed was reasonable and reliable,and the evaluation results were basically in line with the actual flood in Pakistan in 2010.2)The flood risk was greatly affected by precipitation and topography,and the spatial distribution of the flood risk degree was greater in the southeast than in the northwest in CPEC.3)Medium and high-risk areas accounted for 28.5%of the study area,mainly the areas with small topographic changes,dense river network,and concentrated population distribution,which indicates the attention should be paid to the impact and harm of floods on social and economic development in CPEC.4)The region's high-risk areas of flood disasters expanded with the interdecadal changes from 1990 to 2010.Changes in the degree of risk over time could be obtained by comparing these results.Among them,rainfall was the most obvious influence factor.Its intensity range and duration were important and played a decisive role in the possibility of flood disasters.Under the conditions of large intensity,wide area,and long duration,flood risk zones of flood disasters will change.The most obvious was the expansion of medium-high and high-risk areas.5)There were some differences in the scope of flood risk zones in different return periods,but the overall risk trend was consistent.Compared with the high scenario(once in 100 years),the scope of low and medium low-risk areas increased significantly,the scope of medium risk areas expanded slightly,and the area of medium-high and high-risk areas decreased.The decreased areas were mainly distributed in the southern region of the northwest border and eastern Punjab Province.Therefore,it can be concluded that the combined weight assignment based on the improved game theory has realized the optimization calculation of the combination weight under the unit constraint,and the index weight assignment is more scientific and reasonable.