Effectiveness of Existing Drought Indicators

1
Effectiveness of Existing Drought Indicators

• Southeast-ACF Drought Early Warning Information
  System Development Workshop Apalachicola River
  Bay
• April 27-28 2010
• Apalachicola, Florida
• Douglas Le Comte
• NOAA/NCEP Climate Prediction Center
• Douglas.lecomte_at_noaa.gov

2
Outline

• A look at selected current drought indicators
• An honest appraisal (probably politically
  incorrect) of how well they work Dougs Top 5!

3
Atlanta Journal Cartoon
Lake Lanier drought indexthe keep-it-simple
approach
An effective drought index is one that does what
you want it to do. You may want to merge several
drought indicators to meet your needs, e.g., CPC
delivery of historical river basin indices to the
water commissions of GA, FL, AL in January 2001.
4
Desired Features of a Drought Index

• Understandable (what it means and how its
  calculated
• Available daily (short term index) or weekly
  (long-term index)
• Correlates with impacts
• Range of values easy to understand (e.g.,
  percentiles, anomalies).
• Appropriate spatial resolution
• Can be imported into GIS (e.g. GeoTIFF)
• Has a long history and the archive is easy to
  find

5
Historical Perspective

• The Drought Phenomenon
• Definition Sectors Impacted Indices
• Early 20th Century Drought Indices
• Incorporated some measure of precipitation over
  a given time
• Examples
• 15 Consecutive Days with No Rain
• 21 Days or More with Precipitation less than
  One-Third (or 30) of Normal
• Annual Precipitation less than 75 of Normal
• Monthly Precipitation less than 60 of Normal
• Any Amount of Rainfall less than 85 of Normal

• Other important climatic elements (soil moisture,
  evapotranspiration, etc.) were not readily
  available at that time
• R. Heim (Inter-Regional Workshop on Indices and
• Early Warning Systems for Drought
  WMO/NDMC/NOAA/UNCCD/USDA
• Lincoln, NE, USA December 8-11, 2009)

6
These Basic Indicators Are Still Widely Used Today
7
Historical Perspective

• Palmers Drought Index (1965) and Crop Moisture
  Index (1968)
• CMI is computed on a weekly basis for
  agricultural applications
• Moisture Stress Index
• Developed in 2003 by NCDC
• Annual index for corn and soybeans that relates
  moisture stress (drought and extreme wetness) to
  crop productivity

• Computed from Palmer Z Index and crop yield
• Economic impacts

8
Post-Palmer Era

• Indices Based On
• Additional Observed Variables
• Advanced Statistical Methodologies
• More Sophisticated Models
• Data from New Observational Platforms

9
Standardized Precipitation Index (SPI)

• Developed by Colorados McKee et al. (1993)
• Based on Probability Distribution of
  Precipitation, with Normalization
• Can Compute for Different Time Scales for
  Differing Applications

http//www.ncdc.noaa.gov/oa/climate/research/preli
m/drought/spi.html http//www.wrcc.dri.edu/spi/spi
.html
http//www.agr.gc.ca/pfra/drought/mapscc_e.htm
10
Soil Moisture

• Soil Moisture Observation Networks Limited in U.S.

http//www.wcc.nrcs.usda.gov/scan/
http//www.cpc.ncep.noaa.gov/products/monitoring_
and_data/topsoil.shtml
11
Soil Moisture

• Rely on Modeled Soil Moisture

CPC Leaky Bucket
Leaky Bucket Model
http//www.cpc.noaa.gov/soilmst/leaky_glb.htm http
//www.cpc.ncep.noaa.gov/products/Soilmst_Monitori
ng/US/Soilmst/Soilmst.shtml http//mrcc.sws.uiuc.e
du/cliwatch/drought/moisture.htm http//www.agr.gc
.ca/pfra/drought/nlslmr_e.htm
12
Using the NLDAS Ensemble Mean Soil Moisture as a
Drought Indicator
Shallow layers of soil moisture can be used as
agricultural drought indicators. Model runoff as
hydro indicators.
http//www.emc.ncep.noaa.gov/mmb/nldas/drought/
Ensemble Mean
NCEP Noah
NASA Mosaic
Princeton VIC
OHD SAC
Percentile soil moisture levels correspond to
U.S. Drought Monitor levels D0 to D4
13
NOAA/NWS Advanced Hydrologic Prediction Service

