Identifying the spatio-temporal pattern of drought characteristics and its constraint factors in the Yellow River Basin

The Yellow River Basin (YRB) constitutes an important grain-producing area, as well as an economic core area in China. However, drought and water shortage have always been important factors restricting the agricultural sustainable development of this region. The intensity of agricultural drought in the YRB and the factors affecting it need to be urgently clarified. Based on this, this study attempts to assess the spatio-temporal characteristics of agricultural drought in the YRB using the improved traditional temperature vegetation dryness index (ITVDI) and to identify the type and intensity of constraint effects of human activities (population, GDP, land use type), climate change (temperature, precipitation) and topography (elevation, slope) on drought in the YRB with the constraint line method. Results appeared that from 2001 to 2015, the drought intensity in the YRB exhibited fluctuating changes, while it decreased after 2015. Spatially, the drought characteristics of the YRB existed large heterogeneous, with drought showing a significant increase in the northwestern and downstream areas and a significant decrease in the central part of the basin. There were eight types of constraining effects of human activities, climate change, and topographic factors on drought intensity in the YRB, including U-shaped curve, positive linear, negative linear, logarithmic, negative convex, positive convex, exponential, and hump-shaped. Precipitation and temperature were the most direct factors affecting drought intensity and would exert a binding effect on both the upper and lower boundaries of drought intensity. The constraint effect of precipitation on drought varied in space. In the upstream and midstream, precipitation constrained the upper and lower boundaries of ITVDI, and the constraint effect increased with the increase of precipitation. In the downstream, the constraint effect of precipitation did not change significantly, and the maximum value of ITVDI was always less than 0.8. Temperature had upper and lower boundary constraint effects at three scales. The constraining effects of elevation, slope, GDP, and population were stable across scales. It should be noted that the constraint effect of land use type on drought exhibited U-shaped, and the threshold point of minimum values appeared in the areas covered by grassland and water. The constraint line method can improve the drought monitoring ability of ITVDI, as well as help to clarify the constraint factors and mechanism of drought intensity in the YRB and provide scientific decision-making support to mitigate regional drought and promote regional sustainable development.