If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact us at firstname.lastname@example.org.
Identifying Indicators and Timescales of Agricultural Drought in the Northeast United States using Crop Area-weighted Drought Indices
AltmetricsView Usage Statistics
Critical indices of agricultural drought in the Northeast United States (Northeast) are examined with a particular focus on the drought events in 2016 and 2020. We evaluate relationships of drought indices that are based on precipitation (PRCP), soil moisture (SM), and evaporative demand (E0) and their optimal timescales to agricultural-impacts data. We resampled forty years (1981-2020) of these daily climate drivers onto crop-cover maps provided by the National Agricultural Statistics Service (NASS) and computed standardized drought indices, namely the Standardized Precipitation Index (SPI), the Standardized Soil Moisture Index (SSMI), and the Evaporative Demand Drought Index (EDDI). Further indices that reflect actual evapotranspiration (ET) and the balance between PRCP and E0 — the Landscape Evaporative Response Index (LERI) and the Standardized Precipitation-Evapotranspiration Index (SPEI), respectively--were also included. Indices were estimated at timescales ranging from one week to twelve months. All climate drivers were averaged to state and county levels before their use in estimating the drought indices. Five of the most widely cultivated crops in the Northeast were selected--the most prominent being hay and pasture. Forty years (1981-2020) of annual hay yield and twenty-six years (1995-2020) of weekly pasture-condition records were retrieved from NASS. Hay yields were detrended and correlated to each set of drought indices, while pasture conditions were compared to the progression of indices within the drought years of 2016 and 2020. Results show the strongest correlation (Rmax) around 0.5-0.7 between drought indices and hay yield, with SPI correlations being positive and EDDI correlations negative. The Rmax values and associated distribution of dates and timescales were derived for states and counties and exhibited distinct regional spatial variation. Hay yield showed the strongest response to shorter timescales for SSMI and EDDI than for SPI and SPEI. Weekly pasture conditions in most states exhibited a strong response to the 2016 and 2020 droughts. Characteristics of how the degradation of pasture through the growing season tracks the progression of drought indices across different states were examined. Respectively, pasture conditions show the strongest response to 1- to 2-week SSMI, 2-week to 1-month EDDI, and 1- to 3-month SPI. At the 1-week timescale, fluctuations of LERI generally tracked fluctuations of SSMI throughout both 2016 and 2020. The times when major pastureland degradations began were usually accompanied by abnormally low LERI, low SSMI, and high EDDI, regardless of SPI.