• Hybrid Product Integrates Radar Precipitation
  Estimates with In Situ Rain Gage Measurements

http//water.weather.gov/index.php
14
U.S. Objective Blends

• Developed as support for U.S. Drought Monitor
• Operationally integrate multiple indicators in a
  weekly update using a percentile ranking method
• Produced weekly using CPCs real-time daily and
  weekly climate division data and NCDCs monthly
  archive of indices for 1932-2000
• All parameters are first rendered as percentiles
  with respect to 1932-2000 data using a percent
  rank method
• The parameters (drought indicators) are then
  combined using a weighting scheme
• Short- and Long-Term Blends produced

15
http//www.cpc.ncep.noaa.gov/products/predictions/
tools/edb/droughtblend-access-page.html
16
Drought Assessments based on Aggregating
Indicators
DWR NC
17
Hydrologic Indices

• Based on Streamflow or Groundwater
• Groundwater Observing Network Coarse

http//watermonitor.gov/
18
Hydrologic Indices

• Based on Streamflow or Groundwater
• Streamflow Need to Remove Peak Flow from
  Hydrograph
• Streamflow-based drought indicators usually based
  on low-flow (base flow) or some averaged flow

Flood Monitoring
Drought Monitoring
Monthly Average Streamflow
14-Day Average Streamflow
http//watermonitor.gov/
19
Hydrologic Indices

• Based on Streamflow or Groundwater
• Reservoir Data
• State data, no national access system
• Managed for different purposes (irrigation, urban
  water supply, flood control, etc.)

http//watermonitor.gov/
Arizona reservoir levels, September 2009, percent
of capacity.
20
Remotely-Sensed (Satellite-based) Indices

• Provide Global Coverage
• Spatially Consistent
• Reflected Intensities at Specific Wavelengths

http//www.agr.gc.ca/pfra/drought/mapscc_e.htm
http//www.star.nesdis.noaa.gov/smcd/emb/vci/VH/in
dex.php
21
Vegetation Drought Response Index (VegDRI)

• Hybrid Drought Index that Integrates
• Satellite-based observations of vegetation
  conditions
• Climate-based drought index data
• Biophysical characteristics of the environment

http//drought.unl.edu/vegdri/VegDRI_Main.htm
(Source Wardlow, 2008)
22
My Top 5 Countdown
23
No. 5 KBDI
http//flame.fl-dof.com/fire_weather/KBDI/4km_main
.html
Timely Daily Good short term dryness indicator
April 23, 2010 Please Note Due to technical
difficulties the KBDI and Rainfall maps are
inaccurate. We apologize for any inconvenience.
24
No. 4 CPC Drought Indicator Blends
http//www.cpc.ncep.noaa.gov/products/predictions/
tools/edb/droughtblend-access-page.html
Weekly, based on Climate Division data weights
somewhat subjective not operational
25
No. 3 Soil Moisture Model (CPC)
http//www.cpc.ncep.noaa.gov/products/Soilmst_Moni
toring/US/Soilmst/Soilmst.shtml
Daily, with 1-day lag, based on CD data 1.6m
depth good mid-range indicator. Archive
available. But, sometimes not realistic (e.g.,
see extreme drought in upstate NY).
26
No.2 NLDAS Soil Moisture
Promising! And yet min 5-day lag, sometimes much
longer non-operational very resource intensive,
so computer crashes result in huge delays.
http//www.emc.ncep.noaa.gov/mmb/nldas/drought/
27
No. 1 USGS Streamflow
http//waterwatch.usgs.gov/
So much to like! Current, real time, daily to
28-days, reliable, percentile format, zoomable,
28
A Variety of Format Options
Can show in Google Earth as well as Google Maps
29
Questions?
Douglas.lecomte_at_noaa.gov