Comparative study on flash flood hazard assessment for Nam Ou River Basin, Lao PDR

• Published in: Natural hazards (Dordrecht) 2020-07, Vol.102 (3), p.1393-1417
• Main Authors: Du, Jun, Fan, Zhong-jie, Pu, Jian
• Format: Article
• Language: English
• Subjects: Autocorrelation; Casualties; Civil Engineering; Coefficients; Comparative analysis; Comparative studies; Correlation; Correlation analysis; Developed countries; Developing countries; Earth and Environmental Science; Earth Sciences; Economic development; Economics; Environmental Management; Extrapolation; Flash flooding; Flash floods; Flood hazards; Floods; Geophysics/Geodesy; Geotechnical Engineering & Applied Earth Sciences; Hazard assessment; Hydroelectric power; Hydrogeology; LDCs; Local knowledge; Natural Hazards; Original Paper; Rain; Rainfall; Regional development; River basins; Rivers; Scale effect; Spatial analysis; Watersheds
• ISSN: 0921-030X; 1573-0840
• DOI: 10.1007/s11069-020-03972-3

Laos is a mountainous, rainy and less developed country in Southeast Asia. In Laos, floods represent a major constraint on social economic development, causing a large number of casualties and property losses each year, among which the impact from flash floods is also very prominent. Especially in recent decades, with the development of social economy and the intensification of hydropower development, the serious threat from flash floods is becoming more and more obvious. However, there is no fundamental defence system for flash floods been established yet in this country, and the basic knowledge on local flash flood development is also ignored. For filling this gap, taking the Nam Ou River Basin as an example, this paper tries to find out the most helpful hazard assessment method for current Laos based on comparative analysis from the flash flood potential index, the calculation of curve number–rainfall erosivity and the extrapolation method. The results show that the extrapolation method based on spatial lag model constructed by the data from Yunnan, China, presents the most reliable outcome compared with the other two methods, indicating the spatial autocorrelation model can also be useful for extrapolation, effectively. Besides, the scale effect of different potential impact elements on flash flood, i.e. the spatial correlation between each element and the distribution of flash flood events at different spatial statistic units, was also preliminary studied. It is found that the degree of correlation in spatial analysis depends on the short board effect, i.e. only the element that restricts the regional flash flood developmental system can be the key factor. The correlations of most watershed elements increase or decrease directly with the rise of watershed scale, and the values of coefficients tend to be stable at large watershed scales.