Drought Risk Assessment Method And Its Application Based On Cloud Model

Drought has been one of the most serious natural disasters in China and other parts of the world.With the increase of global climate change and human activities,the damage of drought also shows an increasing tendency.Due to the fact that drought is affected by many factors,the existing uncertainty is the key to drought risk.Therefore,based on the cloud model theory proposed by Li Deyi,the uncertainty assessment of drought risk in three levels of region,including Bengbu city,Anhui Huaibei plain and Anhui province,was researched through the improved similarity of cloud,clouded information diffusion and cloud reasoning method.The main results are shown as follows:(1)Through index selection,expert scoring,and analytic hierarchy process,the index system is introduced,and then,based on the index system and the basic data,the cloud similarity is im proved,finally,the drought vulnerability asse ssment model based on improved cloud sim ilarity is established and applied to the assessm ent of drought vulnerability in Bengbu city over period of 2001 to 2010.The research results show that: the assessment results calcul ated through im proved cloud sim ilarity are close to that calculated by fuzzy comprehensive evaluati on method,and are m ore conservative.The original data characteristics of the assessment indexes can be retained.And it is possible to avoid the phenomenon of losing fidelity caused by indexes data normalization and different indexes dimension.(2)Based on clouded information diffusion method,drought vulnerability assessment model based on clouded infor mation diffusion is established and used to estim ate the probability density distribution of drought vu lnerability in Anhui province over period of 1988 to 2012.The m ethod is a lso compared with the probability statistics an d bootstrap method.The results show that: th e assessment results calculated through clouded information are very close to the prob ability density distribution of historical drought loss rate in Anhui province.This could avoid the problem that probability density could not be directly estimated due to original incomplete data.(3)Based on the R=P×C model which is improved through the cloud m odel,drought and drought da mage are m erged through cloud reasoning.And drought vulnerability assessment model based on cloud reasoning is established to assess drought risk.The results show that: the assessm ent calculated through cloud reasoning are close to the historical real disaster condition.And this could avoi d the problem that weights determined by traditional method were more subjective and the result of the calculations were very different because of the undoubtedly “×”.(4)Based on the digital features of cloud model,temporal-spatial distribution model is established,and is used to quantitatively analyze the uncertainty of reference crop evapotranspiration of Huaibei plain at temporal-spatial scale.The results show that: the assessment based on cloud m odel is sim ilar to the result of traditional m ethod.In addition,the cloud feature digi tal entropy En and super entr opy He are used to further quantitatively describe the uncertainties such as the degree of dispersion and stability in the analysis process.In summary,drought risk assessm ent based on cloud m odel theory and m ethod can not only obtain the s imilar results as the tradition al method but als o quantitatively describe the uncertainties include degree of dispersion and stability by digital features of cloud model.The method is feasible and ef fective,and can provide the scientific basis and the theoretical foundation for the drought risk uncertainty management.