PowerPoint Presentation

Quick Drought Response Index (QuickDRI) A Composite Index for Monitoring Rapid-Onset Agricultural DroughtTsegaye Tadesse1, A/ProfessorClimatologist/Remote Sensing Expert, UNL/NDMCBrian Wardlow2, Mark Svoboda1, Jesslyn Brown3, Martha Anderson4, Chris Hain5, Matt Rodell6, and David Mocko6

1National Drought Mitigation Center (NDMC), University of Nebraska-Lincoln2Center for Advanced Land Management Information Technologies (CALMIT), University of Nebraska-Lincoln3USGS Center for Earth Resources Observation Science (EROS)4USDA Agricultural Research Service (ARS)5Earth System Science Interdisciplinary Center, University of Maryland6NASA Goddard Space Flight Center (GSFC)

2015 U.S. Drought Monitor Forum Reno, NevadaApril 14-16, 2015

Quick Drought Response Index (QuickDRI)QuickDRI is a hybrid drought index that monitors rapid, short-term changes in agricultural drought conditions through the integration of: - satellite-based observations of vegetation conditions- evapotranspiration (ET) estimates from satellite- root-zone soil moisture (satellite-estimated or modeled) - climate-based drought index data - biophysical characteristics of the environment.

Goal: Use recently available remote sensing products that are shorter-term indicators of drought-related environmental conditions to develop complimentary, operational tool over the CONUS to VegDRI that characterizes shorter-term, rapid-onset agricultural drought conditions on the order of weeks to a month.

Vegetation Drought Response Index (VegDRI)VegDRI is a composite drought index that integrates: - satellite-based observations of vegetation conditions - climate-based drought index data - biophysical characteristics of the environmentto produce 1-km spatial resolution maps that depict drought-related vegetation stress.

Timeline/Milestones:2002: USGS-funded proof-of-concept work2005: Development of operational VegDRI for CONUS (USDA RMA and NASA)2009: Completion of operational VegDRI tool and products

3

VegDRI ResultsUsed by the operational USDM map development processTechnology transfer to other regions of the world.CanadaCentral Europe (Czech Republic prototype)Greater Horn of Africa/EthiopiaIndiaMexicoChinaSouth KoreaVegDRI has proven to be a useful indicator of seasonal agricultural drought conditions, but has limited ability to characterize rapid, short-term changes in conditions:inherent lag of NDVI response to drought stress longer time interval of climate data inputs

Canada-US VegDRIJuly 30, 2002

Czech Republic VegDRI PrototypeJune 22, 2014

Ethiopia/ Greater Horn of AfricaVegOut PrototypeAugust, 2011

Emerging Satellite-based Drought Monitoring ToolsOver the past 10+ years, there has been a rapid development of remote sensing-based drought monitoring tools characterizing different parts of the hydrologic cycle that influence drought conditions.

ExamplesEvaporative Stress Index (ESI)GRACE Terrestrial Water Storage (TWS) anomaliesNLDAS soil misture anomalies

QuickDRI MethodologyIndicators responding on shorter time scales (1 to 2 months)

QuickDRI models developed from analysis of 12-year time-series data of dynamic variables over CONUS for period of 2000 to 2012.Biophysical variables static over the entire history. QuickDRI Training Database Development

Locations of the 2,427 weather stations used to extract the training data for the QuickDRI models.Vegetation IndexETSoil MoisturePrecip.Biophysical

Example of QuickDRI Model from CubistExample of QuickDRI models output from regression tree analysis (Cubist) of historical input variables.Dependent variableSPEI (1 month)

Independent variables:SPI (1 month)ESI anomaly (4 week)SVIVIC soil moistureStart of Season Anomaly (SOSA)

Static independent variables:LULCIrrigation statussoil AWCelevationecoregion

June 26, 2012

June 25, 2012USDMVegDRIFigure 1. Comparison of USDM and VegDRI vs. QuickDRI (Model R_N) for end of June 2012.

