Evaporative Demand Drought Index (EDDI)

About

latest realtime map

What is EDDI?

The Evaporative Demand Drought Index (EDDI) is an experimental drought monitoring and early warning guidance tool. It examines how anomalous the atmospheric evaporative demand (E0; also known as "the thirst of the atmosphere") is for a given location and across a time period of interest. EDDI is multi-scalar, meaning that this period—or "timescale"—can vary to capture drying dynamics that themselves operate at different timescales; we generate EDDI at 1-week through 12-month timescales.

This webpage offers a frequently updated assessment of current conditions across CONUS, southern parts of Canada, and northern parts of Mexico; a tool to generate historical time series of EDDI for a user-selected region; introductions to the EDDI team; and a list of resources for users to explore EDDI and its applications further.

Why use EDDI?

EDDI can offer early warning of agricultural drought, hydrologic drought, and fire-weather risk by providing near-real-time information on the emergence or persistence of anomalous evaporative demand in a region. A particular strength of EDDI is in capturing the precursor signals of water stress at weekly to monthly timescales, which makes EDDI a strong tool for preparedness for both flash droughts and ongoing droughts.

How often is EDDI updated?

Currently, EDDI is generated daily—though with a 5-day lag-time—by analyzing a near-real-time atmospheric dataset. This lag-time results from the procedures to quality control the meteorological data used to estimate evaporative demand. There is also an ongoing effort to forecast EDDI based on seasonal climate-forecast information.

Acknowledgements

This work is supported in part by grants from (i) NOAA's Joint Technology Transfer Initiative (JTTI) for the project titled "Operationalizing an Evaporative Demand Drought Index (EDDI) service for drought monitoring and early warning;" (ii) NOAA's Sectoral Applications Research Program (SARP): Coping with Drought in Support of the National Integrated Drought Information System (NIDIS) program for the project titled "Developing a wildfire component for the NIDIS California Drought Early Warning System;" (iii) DOI's North Central Climate Science Center for the project (Grant #G14AP00182) titled "Ecological Drought, Climate Extremes and the Water Cycle across Timescales;" and (iv) Western Water Assessment, an NOAA RISA program, for the project titled "Enhancing the usability of EDDI," with funding originating from NIDIS.

Any issues with accessing the plots and other information on this page are welcome and should be sent to esrl.psd.data@noaa.gov.

NIDIS logo NOAA logo WWA logo CIRES logo CU logo DRI logo NCCSC logo

Plot EDDI Maps of the Continental U.S.


If you use these plots in publications, we ask that you acknowledge the Physical Sciences Division. For example: "Image provided by the NOAA/ESRL Physical Sciences Division, Boulder, Colorado, from their web site at: https://www.esrl.noaa.gov/psd/."
Current Synopsis for October 10, 2018*
During the week ending on October 10 (1-week EDDI), the pattern of atmospheric evaporative demand divided CONUS into two distinct regions: across the Western US, Great Plains and northern tier of states from WA to MI, the atmosphere was significantly wetter than normal, (evaporative demand lower than normal), with near-record wetness (EW4, or evaporative demand at or below the 2nd percentile) observed across a wide region including eastern MN, IA, the Dakotas and NE, and parts or all of CO, UT, WY, MT, and the interior Pacific Northwest. CO, UT, northeastern NV, MT. This represents significant, multi-category improvement over the central western US, from the High Plains of CO west to eastern CA and from northeastern NM north and west to southeastern ID, and including western WA. With some minor drier-than-normal exceptions of ED0 (70-80%ile) and ED1 (80-90%ile) in the Chihuahuan Desert of southern NM and western TX, the Mojave Desert of southeastern CA, and the Sacramento Valley of northern CA, the remainder of this region is wetter than normal. The remainder of CONUS, including the southern Midwest the southeastern states, and eastern US to the west of the Appalachians appear drier than normal at this timescale, with isolated hotspots of ED3 (95-98%ile) around southern MO and IL, southern OH, eastern TN and KY, northern WV, northwestern PA, and far-southern FL. The region to the east of the Appalachians appears predominantly neutral (evaporative demand at 30-79%ile) from the Carolinas to ME. Note that the recent heavy rains in the eastern US from FL to NJ subsequent to the October 10 landfall of Hurricane Michael are not reflected on this map.

