| The Southwest region is an important grain producing area. Meanwhile, it is one of the areas which are severely affected by drought disaster. Due to the complicated topography, variable climate, uneven spatial and temporal distribution of precipitation, regional and periodic drought often occurs in the Southwestern, causing huge losses to the local agricultural production and peoples lives. Therefore, research on a better adaptive drought monitoring method and taking appropriate drought monitoring not only contribute to drought relief and disaster prevention of the local government, but also have an important practically significance on agricultural production and people’s quality of live.This paper has systematically analyzed and summarized the drought monitoring methods in domestic and overseas. Based on historical drought and soil moisture data, qualitatively and quantitatively analyzing the response of drought monitoring indexes, this paper discusses the suitability of meteorological drought monitoring index and remote sensing drought monitoring index in southwest China. On this basis, by making full use of the meteorological site data and remote sensing data, comprehensive drought monitoring index (DI) which bases on multi-source data is developed. This index can rich drought monitoring mechanism, optimize single drought monitoring index, improve monitoring accuracy, and is more suitable for complex terrain regional drought monitoring. Finally, drought in southwest area has been taken as an example for analysis.This paper has obtained the following results and conclusions:1、The qualitative analysis between meteorological monitoring index and historical drought data indicates that: Percentage of Precipitation Anomaly Index(Pa) and the Standard Precipitation Index(SPI)and Relative Moisture Index (MI), based on meteorological stations interpolation method, have good consistency in spatial variation. They all have reflection on the drought, but there is a big difference in the drought degree. The degree reflected by Pa and SPI is light. Pa has certain mutation and hysteresis, which cannot reflect the gradual character of drought well. In low temperature area, the degree reflected by MI is worse. In addition, the results of drought spatial distribution obtained though site spatial interpolation, inevitably inducing drought buffer zones, which show that the drought monitoring method based on single meteorological drought monitoring index has certain limitation. In general, the monitoring results of MI are consistent with the actual drought. MI can reflect actual drought better than SPI and Pa.2、The quantitative analysis between remote sensing monitoring index and historical drought data indicates that: Vegetation Water Supply Index (VSWI)and Normalized Difference Water Index(NDWI) have good consistency with historical drought data. In spatial distribution of drought, VSWI and NDWI is roughly similar, but VSWI has wider drought range than NDWI, indicating that VSWI is more sensitive to drought and exists more uncertain problems in drought monitoring. In drought degree, VSWI and NDWI are lighter than actual drought. NDWI is more determined and reliable than VSWI, but they both cannot truly reflect the area which is covered by cloud and snow. VSWI is affected by cloud and snow worse than NDWI, therefore, NDWI is to be given priority in remote sensing drought index.3、This paper has made correlation analysis of meteorological drought monitoring index, remote sensing monitoring index and soil relative moisture data using a pixel-to-weather station paired correlation approach. The result shows that in a month time scales, meteorological drought index MI and soil relative humidity has a highest correlation, the Pearson correlation coefficient is 0.477; remote sensing index NDWI and soil relative humidity has a highest correlation, the Pearson correlation coefficient is 0.416. In addition, the correlation of the same type of drought monitoring index is higher than different type. This demonstrates that there are complementary between different types of index, because different type indexes represent different drought information.4、Through qualitative and quantitative analysis, This paper sets up comprehensive drought monitoring index (DI) on the basis of NDWI and MI. The estimate value of the DI has good correlation with soil moisture measured values and the correlation coefficient are all through p<0.05 significance level, which shows that it is good for drought monitoring in southwest area. Furthermore, the correlation coefficients between DI values and soil moisture values are significantly higher than MI or NDWI in all altitudes. This demonstrates that the fitting effect of DI based on multiple data sources with soil moisture values is better than index based on single data source in different altitudes. That is to say, DI suits better in southwest area.5、Using DI monitoring the drought in southwest in 2009-2010, the process of development and regression of the drought has been reappeared. The severe drought mainly concentrates in south region of Sichuan province, large part of Yunnan province and western region of Guizhou province. The drought evolution process is as follows: the beginning of the drought (2009.9-2009.10), the development of drought (2009.11-2010.2) and the recession of drought (2010.1-2010.5), the end of drought (2010.6). In order to further verify the reliability of the DI model, the paper analyzes DI value and soil moisture with the typical month, and the results are all through p<0.01 significance level. By Comparatively analyzing the drought spatial distribution with DI index and the results of other researchers with SPEI index at March in 2010, the paper finds that the range and degree of drought are consistent with the results of other researchers, which indicates that comprehensive drought monitoring index can be used in drought monitoring in southwest China.6、Through the analysis between DI and rainfall, the paper finds that DI values are significantly positively correlated with the curve of rainfall, which shows that DI index is sensitive to rainfall. Through the analysis between DI and the atmosphere relative humidity, the paper finds that DI values have high dependency with the relative humidity of the atmosphere, which are all through p<0.05 significance level. From drought area to humid area, the correlation coefficient between DI and the atmosphere relative humidity slowly decreases, indicating that DI index is more sensitive to atmosphere relative humidity in severe drought area. It also indirectly shows that the DI index can reflect the drought in southwest. |
PDF Full Text Request