QuickDRI Model Results for late-June 2012(Drought Year)QuickDRI_June 30, 2012

vma_2012_w26sosa_2012svi_12wk26Figure 2. Comparing and contrasting the input climate and satellite data with QuickDRI (Model R_N) map at the end of June 2012.es08_2012w26spi1_2012_26

es04_2012w26

QuickDRI_June 30, 2012

awc_1kmdem_1kmus_econlcd06_1kmmirad07_1kmFigure 3. Contrasting the input static biophysical parameters with QuickDRI (Model R_N) map at the end of June 2012.

QuickDRI_June 30, 2012

July 29, 2012

July 23, 2012USDMVegDRIFigure 4. Comparison of USDM and VegDRI vs. QuickDRI (Model R_N) for end of July 2012.

QuickDRI_July 30, 2012QuickDRI Model Results for late-July 2012(Drought Year)

vma_2012_w30sosa_2012svi_12wk30Figure 5. Comparing and contrasting the input climate and satellite data with QuickDRI (Model R_N) map at the end of July 2012.es04_2012w30es08_2012w30

spi1_2012_30

spi2_2012_30

QuickDRI_July 30, 2012

awc_1km

dem_1km

us_eco

nlcd06_1kmmirad07_1kmFigure 6. Contrasting the input static biophysical parameters with QuickDRI (Model R_N) maps at the end of July 2012.

QuickDRI_July 30, 2012

Aug. 28, 2012

Aug. 20, 2012USDMVegDRIFigure 7. Comparison of USDM and VegDRI vs. QuickDRI (Model R_N) for end of August 2012.

QuickDRI Model Results for late-Aug. 2012(Drought Year)QuickDRI_August 27, 2012

vma_2012_w34

sosa_2012svi_12wk34Figure 8. Comparing and contrasting the input climate and satellite data with QuickDRI (Model N) map at the end of August 2012.es08_2012w34spi1_2012_34spi2_2012_34

QuickDRI_Aug. 30, 2012es04_2012w3417

awc_1km

dem_1km

us_eco

nlcd06_1kmmirad07_1kmFigure 9. Contrasting the input static biophysical parameters with QuickDRI (Model R_N)) map at the end of August 2012.

QuickDRI_August 27, 2012

Example of Early-Onset Agricultural Drought Development Detected by QuickDRI

Selection of Benchmarking Areas for QuickDRI Evaluation and Validation

Benchmarking areas and time periods where earlier indicators of drought would have been valuable to the USDM will be selected in consultation with USDM authors and other drought experts.

Benchmark areas for the 2011 drought over U.S. Southern Great PlainsSpatial Subset of QuickDRI inputs and results, VegDRI, and USDM over Benchmark Area #4

Targeted Applications for QuickDRIUSDA Livestock Forage Disaster Program (LFP)

The 2008 and 2014 Farm Bills mandated the USDA Farm Service Agency (FSA) RFP use the USDM as the primary trigger to establish county-level eligibility for financial assistance to cover drought-related grazing losses. U.S. Drought Monitor (USDM)

Current state-of-the-art drought monitoring tool for the U.S. that is used in federal and state programs and by the media to communicate drought conditions.

Composite index that incorporates various types of data through expert analysis by drought experts.

Challenge: Increasing demand to use the USDM for county- to subcounty-level decisions requiring higher spatial resolution information to characterize more spatially-detailed drought patterns.

Current QuickDRI ActivitiesTesting a subset of QuickDRI models for several flash drought events (e.g., 2011 and 2012), as well as for other drought and non-drought periods.

Collecting drought-related information (e.g., drought impacts, in situ observations, and expert insights) that characterizes the development and impact of flash drought events for benchmark areas to compare the QuickDRI model results.

Evaluating how QuickDRI results would improve the targeted decision making processes (USDM and USDA Range Forage Program) for retrospective flash drought events.

Engaging a network of key drought experts and decision makers to evaluate QuickDRI results and define informational products and other applications.

Developing operational QuickDRI processing system (at USGS EROS) and data/product dissemination mechanisms.

Questions?