For the month ending on October 10 (1-month EDDI), conditions across CONUS show continued intense dryness (ED4, or >98%ile) persisting across a region that includes almost all of UT, parts of southwestern WY, the western half of CO, parts of southern NV and eastern CA. This region is surrounded by a larger region of ED3-ED2 (90-95%ile) that extends over about twice the area, including all of eastern CO, parts of northeastern NM, the southwestern quadrant of WY, and southeastern ID. In this region, conditions are improving over CA, NV, southern ID, WY, and western areas OR and WA. Meanwhile, conditions around the lower Colorado River (southern NV, southeastern CA, and southwestern AZ), a region of southeastern CO and northeastern NM are degrading. Another area of abnormal dryness generally at ED0 (70-80%ile) to ED1 (80-90%ile), stretches in patches from MO northeastward across the Ohio Valley into western NY. Conditions here are deteriorating relative to last week. Also dry and deteriorating is FL, where dryness peaks at ED3 in far-southern parts around Miami. Note again, that this period does not include the effects of the landfall of Hurricane Michael on October 10. Wetter than normal or neutral conditions prevail across much of the rest of CONUS—particularly in the Gulf Coast of TX, central MT, the southern Great Plains, IA and southern MN, and the mid-Atlantic Seaboard.

The seasonal timeframe (3-month EDDI) ending on October 10 shows strong lingering drought in two regions of the west: a region from central CA, where ED4 (>98%ile) conditions prevail, stretching northward at ED1 (80-90%ile) to ED2 (90-95%ile) up across CA, northwestern NV, southern ID, most of OR, and through western WA to Canada; meanwhile, the drought in CO, UT, and southwestern WY lingers at ED3-ED4 (>95%ile), and stretches at ED1-ED2 (80-95%ile) across western CO, eastern UT, southwestern WY, and southern ID. Strong and lingering drought remains over northern NY (ED3, or 95-98%ile), and across northern New England (peaking at ED2, or <95%ile) in eastern ME. Southern MO remains in drought at ED0-ED2 (70-95%ile). Other regions in drier-than-normal conditions include, the Red River of the North in eastern ND and northwestern MN (peaking at ED2, or 90-95%ile), and peninsular FL, where ED2 (90-95%ile) conditions now cover the peninsula, a deepening of drought over last week. Wetter-than-normal conditions continue to prevail across a region stretching from southern MT across northeastern WY, down the central Great Plains, and a small region of wetter-than-normal conditions around the DC area. The most significant changes since this time last week appear to be a lessening in intensity around the drought area of UT and CO, in NY and northern New England, and along the northern CA coast. While the dryness in peninsular FL appears to be increasing, this does not reflect any recent impacts from Hurricane Michael which made on October 10, after these latest data became available.
Updated: October 16, 2018
*The latest date for which data were available, generally 5 days prior to update.

Description of Maps and Underlying Data

The EDDI maps displayed here use atmospheric evaporative demand (E0) anomalies across a timescale of interest relative to its climatology to indicate the spatial extent and severity of drought. This page provides access to near-real-time (with a five-day latency, i.e., the most recent information is five days old) EDDI plots with timescales that measure E0 anomalies across the 1 to 12 weeks and 1 to 12 months prior to the most current date. The colors indicate the frequency at which the observed E0 anomaly has occurred in the climatology, with warm colors indicating conditions that are drier than normal and cool colors indicating wetter-than-normal conditions. As an example, the ED4 category indicates that the current E0 anomaly has only been observed less than 2% of the time in the past 38 years (1979-2016), which represents the most severe drought conditions; the EW4 category means indicates that the anomaly has been exceeded 98% of the time, which represents the wettest conditions. For plotting purposes, EDDI values are binned into different percentile categories analogous to the US Drought Monitor plots—however, in case of EDDI plots, both drought and anomalously wet categories are shown.

E0 is calculated using the Penman Monteith FAO56 reference evapotranspiration formulation (0.5m tall reference crop), driven by temperature, humidity, wind speed, and incoming solar radiation from the operational North American Land Data Assimilation System (NLDAS-2) dataset. For a particular time-window, EDDI is estimated by standardizing the E0 anomalies relative to the the same accumulation time-window in the whole period of record (1979-present), using a rank-based non-parametric method described in Hobbins et al. (2016). EDDI data are available at a ~12-km resolution (0.125° lat and long) across CONUS since January 1, 1980, and are updated daily.

