Quantifying the relative importance of multiple drought indicators in the U.S. Drought Monitor as a function of location and time of year

The National Integrated Drought Information System (NIDIS) is tasked with providing dynamic,
accessible, and authoritative drought information for the Nation. Currently, the US Drought Monitor
(USDM) is widely recognized as the definitive source for synthesized monitoring of the onset, severity,
extent, and recovery of drought in the US. However, because the USDM is constructed using a
combination of objective and subjective methods by a centralized group of USDM authors (Svoboda et
al., 2002), there is no easy way for state and local stakeholders to know which individual indicators are
most representative of drought status. The recently developed Drought Risk Atlas (Svoboda et al.,
2015) is a step in this direction, because it allows users to plot the USDM alongside other conventional
indicators such as the Standardized Precipitation Index (SPI) and the Palmer Drought Severity Index
(PDSI) at selected points. Further, because the USDM considers drought indicators from a variety of
sources, the traceability of these individual indicators to the final product is not well-understood.

As stakeholders continue to make decisions based on the USDM, there is a desire for increased
quantitative understanding of the physical mechanisms that affect the determination of drought. For
example, as we move from meteorological, to agricultural, to hydrological drought, are there different
physical indicators that become more important (e.g., evapotranspiration, soil moisture, groundwater)?
Further, there has been a proliferation of drought indicators, and it is not clear to users or decision-
makers which set of physical indicators are most critical for a given location in a particular season
(Heim, 2002; Zargar et al., 2011). If local stakeholders can have information about which drought
indicators are most critical for monitoring and predicting evolving droughts in their areas of
responsibility, then they can better prepare for and mitigate existing droughts. Examples of physical
indicators for which there are multiple sources of data include:

● precipitation (e.g., SPI)
● evaporative demand/stress (e.g., Evaporative Drought Demand Index (EDDI), Evaporative Stress
Index (ESI))
● soil moisture (e.g., North American Land Data Assimilation System (NLDAS), Climate
Prediction Center (CPC), Soil Moisture Active Passive (SMAP))
● groundwater resources (e.g., U.S. Geological Survey (USGS), Gravity Recovery and Climate
Experiment (GRACE)-based drought indicators)
● runoff/streamflow (e.g., Standardized Runoff Index (SRI))
● vegetation (e.g., Vegetation Drought Response Index (VegDRI))
● snowpack (e.g., Snow Telemetry (SNOTEL))

Accordingly, the overarching goal of this project is to determine the roles of multiple physical indicators
in replicating the drought conditions in the USDM and quantify the information explained by each
indicator as a function of location and season. Knowledge of this information could eventually lead to
customized objective blends for monitoring or more optimized and accurate drought early warning
systems (DEWS).

Climate Risk Area: Water Resources