A new method for assessment of regional drought risk: information diffusion and interval mapping adjustment based on k-means cluster points

Abstract

Abstract Aiming at the defect that there is no ability for the conventional weighted comprehensive assessment method (WCA) to grade drought risk directly, a method based on k-means cluster points to realize the classification of drought risk is proposed in this paper. On the basis of calculating the drought risk value of cluster points, the inverse distance weight interpolation method (IDWI) and multidimensional normal diffusion method (MND) were used to quantify the drought risk value, and the discrimination between the risk value and grade was improved by interval mapping adjustment (IMA). In this paper, the drought risk of Anhui Province from 2000 to 2020 was calculated to verify the above method. The results show that: (1) The drought risk quantification method based on information redistribution of k-means cluster point can not only realize automatic risk classification, but also re-quantify the risk value of the assessment object in the same risk grade, which makes up for the defects that the conventional WCA cannot carry out grade division and the conventional clustering method cannot assign the risk value of the assessment object. (2) The result of information redistribution based on MND is closer to the actual drought situation and more reasonable than IDWI. (3) The dispersion effect of risk value obtained by information redistribution based on k-means cluster point can be improved by the IMA of drought risk. It improves the discrimination degree of risk value, so that the grades can be displayed more intuitively. The defect of the WCA is overcome by the new method proposed in this paper, the follow-up utilization space is widened, and the thinking of risk quantification in drought risk assessment is broadened.