For more information regarding these plots and access to data, including customized EDDI maps, please contact Mike.Hobbins@noaa.gov (Phone: 303-497-3092)

This is a Research and Development Application

Plot EDDI Time Series of the Continental U.S.

EDDI plot for Boulder, CO

This webtool allows a user to generate historical (1980-latest complete year) timeseries data of the Evaporative Demand Drought Index (EDDI) for a specified region in the Continental United States. The time series is generated as a table for different timescales, i.e. 1 to 12 months of integrated evaporative demand at the end of a given month. This tool also allows users to generate time series plots with user specified timescales.

1 Region
Drag to move map; SHIFT-Drag to select region
N
W   E
S
2 Plot Options
Averaging Period (Months)? Ending Month?
(ending month includes values through the end of that month)
Enter Region Title (OPTIONAL: Used in the plot title: default is the lat/lon range)

Description of Time-Series Data and Plot

This page generates historical (1980-latest complete year) time-series data of EDDI across a user-selected region and timescale. You can select your region as a rectangular area by directly entering latitude and longitude values or by selecting the area in the adjacent map. The main output is a table with EDDI time series for timescales of 1 to 12 months. A plot will also be generated with user- (or default-) entered month and averaging period, which will accompany another table showing the time-series data with user-defined specifications. On the output page, you will have an option to readily replot EDDI time series with different specifications.

To estimate evaporative demand (E0), this tool uses daily reference evapotranspiration based on the Penman Monteith FAO56 method for tall (0.5m) reference crop that is available from the the operational North American Land Data Assimilation System (NLDAS-2) dataset. The spatial resolution of the data is 1/8th degree in latitude and longitude (i.e. approximately 12km). Daily E0 values are then aggregated to monthly values, and spatial averaging is done before monthly EDDI values are computed. At any point in time, there is large spatial variability in EDDI across CONUS. For generating these time series for meaningful assessment, we generally recommend that users select as small an area as possible (e.g., county, climate division, small watersheds), including point locations (i.e., by using a single latitude value in both the N and S boxes, and a single longitude value in both E and W boxes. for bounding the region, thereby getting the time series for just a single representative 12km x 12km pixel).

The main output table also includes reference evapotranspiration value (in mm) for each month used to calculate EDDI. The absolute values of E0 are highest during the warm season, during which there is also a heightened risk of water stress on the land surface in water-limited regions. For drought-related impacts, users can use time series of different EDDI timescales to compare with their own historical impacts data.

For further assistance or technical advice on the use of these timeseries data for research applications please contact Imtiaz.Rangwala@noaa.gov and Candida.Dewes@noaa.gov.

This is a Research and Development Application

EDDI Map Archive

Select skipping size:   1 week  30 days
FTP ASCII Data for This Plot All Data & Images
If you use these plots in publications, we ask that you acknowledge the Physical Sciences Division. For example: "Image provided by the NOAA/ESRL Physical Sciences Division, Boulder, Colorado, from their web site at: https://www.esrl.noaa.gov/psd/."

Description of Maps and Underlying Data

The EDDI maps displayed here use atmospheric evaporative demand (E0) anomalies across a timescale of interest relative to its climatology to indicate the spatial extent and severity of drought. This page provides access to historic and near-real-time (with a five-day latency, i.e., the most recent information is five days old) EDDI plots with timescales that measure E0 anomalies across the 1 to 12 weeks and 1 to 12 months prior to the most current date. The colors indicate the frequency at which the observed E0 anomaly has occurred in the climatology, with warm colors indicating conditions that are drier than normal and cool colors indicating wetter-than-normal conditions. As an example, the ED4 category indicates that the current E0 anomaly has only been observed less than 2% of the time in the past 38 years (1979-2016), which represents the most severe drought conditions; the EW4 category means indicates that the anomaly has been exceeded 98% of the time, which represents the wettest conditions. For plotting purposes, EDDI values are binned into different percentile categories analogous to the US Drought Monitor plots—however, in case of EDDI plots, both drought and anomalously wet categories are shown.

