Temporal and Spatial Changes of Drought in Beijing-Tianjin-Hebei Region Based on Remote Sensing Technology

Abstract

Abstract. Drought is an extremely complex natural disaster phenomenon. Sustained drought will lead to the aggravation of water shortage, food production reduction, land desertification and ecological crisis, which will have a great impact on social and economic development, industrial and agricultural production and ecological environment. In recent years, human activities have intensified, the global climate has been warming, and the frequency and intensity of extreme events such as drought have been continuously strengthened, which has been widely concerned by scholars at home and abroad. The Beijing-Tianjin-Hebei region, as the political and cultural center of China, is seriously affected by the drought, which has a certain impact on the social economy of the region. Therefore, it is of great practical significance to carry out remote sensing drought monitoring and understand the temporal and spatial characteristics of drought. Based on MOD13A2 and MOD11A2 in MODIS products, this paper constructed the NDVI-Ts feature space to obtain the Temperature Vegetation Dryness Index (TVDI) in Beijing-Tianjin-Hebei region, and analyzed the spatial and temporal variation characteristics of drought in Beijing-Tianjin-Hebei region from 2001 to 2014 with TVDI as the drought index. The results showed that: (1) the NDVI-Ts feature space in the Beijing-Tianjin-Hebei region showed that when the value of NDVI rose from 0 to 1, the maximum temperature of the surface gradually decreased, while the minimum temperature gradually rose. Both of them had a linear relationship with NDVI, and their spatial shape was similar to a triangle, indicating that the study of drought in Beijing-Tianjin-Hebei was effective by using this spatial feature; (2) the Beijing-Tianjin-Hebei region experienced drought every year in the period of 2001–2014, and the drought was more serious in 2001, 2004, 2008, 2009, 2011 and 2012; (3) The drought in the Beijing-Tianjin-Hebei region was widespread, and the severely arid regions were mainly concentrated in the southern and central regions. In terms of drought intensity, the most frequent occurrences in the period of 2001–2014 were Xingtai, Hengshui, Shijiazhuang, and Handan, and the annual drought in 14 years; the least number of occurrences was Chengde, a total of 4 times; the number of occurrences was second only to Chengde, Tangshan, Zhangjiakou, Qinhuangdao, a total of 7 times. Our results indicated that the Beijing-Tianjin-Hebei region was prone to drought and the drought degree was severe.

Trend Prediction of Vegetation and Drought by Informer Model Based on STL-EMD Decomposition of Ha Cai Tou Dang Water Source Area in the Maowusu Sandland

“…The complexity of model optimization and the limitations of computational speed remain the main challenges for machine learning models, such as random forest, support vector machine (SVM), and support vector regression (SVR), which are currently widely used to predict trends despite their effectiveness in predicting vegetation metrics [14]. For instance, Lin et al used LST and the NDVI to characterize large-scale drought conditions [15]. The correlation with soil moisture can be better derived through the TVDI, and this is widely used in remote sensing drought, which is suitable for long time-series analysis [15]; Gidey et al effectively improved the NDVI in-version accuracy through the use of the SVR algorithm, although the model has several drawbacks, including a slow convergence speed and a highly complex optimization process [16].…”

“…For instance, Lin et al used LST and the NDVI to characterize large-scale drought conditions [15]. The correlation with soil moisture can be better derived through the TVDI, and this is widely used in remote sensing drought, which is suitable for long time-series analysis [15]; Gidey et al effectively improved the NDVI in-version accuracy through the use of the SVR algorithm, although the model has several drawbacks, including a slow convergence speed and a highly complex optimization process [16]. Chen et al introduced an SVM model to simulate the quantification of the effect of LST on vegetation cover, but there were machine learning shortcomings.…”