中国自然灾害灾情统计与自组织临界性特征

基于自组织临界性(SOC)思想,在分析中国自然灾害灾情统计特征的基础上,分析并揭示了中国自然灾害灾情的自组织临界性特征。在此基础上,建构了中国自然灾害灾情(包括松花江等七大流域洪涝灾害受灾县数、洪涝灾害伤亡人数、干旱伤亡人数、各类自然灾害总的伤亡人数等)标度-频度分形关系模型。最后指出,中国自然灾害灾情的自组织临界性特征是客观存在的,而标度-频度的幂律值则是一个相对值。

Statistics and Self-organized Criticality Characters of Natural Disasters in China

Bak P,Tang C and Wiesenfeld K enunciated the concept of Self-organized criticality(SOC)in his paper"Self-organized criticality:an explanation of 1/f noise?"published in "Physical Review Letters" in the 1980s. The SOC concept was presented for the first time in that paper and has been applied to many fields ever since such as earthquake,weather and population etc. It has been adopted thousands upon thousands times. SOC is a new method to observe and explain the complexity and mechanism of nature. The negative power is the direct proof of SOC,fractal and disaster are two exterior representations of SOC. A SOC system can be forecasted. China is an agricultural country to some degree. However,natural and social factors result in the frequent natural disaster,which has greatly affected the agriculture and economy of China. Thus,it is useful to study the characters of statistics and SOC of natural disasters of China. In this paper,the characters of statistics and SOC of exposure of natural disasters of China are discussed based on the concept of Self-organized criticality and fractal. Furthermore,the relationships between the scale and frequency of cities covered by flood,died people by drought and flood are constructed. For example,the relationship between the scale and frequency of cities covered by flood is logN=4.0957-0.577logr,that of died people by drought is logN=3.9486-0.5359logr,and that of died people of all disasters is logN=4.8645-0.6526logr. The paper also discriminates the relationship between the negative power and scales,it is found that the character of self-organized criticality of China disaster exposure exists objectivity,but the value of the negative power of relationship between the scale and frequency may change with different scales.