Retiming Evaporative Stress Index to Vegetation Phenology in Iowa Croplands

Vegetation health indices can be used to infer productivity, which is useful for yield estimation and land surface modeling. The Evaporative Stress Index (ESI) is an estimation of vegetation water stress. ESI is the standardized anomaly of evapotranspiration (ET) relative to potential ET (fPET), and is calculated operationally over the continental United States (CONUS) by NOAA and globally by NASA to support drought and crop monitoring on two, four, eight, and twelve week products. ET is estimated by the Atmosphere–Land Ex-change Inverse (ALEXI) model using thermal infrared observations of Earth’s surface from the Geostationary Operational Environmental Satellites (GOES), meteorological data from the Climate Forecast System Reanalysis, and leaf area index estimation from MODIS/VIIRS. Negative ESI is indicative of reduced ET, higher surface temperature, and water stress, while positive ESI generally indicates well-watered and healthy vegetation conditions. Currently, ESI is computed considering climatological anomalies in ET based on day of year; however, over highly managed agricultural landscapes, small shifts in planting and emergence date can introduce false anomalies due to differences in phenology rather than moisture status. This study aims to retime these climatologies and anomalies based on vegetation development rather than calendar date. Vegetation emergence (also known as greenup) is estimated by the Terra and Aqua fusion MODIS land cover dynamics yearly product, which is available from 2001 to 2022. Initial testing is conducted over sites in Iowa, USA, with contrasting characteristic green-up dates and annual fPET curve widths. Vegetation emergence is often around 50 days later in cropland than in forests, as farmers may wait on soil to warm and dry before planting. There is also considerable variability in greenup date between years in one location, which can be around 20 to 30 days. When timed to greenup, ET rises at a more consistent part of the year relative to timing by day of year. When annual fPET curves are aligned on greenup date, variability in ESI is reduced, and correlations U.S. Drought Monitor increase. Impacts of modifying the time unit post-greenup from calendar day to degree day is also evaluated. These findings will be used to improve the information content in the operational national and global ESI products, better capturing actual stress signals over managed landscapes. We have found that retiming indices to phenology improves agreement in highly agricultural areas, while retiming to phenology does not significantly affect agreement between vegetative indices in a mixed cropland - forested area. Phenological ESI can be computed across CONUS, and there is the potential for this product to be posted operationally. In croplands, National Agricultural Statistics Services (NASS) crop condition reports are compared to ESI, shedding light on the condition of individual crops within a county. This study also aims to focus on ESI in grazinglands, as changes in management may result in grass recovery and productivity. ALEXI can also be run with microwave inputs for land surface temperature, offering another flavor of ESI.