COMPUTATION--Watershed Simulation Models

In Reply Refer To:                                          October 31, 1983
EGS-Mail Stop 415


Subject: COMPUTATION--Watershed Simulation Models

The Surface Water Branch, with Alan Lumb as coordinator, is taking a more
active role in the support and development of watershed simulation techniques.
Attention is being given to distributing the software, providing assistance
in the use of the software, setting goals for future software, and providing
test cases for quality control of software.  With the Distributed Information
System (DIS) we have become keenly aware of the difficulty of these tasks.
Your support and cooperation is vital.  Each of these tasks are described so
that you may understand how we plan to provide assistance and how you might
assist our efforts.

Currently, four software packages are being distributed on tape by the Surface
Water Branch.  They are DR3M, HSPF, PRMS, and ANNIE.  DR3M and DR3MQUAL are
the Distributed Routing Rainfall-Runoff Model--Version II and the Multi-Event
Urban Runoff Quality Model which are described in Water-Resources Investigations
Report WRI 82-344 and WRI 82-764, respectively, that were developed for
small urban watersheds.  The Hydrologic Simulation Program-Fortran (HSPF) is
a more comprehensive watershed model especially useful for large drainage
basins and water-quality simulations.  HSPF is being supported in cooperation
with the U.S. Environmental Protection Agency.  The Precipitation Runoff
Model System (PRMS) is described in a WRI report that has just been submitted
for Director's approval.  PRMS is especially useful for smaller watersheds
to analyze impacts of land use.  ANNIE is not a watershed model, but an inter-
active program (that is user friendly and programer friendly) for managing
data associated with the use of the watershed models.  ANNIE can plot data,
reformat data, create a time-series data base to be used by the models, and
create, check, and update input to the models.  ANNIE is modular, with a set
of utility subroutines to make it easier to add new features.  ANNIE is a
relatively new product although it has been in test stages for almost a
year.  New, upward compatible features are continually added to ANNIE and
will continue to be added over the next several years.  A draft users manual
and programers manual for ANNIE is available from the Surface Water Branch.

Assistance with watershed simulation models will be provided by the Regional
Surface Water Specialists, the Surface Water Branch, and research groups.
Questions on what model to use for a given application and problems with
specific applications should be referred first to the Regional Surface Water
Specialists.  Problems with the software should be referred to the Surface
Water Branch.  If there is a problem with the hydrologic process algorithms,
the problem will be referred to the Branch and then to the appropriate research
group for resolution.  The research groups particularly need to be informed
when model algorithms do not appear to represent the observed process.  Until
the plans and details of the DIS supported library are imlemented, the Branch

will continue to distribute the watershed modeling software.  Thirteen Districts
and three Regions have requested and received tapes of the software.  Since
the software and test runs involve almost 200,000 records, the software is
being distributed by tape.  Updates also can be received by tape.  A file of
tape distributions is maintained by the Branch so that it is known who has
what version.  For this reason all updates should be received from the Branch
and not another District.  If updates only involve a few lines of codes,
those updates may be transmitted by phone or memorandum.  Additional PRIMOS
command files are on the tape so that code can be easily edited, compiled,
and segmented at the site.  When the DIS network is implemened, the policy
on updates for watershed simulation models will be revised.

Test runs are developed for two of the software packages and are being developed
for the other two packages.  Continued improvement to the test runs will be
made so that each option in each model is tested.  It is expected that each of
these tests would be performed upon receiving a tape or updates to the models.

Goals and designs for future software are being addressed by the watershed
simulation models group composed of Research and Surface Water Branch staff.
This group will be expanded to include representation from the Districts.
Our goals are to provide a single modular software package that uses a common
data base and standard Fortran.  Meeting these goals should ease the task of
software to other computer systems, and make enhancements more readily avail-
able.  There is both a long-term and a short-term goal.  The short-term
goal will provide an intermediate integration of the four models so output
from a process in one model could be used as input to a process in another
model (see attached Figure).  This will be accomplished with ANNIE and a
common time-series data-management system.

Current software is quite adequate for most watershed simulation needs such
as record extension, evaluation of the impacts of urban development or surface
mining, and forecasting reservoir inflows.  Water-quality algorithms include
sediment, nutrient cycles, dissolved oxygen, pH, conservatives, pesticides,
and temperature but have not been tested as thoroughly as the water-quantity
algorithms.  Watersheds from 1 acre to 62,000 square miles have been used.

Much remains to be added to the software packages, especially preprocessing
and postprocessing of time-series data by ANNIE.  This is where assistance
is needed.  Attached sheets summarize the status of the software plans.
Please let us know of any additions you may have for this list and of any
District staff that would like to work on any of the tasks outlined in the
software plans.  A 2-day workshop in January 1984 will be given to explain
the Fortran coding conventions and use of the utility subroutines in ANNIE.

                                           Marshall E. Moss
                                           Chief, Surface Water Branch


Distribution:  A, B, FO

                                SOFTWARE TASKS

Task                                 Assignment                Date

PRMS test runs                       Branch                    December 1983

Utility subroutine for ANNIE         Contractor                October 28, 1983
  to more easily read/write
  from/to TSS file

Output simulated flows               Kate Flynn                November 1983
  from PRMS to file of
  PLTGEN format

Add J407 to ANNIE to                 Kate Flynn                November 1983
  read PLTGEN files for
  flood-frequency analysis

Add Tektronix, Houston               Branch                    December 1983
  Instrument, Zeta, and Tab
  Graphics plotting to ANNIE

Add selected statistical             Unassigned                January 1984
  analysis of time-series
  to ANNIE using recently
  acquired proprietary software

Add existing subroutines to          Unassigned                December 1984
  ANNIE to fill in missing
  precipitation record

Add error analysis and               Unassigned                February 1984
  sensitivity subroutines
  to ANNIE

Add to ANNIE subroutines to move     Branch                    December 1983
  data directly from the interim
  WATSTORE files to the TSS file

Test Fortran version of A634         Georgia District          January 1984

Add the Fortran version of A634      Unassigned                January 1984
  to the group of models, ANNIE
  and TSS file

Reformat simulated data from the     Unassigned                Summer, 1984
  TSS file to files for unsteady-
  flow models

Add subroutine to ANNIE to read      Branch                    November 1984
  new NOAA tapes with "element"


Task                                 Assignment                Date

Include DR3M in ANNIE and            Kate Flynn                Janaury 1984
  the TSS file

Include PRMS in ANNIE                Linda Saindon             January 1984

Utility subroutine for ANNIE for     Contractor                December 16, 1983
  multi-response to question

Add subroutines to ANNIE to          Unassigned                March 1984
  compute potential ET and solar
  radiation from other meteoro-
  logic time series

Provide interface to GIS soft-       Unassigned                Fall, 1984
  ware to determine watershed

Add software to utilize radar        Unassigned                1985
  to improve rainfall input
  to watershed models

Add utility routine to ANNIE         Unassigned                April 1984
  for low-flow frequency
  analysis of simulated flow

Add utility routines to ANNIE        Unassigned                February 1984
  to plot results of flow-
  duration analysis

Add utility routines to ANNIE        Unassigned                February 1984
  to plot observed vs. simulated
  peak flows, storm runoff
  volumes, annual flows, and
  monthly flows