基于正态云信息扩散方法的水旱灾害风险评估模型——以徐州市为例

Risk Evaluation Model of Flood and Drought Disaster Based on Normal Cloud Information Diffusion Method:A Case Study in Xuzhou City

由于自然灾害风险评估中存在案例样本稀少、信息难以获取等特点,为了降低各种不确定性因素影响,提高灾害风险评估的合理性和可靠性,建立基于正态云信息扩散方法的灾害风险评估模型,用一组云滴来反映灾害风险的超越概率。针对信息扩散原理中仅考虑模糊不确定的不足,综合考虑信息扩散过程中模糊性和随机性的关联性,用正态云扩散函数代替正态扩散函数,发展了模糊信息扩散方法。最后,结合1990-2012年徐州市农作物水旱受灾面积资料,将该模型应用于徐州市农业自然灾害风险评估,计算结果表明,徐州市农业水旱灾害风险趋势基本一致,干旱灾害的风险略大于洪涝灾害,农业受损程度主要集中在35%以内,受损程度在35%以上的可能性很小,这与徐州历史上水旱灾害发生的实际基本相符。

Aiming at the difficulties in natural disaster less sample cases and rare collectable information,and in order to reduce the error caused by uncertainties and improve the rationality and reliability,a disaster risk assessment model based on normal cloud information diffusion method,and a set of cloud drops are used to estimate the exceeding probabilities of disaster loss.Since the principle of information diffusion model only considers the shortage of fuzzy uncertainty,the relevance of fuzziness and randomness is considered in the information diffusion process by the normal cloud diffusion function instead of normal diffusion function.The model is applied in a case study of agricultural flood and drought risk in Xuzhou city of Jiangsu province.The results show that the risk of agricultural drought and flood in Xuzhou are basically the same,the drought disaster risk is slightly higher than the flood disaster,agricultural damage mainly concentrated in less than 35%,the extent of the damage probability of more than 35%is very small.The model evaluation results are basically consistent with the Xuzhou real situation of the flood and drought disasters.