Drought variation characteristics in Ningxia from 2000 to 2010 based on temperature vegetation dryness index by remote sensing

Abstract: Drought is a period of deficit in precipitation with impact on agriculture, water resources and natural ecosystems. It has always an impact on agricultural and ecological fields and causes serious environmental consequences worldwide. Drought has been a prevalent concern for local government in Ningxia Hui Autonomous Region over the last decades. Therefore, monitoring drought characteristics and exploring its change trend in Ningxia are crucial for local agricultural production and ecological construction. Based on the difference of the temperature of day and night (ΔLST), which was calculated from day and night land surface temperature (LST) of the moderate resolution imaging spectroradiometer (MODIS) and normalized differential vegetation index (NDVI) data, the monthly temperature vegetation dryness index (TVDI) in Ningxia from 2000 to 2010 was produced. The time series TVDI data were used to monitor the drought characteristics in Ningxia during the past 10 years, and the relationships of TVDI with meteorological factors, meteorological drought index and agricultural disaster-affected area were also analyzed in this paper. The results showed that there was a typical triangular shape between the ΔLST and NDVI which is suitable for building the TVDI model. The dry edges were obtained by a least squares regression that had significant negative slopes (P<0.05), and the wet edges almost were constant isoline which was an optimal theoretical condition. The time series TVDI showed that the drought spatial characteristics were different in 3 typical ecosystem regions. Severe drought with the most frequency and high intensity usually occurred in the arid zone of middle Ningxia. The mountains area of southern Ningxia was moderate affected by drought, for drought occurred with lower frequency and intensity. By contrast, the Yellow River irrigation area of northern Ningxia was slightly affected by drought during the past 10 years, for there was a typical inland oasis and convenient irrigation condition from the Yellow River. By the way, the forest regions of Liupan Mountain in the south and Helan Mountain in the north were also less affected by drought because this ecosystem could adjust moisture by itself. From the perspective of whole Ningxia, there were 3 extreme drought events in 2000, 2005 and 2009 respectively that were detected by time series TVDI data, and this result was consistent with the previous studies. Based on the methods of the linear regression of TVDI and the ensemble empirical mode decomposition (EEMD), the change trends of drought during the past 10 years were detected respectively in annual and quarterly scales. The annual mean drought intensity was weakening from 2000 to 2010, but the monthly extreme drought event of each year was enhancing. The mean drought intensity in spring and summer was significantly enhancing, but in autumn and winter was significantly weakening. The correlation analysis results showed that the variation of TVDI mainly depended on the annual precipitation rather than annual average air temperature, and this phenomenon was related to the land use pattern and vegetation type in Ningxia. The disaster-affected area had a significant correlation with the annual mean drought intensity rather than monthly extreme drought event of each year. This phenomenon indicated that the long-time drought process, prolonged for more than one month, would impact the agricultural production in Ningxia. The seasonal results showed that summer drought was most likely to lead to the reduction of agricultural production, and then were the spring and autumn drought, while the winter drought had no effect on the agricultural production in Ningxia. This work can provide theoretical basis for the local government to make a proper strategy when combating drought, scheduling irrigation and providing disaster relief.