Multisource data based agricultural drought monitoring and agricultural loss in China
Qiang Zhang, Huiqian Yu, Peng Sun, Vijay P. Singh, Peijun Shi
[Qiang Zhang, Huiqian Yu, Peijun Shi] Key Laboratory of Environmental Change and Natural Disaster, Ministry of Education, Beijing Normal University, Beijing 100875, China.
[Qiang Zhang, Huiqian Yu, Peijun Shi] State Key Laboratory of Earth Surface Processes and Resource Ecology, Beijing Normal University, Beijing 100875, China.
[Qiang Zhang, Huiqian Yu, Peijun Shi] Faculty of Geographical Science, Academy of Disaster Reduction and Emergency Management, Beijing Normal University, Beijing 100875, China.
[Peng Sun] College of Territorial Resources and Tourism, Anhui Normal University, Anhui 241000, China.
[Vijay P. Singh] Department of Biological and Agricultural Engineering, Zachry Department of Civil Engineering, Texas A&M University, College Station, TX, USA.
Abstract: The Drought Severity Index (DSI) was computed, based on NDVI and ET/PET datasets from 2000 to 2014 over China, and was used for agricultural drought monitoring. Trends of droughts were characterized using the modified Mann-Kendall trend test, and the Pearson correlation method was used to determine the relation between the crop yield and drought-affected areas monitored by DSI. Results indicated the following: 1) DSI satisfactorily monitored agricultural droughts in both space and time, but could not decide the end of a drought, because drought-induced vegetation degradation could not recover in due time and generally a time lag of one or two months can be expected for the ending time of DSI-based droughts. 2) Two time intervals during the study period were identified with serious drought conditions, i.e. 2000–2001 and 2007–2009, and typical drought years, such as 2000, 2001, 2007, and 2009, were successfully monitored by DSI. Furthermore, droughts occurred mainly in northeastern, northern and southwestern parts of China. (3) Maize and wheat are the dominant crops in regions with high frequency of droughts and are hence under risk of drought hazards. Moreover, during the growing season maize and wheat are more sensitive to drought hazards and substantial agricultural losses can be expected due to droughts. (4) Agricultural irrigation introduced uncertainty in agricultural drought monitoring using drought indices. Therefore, no agricultural losses but frequent droughts were observed in northern China. This study provides a theoretical basis for application of DSI to agricultural monitoring across China and perhaps in other regions of the globe.
Published in: Global and Planetary Change, 2019, 172: 298-306. https://doi.org/10.1016/j.gloplacha.2018.10.017.