E0 is calculated using the Penman Monteith FAO56 reference evapotranspiration formulation (0.5m tall reference crop), driven by temperature, humidity, wind speed, and incoming solar radiation from the operational North American Land Data Assimilation System (NLDAS-2) dataset. For a particular time-window, EDDI is estimated by standardizing the E0 anomalies relative to the the same accumulation time-window in the whole period of record (1979-present), using a rank-based non-parametric method described in Hobbins et al. (2016). EDDI data are available at a ~12-km resolution (0.125° lat and long) across CONUS since January 1, 1980, and are updated daily.

Click here for access to these CONUSwide plots and here for access to the CONUSwide data. For more information regarding the plots and data, including arranging for customized EDDI maps, please contact Mike.Hobbins@noaa.gov (Phone: 303-497-3092).

This is a Research and Development Application

EDDI Project Team

Joe Barsugli

Joe Barsugli  •  joeseph.barsugli@noaa.gov  •  303-497-6042

Joe is a Research Scientist at CIRES and NOAA’s Physical Sciences Division. Trained in climate theory and modeling, he works at the technical interface connecting climate science with the practitioners and technical staff who are informing planning for water and land management in the Colorado region, and connecting researchers to the problems faced by managers.
Candida Dewes

Candida Dewes  •  candida.dewes@noaa.gov  •  303-497-4236

Candida is a Research Scientist at NOAA’s Physical Sciences Division and CIRES/Western Water Assessment at the University of Colorado, Boulder. She has extensive research experience in climate variability and climate change and their impacts on socio-ecological systems. Her current research focuses on regional-scale land surface processes contributing to drought, and in particular, the variability of evaporative demand under climate change. She is also interested in the role of rain versus snow in mountainous terrain how these precipitation types impact regional water resources.
Mike Hobbins

Mike Hobbins  •  mike.hobbins@noaa.gov  •  303-497-3092

Since obtaining his Ph.D. in Hydrologic Science and Engineering from Colorado State University in 2004, Mike has worked in research into evapotranspiration, evaporative demand, and drought. As a Research Scientist for NOAA’s Physical Sciences Division and the Cooperative Institute for Research in Environmental Sciences (CIRES) at the University of Colorado in Boulder, CO, his recent work supports drought early warning across the US for the National Integrated Drought Information Systems (NIDIS) and famine early warning across the globe for the Famine Early Warning Systems Network (FEWS NET), including the development and dissemination of reanalyses of evaporative demand; the development of the Forecast Reference Evapotranspiration (FRET) product for daily and weekly evaporative demand forecasts across the US; and the development of the EDDI.
Justin Huntington

Justin Huntington  •  justin.huntington@dri.edu  •  775-673-7670

Justin Huntington is an associate research professor of Hydrology at the Desert Research Institute, Reno, Nevada. His research interests are focused on remote sensing, land surface energy balance measurement and modeling, drought monitoring, and hydrologic modeling. His research primarily supports water use, water demand, and drought mapping and prediction efforts funded by the U.S. Bureau of Reclamation, U.S. Geological Survey, U.S. Bureau of Land Management, NASA, NOAA, and Google. He is one of 25 members of the 2012-2017 Landsat Science Team.
Jeff Lukas

Jeff Lukas  •  lukas@colorado.edu  •  303-735-2698

Jeff is a Research Integration Specialist with the Western Water Assessment program at CIRES, based out of the University of Colorado Boulder. For the past 15 years, Jeff has worked closely with water managers and other resource decision-makers in the Rocky Mountain West to help them understand and prepare for climate-related vulnerabilities by interpreting and applying paleoclimate data, historical climate records, and climate projections. He was lead author of the 2014 Climate Change in Colorado report for the Colorado Water Conservation Board, which summarized the latest science on climate trends and projections for the state. Jeff was initially trained in forest ecology (M.S., Forestry, University of Montana) and conducted fire history research, later shifting into applied climatology and hydrology.
Daniel McEvoy

Daniel McEvoy  •  daniel.mcevoy@dri.edu  •  775-673-7682

Daniel is a researcher with the Western Regional Climate Center. His research interests are interdisciplinary and span the fields of climate, hydrology, and meteorology. They include advancing drought monitoring technology, seasonal drought prediction, the role of evaporative demand on drought, quality and uncertainty assessment of weather observations, and climate modeling.
Charles Morton

