Mapping the sensitivity of agriculture to drought and estimating the effect of irrigation in the United States, 1950–2016

Junyu Lu, Gregory J. Carbone, Xiao Huang, Kirsten Lackstrom, Peng Gao

Research output: Contribution to journalArticlepeer-review

Abstract

Drought is a devastating natural hazard posing great threats to agriculture. Identifying the spatial pattern of agricultural sensitivity to drought can provide scientific information for decision-makers to prepare droughts, allocate resources, and mitigate impacts. Here, we use long-term state- and county-level crop data for the 10 major crops: corn grain, soybeans, hay, spring wheat, winter wheat, cotton, corn silage, sorghum, barley, and rice in the United States from 1950 to 2016. First, we perform a correlation analysis between crop yield anomalies and two drought indices (Standardized Precipitation Evapotranspiration Index (SPEI) and Standardized Precipitation Index (SPI)) to identify the sub-seasonal pattern of agricultural sensitivity to drought stress. SPEI performs better than SPI. For most crops, the sensitivity to drought increases in the early period, peaks at the critical months, and then declines. July is the most critical month for crop growth for most crops. Among all crops, soybean and corn grain are most sensitive to drought. Second, we develop an Agriculture Drought Sensitivity Index (ADSI) to quantitatively measure the sensitivity of agriculture to drought stress based on the statistical relationship between the ten major crops and SPEI. We demonstrate that there exists a very strong spatial correspondence between higher sensitivity to drought and the lower percentage of acres irrigated, and vice versa. Also, for those regions with limited irrigation, the sensitivity is higher in arid/semi-arid regions and lower in humid regions in summer. Third, given the importance of irrigation, an analysis of covariance (ANCOVA) shows that the irrigated crop yields have much higher long-run mean yields than non-irrigated crop yields. Fourth, to investigate how irrigation affects drought sensitivity, a panel data regression model shows that the responses of crop growth to drought are non-linear for all crops. Non-irrigated crops are more sensitive to droughts than the irrigated crops, particularly in severe drought conditions. This provides quantitative incentive to use irrigation as an important adaptation and coping strategy to mitigate the drought impacts on agriculture in the US.

Original languageEnglish (US)
Article number108124
JournalAgricultural and Forest Meteorology
Volume292-293
DOIs
StatePublished - Oct 15 2020

Keywords

  • Analysis of covariance (ANCOVA)
  • Drought risks mitigation
  • Gridded standardized precipitation evapotranspiration index (SPEI)
  • Gridded standardized precipitation index (SPI)
  • Panel data regression model

ASJC Scopus subject areas

  • Forestry
  • Global and Planetary Change
  • Agronomy and Crop Science
  • Atmospheric Science

Fingerprint Dive into the research topics of 'Mapping the sensitivity of agriculture to drought and estimating the effect of irrigation in the United States, 1950–2016'. Together they form a unique fingerprint.

Cite this