Current Website: http:www.hydro.washington.eduforecastmonitor

1
NASA/NOAA Workshop Predicting Drought on
Seasonal to Decadal Timescales (Silver Spring,
MD), May 2005
An Experimental Surface Water Monitoring System
for Continental US Andy W. Wood, Ali S. Akanda,
and Dennis P. Lettenmaier
Univ. of Washington
2
1
Methods
Index Station Method Gridded Forcing Creation

• OVERVIEW
• We have implemented an experimental real-time
  surface water monitoring system that uses the
  Variable Infiltration Capacity (VIC) macroscale
  hydrology model (Liang et al., 1994) for
  real-time estimation of soil moisture and snow
  water equivalent (SWE) over the continental U.S.
  at 1/2 degree spatial resolution.
• The monitoring system will stage the DROUGHT
  ANALYSIS PRODUCTS described in the accompanying
  poster (Andreadis et al.) in real-time, as well
  as maintain an archive of comparative products
  extending back to 1915.
• The monitoring system is driven by daily
  precipitation temperature minima and maxima
  from 2131 Coop stations, and has a 1-day update
  lag, with updates each day by 2 p.m. PST.
• The system will also stage weekly outlooks based
  on similar methods to the ensemble forecasting
  techniques used in the UW west-wide seasonal
  hydrologic forecast system.

Index Station Forcing Approach
Real-time Nowcasting Information Flow
VIC Hydrologic Model (Liang et al., 1994)
NOAA ACIS Prcp Tmax Tmin Coop Stations
A primary goal of this system was to ensure
consistency between real-time and retrospective
outputs, hence the input stations were limited to
those with BOTH reliable real-time reporting and
historical records extending back at least 40
years. Because this results in many fewer
stations than are available with less stringent
criteria, we use the input stations indirectly.
First, we estimate precipitation percentile (for
month long periods) and (daily) temperature
anomalies and interpolate them to a 1/8 degree
grid. Then we extract corresponding values from
a 1/8 degree climatological forcing PDF created
using a larger set of station inputs (e.g., not
all reporting in real-time, taken from the
Andreadis et al. effort described in an
associated poster). Finally, temporal
disaggregation of the precipitation period values
is needed. Currently, the PDF is based on
1960-2003, and the stations are fairly consistent
for this period however, this choice may be
revisited.
NOAA ACIS Prcp Tmax Tmin Coop Stations
1930s
1955
Nowcast produced with 1-2 day lag from current
Index Meth.
VIC Retrospective Simulation Daily, 1915 to Near
Current
VIC Real-time Simulation (1 month long)
Hydrologic State
Hydrologic State (-1 Day)
Hydrologic values, anomalies, percentiles w.r.t.
retrospective PDF
vals, anoms -iles w.r.t. PDF
climatology (PDF) of hydrologic values w.r.t.
defined period
3
5
Surface Water Monitoring System Products
Comparison with Other Datasets
CPC
UW
Examples of Current Products
From a qualitative standpoint, we find good
agreement with current real-time CPC soil
moisture monitoring tool (recent plots shown at
right)
ARCHIVES
CHANGES IN CURRENT CONDITIONS
March 2002 Virginia experiences severe drought,
many well failures
March 1997 La Nina conditions bring the
highest recorded snowfall to the PNW
July 2002 the western U.S. drought centers on
Colorado
August 1993 the highest recorded flow on the
Mississippi R.
The change in conditions over the past week, 2
weeks and 1 month help to characterize the
evolving water balance.
Colorado Cell (39.25, -108.25) Andreadis et
al. SW Monitor
Andreadis et al.
The Andreadis et al. retrospective soil moisture
dataset shows good spatial and temporal
consistency with the SW Monitor
simulations (temporal comparison,
left) (spatial comparison, right)
In Development
TIMESERIES ANAYSIS FOR PIXELS OR REGIONS
For a pixel or region (e.g., Upper Colorado R.
basin), one can track the current hydro-climatic
conditions and contrast them with other water
years or with the seasonal climatological PDF.
4
Current Website http//www.hydro.washington.edu
/forecast/monitor/
The first-cut implementation of the system has
a limited set of products the primary results
shown currently are nowcasts (with a 1-2 day lag
depending on time of day) of soil moisture and
snow water equivalent percentiles. Other
products are in development (see Section 5 at
right).
OUTLOOKS
The recent trend in relative soil moisture and
SWE is shown by the changes in their percentiles
in the last week, 2 weeks and month.
We plan to implement weekly outlooks based on
similar methods to those used in the UW west-wide
seasonal hydrologic forecast system (using CPC,
CFS, NSIPP and/or ESP climate ensembles). At
right are shown a sample ESP 3-month outlook
initialized February 1
An archive from 1915-present of soil moisture and
SWE percentiles on Day 1 of each month is
available, with a simple interface for navigation
References / Acknowledgements
Other Changes / Ongoing Work
Andreadis, K.M., E.A. Clark, A.W. Wood, A.F.
Hamlet, and D.P. Lettenmaier, 2004, 20th Century
Drought in the Conterminous United States ,
Journal of Hydrometeorology (accepted). Liang,
X., D. P. Lettenmaier, E. F. Wood and S. J.
Burges, 1994. A Simple hydrologically Based
Model of Land Surface Water and Energy Fluxes for
GSMs, J. Geophys. Res., 99(D7). The authors
acknowledge the support of NOAA/OGP and the NASA
Seasonal-to-Interannual Prediction Project
(NSIPP).

• improving spatial resolution of the overall
  monitoring system to 1/8 degree.
• completing the automation of the nowcast /
  real-time simulation / plot generations
• incorporating additional products (beyond
  percentiles), including runoff maps, cumulative
  departures from normal, and recovery
  probabilities (derived from the outlook products)
• explore different PDF periods and station
  datasets. (It may be better to screen out more
  stations to achieve greater consistency with the
  pre-1950 record).