Charles Morton  •  charles.morton@dri.edu  •  775-673-7620

Charles Morton is an assistant research scientist at the Desert Research Institute in Reno, NV. His research interests include surface energy balance modeling, hydrology, remote sensing, and cloud computing. For the past 10 years he has worked on numerous projects estimating evapotranspiration in the western United States using satellite remote sensing.
Imtiaz Rangwala

Imtiaz Rangwala  •  imtiaz.rangwala@noaa.gov  •  303-497-6544

Imtiaz is a research scientist at CIRES at the University of Colorado Boulder and NOAA’s Physical Sciences Division. He is a climate scientist with training in assessing and diagnosing regional scale climate change. Using climate observations and models, he works to understand and quantify climate processes relevant to regional warming trends and hydrological processes changes. This specifically ties into understanding climate extremes and changes in water balance in the western U.S., including the Great Plains region, and the how these extremes affect ecosystem response. Other work includes developing approaches to addressing and incorporating future climate change uncertainty into decision making and climate adaptation.
Andrea Ray

Andrea Ray  •  andrea.ray@noaa.gov  •  303-497-6434

Andrea is a research scientist at NOAA's Physical Sciences Division in Boulder, CO. Trained in environment and society geography, she studies user needs for climate forecasts, projections, and knowledge in natural resource management, decision making and adaptation. She often serves as a technical expert on planning and policy teams, working to transition research results into applications. Currently, she is serving on the interagency Climate Projections Team for the USGCRP Climate Resilience Toolkit, and on the Steering Team to develop a NOAA Water Resources Monitor and Outlook to provide enhanced outlooks on the runoff season. She recently led a climate analysis for the U.S. Fish and Wildlife Service to inform their Wolverine Species Status Assessment: Future snow persistence in Rocky Mountain and Glacier National Parks.
Heather Yocum

Heather Yocum  •  heather.yocum@noaa.gov  •  303-497-3917

Heather is a Research Scientist at the University of Colorado and NOAA’s Cooperative Institute for Research in Environmental Sciences (CIRES) in Boulder, CO. An environmental anthropologist and political ecologist, Dr. Yocum studies how culture and social systems impact the way that humans understand and interact with the environment. Since earning her PhD in Anthropology from Michigan State University in 2013, she has researched the production and use of climate and weather information to support natural resource management and decision-making; changing patterns of land use and natural resource management in the face of climate change; and environmental markets and payments for ecosystem services.

Resources

Primary Background Material

Related Links

References

EDDI Development

  • M. Hobbins, A. Wood, D. McEvoy, J. Huntington, C. Morton, M. Anderson, and C. Hain (June 2016): The Evaporative Demand Drought Index: Part I – Linking Drought Evolution to Variations in Evaporative Demand. J. Hydrometeor., 17(6),1745-1761, doi:10.1175/JHM-D-15-0121.1.
  • D. J. McEvoy, J. L. Huntington, M. T. Hobbins, A. Wood, C. Morton, M. Anderson, and C. Hain (June 2016): The Evaporative Demand Drought Index: Part II – CONUS-wide Assessment Against Common Drought Indicators. J. Hydrometeor., 17(6), 1763-1779, doi:10.1175/JHM-D-15-0122.1.

EDDI-Related

  • Dewes, C. F., I. Rangwala, J. J. Barsugli, M. T. Hobbins, and S. Kumar (March 2017): Drought risk assessment under climate change is sensitive to methodological choices for the estimation of evaporative demand. PLoS ONE, 12(3), e0174045, doi:10.1371/journal.pone.0174045
  • McNeeley, S. M., C. F. Dewes, C. J. Stiles, T. A. Beeton, I. Rangwala, M. T. Hobbins, and C. L. Knutson CL (2017): Anatomy of an interrupted irrigation season: Micro-drought at the Wind River Indian Reservation. Clim. Risk Mgt., doi:10.1016/j.crm.2017.09.004.
  • Rondeau, R. J., K. L. Decker, and G. A. Doyle (January 2018): Potential consequences of repeated severe drought for shortgrass steppe species. Rangeland Ecol. Mgt., 71(1), 91–97, doi:10.1016/j.rama.2017.07.002.
  • Shrum, T., W. Travis, T. Williams, and E. Lih (Online February 2018): Managing Climate Risks on the Ranch with Limited Drought Information. Clim. Risk Mgt., doi:10.1016/j.crm.2018.01.002.