Quantifying Uncertainties in Drought Severity to Improve Drought Monitoring (NSF Award 2117433)

Impacts-Based Drought Thresholds for Corn and Soybean Yield in the U.S. Midwest (1982–2022) This dataset contains impacts-based drought thresholds derived from historical relationships between climate indicators and crop yield anomalies for corn and soybean across major agricultural counties in the U.S. Midwest. The thresholds were developed using a 41-year dataset (1982–2022) that integrates crop yield records with drought indicators including the Standardized Precipitation Index (SPI), Standardized Precipitation Evapotranspiration Index (SPEI), and soil moisture observations at multiple depths. Traditional drought monitoring frameworks often rely on fixed statistical thresholds (e.g., SPI < −1) that do not explicitly represent agricultural impacts. In contrast, this dataset identifies data-driven thresholds based on observed yield losses, allowing drought indicators to be linked directly to agricultural impacts. These thresholds can be used to improve agricultural drought monitoring, early warning systems, and climate risk assessments. Dataset Purpose The dataset was created to support research on impact-based drought monitoring, where drought severity categories are determined according to their observed effects on crop production. Specifically, the thresholds correspond to conditions where crop yields fall below specific statistical levels relative to long-term yield distributions. The dataset enables: Identification of drought conditions associated with yield losses below selected percentile levels Comparison between fixed drought thresholds and impacts-based thresholds Regional analysis of agricultural drought sensitivity across counties Development of agricultural drought early warning indicators Study Region The dataset focuses on the primary corn and soybean producing regions of the U.S. Midwest, including counties in: Illinois Indiana Iowa Ohio These states represent a large portion of U.S. grain production and are highly sensitive to drought conditions during the growing season. Data Sources The dataset integrates information from multiple sources: County-level crop yield data for corn and soybean obtained from the USDA agricultural statistics. Climate drought indicators, including SPI and SPEI, derived from long-term precipitation and evapotranspiration datasets (gridMET and PRISM). Soil moisture data at multiple depths representing root-zone moisture availability during the growing season (NLDAS-2). Methodology Overview Yield Data ProcessingCrop yields were detrended to remove technological and management improvements over time. Yield Percentile CalculationFor each county, annual yields were ranked and converted to percentile values within the 1982–2022 period. Impact Threshold DefinitionYield impact categories were defined using percentile thresholds representing increasing severity of agricultural impacts: 20th percentile – moderate yield stress 10th percentile – significant yield loss 5th percentile – severe yield loss 2nd percentile – extreme yield loss Climate Indicator AnalysisCorresponding drought indicator values (SPI, SPEI, soil moisture) were analyzed during critical growing season months (particularly July–August). Impacts-Based Threshold EstimationStatistical relationships between yield percentiles and drought indicators were used to derive indicator thresholds that correspond to specific levels of crop yield loss. Applications This dataset can support research and applications, including: Agricultural drought monitoring Climate risk assessments Crop yield forecasting Drought early warning systems Development of impacts-based drought indices Citation If you use this dataset, please cite: Teleubay, Z., Quiring, S.M., Leasor, Z., 2025. Estimation of impacts-based drought thresholds for the U.S. Corn Belt. Agricultural and Forest Meteorology 372, 110715. https://doi.org/10.1016/j.agrformet.2025.110715 ______________________________________________________________________________________________________________________________________________________________________________________ Effects of Climate Regime and Nonstationarity on Drought Severity Classifications This dataset quantifies how climate regime and temporal nonstationarity influence drought severity classification using widely adopted drought indices, including the Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI). The dataset is derived from three global climate datasets and spanning 1901 - 2016. Dataset Purpose Evaluation of discrepancies between theoretical and observed drought frequencies (SPI_frequencies_TC2.mat, SPEI_frequencies_TC2.mat) Objective drought severity thresholds (thresh_CRU_TC2mat,thresh_GPCC_TC2mat,thresh_UDEL_TC2mat,thresh_SPEI_TC2mat) Drought classification bias (bias_SPI.mat) Study Region This study utilizes three global precipitation datasets at a 0.5° spatial resolution. This study is based on 1901-2016 because this is a common period of record for the three datasets. The datasets used are the Climate Research Unit (CRU) Time Series Version 4.03 , the University of Delaware (UDEL) Terrestrial Precipitation Gridded Monthly Time Series Version 5.01, and the Global Precipitation Climatology Centre (GPCC) Full Data Monthly Product Version 2018. Methodology Overview (methods_GODT.m) Drought Index CalculationSPI and SPEI are computed at 1-, 3-, 6-, and 12-month timescales. Expected vs. Observed Frequency AnalysisObserved drought category frequencies were compared to theoretical expectations based on the assumed probability distributions. Nonstationarity AssessmentDrought thresholds were evaluated across different time periods to quantify the effects of temporal nonstationarity on drought classification. Bias QuantificationDifferences between expected and observed drought frequencies were used to identify systematic biases in drought severity classification. Applications This dataset can support research and applications, including: Agricultural drought monitoring Climate risk assessments Crop yield forecasting Drought early warning systems Development of impacts-based drought indices Citation If you use this dataset, please cite: Leasor, Z., Teleubay, Z., and S.M. Quiring (2026). Effects of Climate Regime and Non-stationarity on Drought Severity Classifications