

Hydrologic Engineering Center



Hydrologic Modeling System

HEC-HMS

Release Notes
Version 4.3
September 2018

Approved for Public Release – Distribution Unlimited

 Hydrologic Modeling System HEC-HMS, Release Notes
2018. This Hydrologic Engineering Center (HEC) Manual is a U.S.
Government document and is not subject to copyright. It may be copied
and used free of charge. Please acknowledge the U.S. Army Corps of
Engineers Hydrologic Engineering Center as the author of this Manual in
any subsequent use of this work or excerpts.
Use of the software described by this Manual is controlled by certain terms
and conditions. The user must acknowledge and agree to be bound by
the terms and conditions of usage before the software can be installed or
used. For reference, a copy of the terms and conditions of usage are
included in the HEC-HMS User's Manual, Appendix E so that they may be
examined before obtaining the software.
This document contains references to product names that are used as
trademarks by, or are federally registered trademarks of, their respective
owners. Use of specific product names does not imply official or unofficial
endorsement. Product names are used solely for the purpose of
identifying products available in the public market place.
Microsoft and Windows are registered trademarks of Microsoft Corp.
Linux is a registered trademark of Linus Torvalds.

ii
 HEC-HMS Version 4.3 Release Notes


Introduction
Version 4.2 of the Hydrologic Modeling System (HEC-HMS) was
completed in August 2016 and released for general use shortly
thereafter. An interim release called Version 4.2.1 was completed
in March 2017 primarily to correct a number of errors.
A number of new capabilities have been under development since
2016. The most significant is the redesign of the optimization trial
to support maximization and minimization problems. A new search
method has also been added to the trial; Markov Chain Monte
Carlo. This new search method provides probabilistic parameter
estimates. There is a new reach routing method, enhancements to
another, plus a new initial condition option for all routing methods.
The Clark and ModClark transform methods offer new capabilities
and it is now easier to work with gridded methods. There is a new
shortwave radiation method in the meteorologic model and the
Penman Monteith evapotranspiration method has been enhanced.
Users have reported some difficulties with several features. The
development team has also continued careful and systematic
testing of the program. The results of that testing in combination
with reports from users have allowed the identification and repair of
various problems. Some of these problems affected simulation
results and are described in detail later in this document.
The combination of new capabilities and error corrections is this
Version 4.3 release. It is currently available for the Microsoft
Windows® operating system and the Linux® operating system.
The program has been designed to be easy to use. However, an
understanding of how the program works is critical to producing
accurate results. It is suggested that a new user read the HEC-
HMS Quick Start Guide (Version 4.3 January 2018). The guide
briefly describes the basic features of the program in enough detail
to begin using it. The guide also includes a step-by-step tutorial.

Installation
The installation package and all documentation are available on the
HEC website at http://www.hec.usace.army.mil/. This new release
is installed independently of any previous versions of the program
so you will need to uninstall the previous versions if you do not wish
to use them anymore. However, you may keep different versions of
the program installed for parallel use or testing if you choose. This
new version will automatically open projects developed with any
previous version of the program. However, once a project has

1
HEC-HMS Version 4.3 Release Notes


been opened in Version 4.3, it will not be possible to open it with

older versions of the program.

Windows
The new installation package is designed to be easy to use. It will
take you through the steps of selecting a directory for the program
files and making other settings. Use the following steps to install
the program on the Microsoft Windows® operating system:
1. Download the installation package from the HEC website to a
temporary folder on the computer.
2. Run the installation program. In Windows Explorer, double-click
the icon for the installation program. You must have
administrator privileges to run the installer.
3. Follow the on-screen prompts to install the program.

Linux
The installation package is delivered as a compressed TAR file. It
has several dependencies that must be installed before extracting
the installation package. The following commands can be used to
configure the dependencies prior to extraction of the installation
package.
Red Hat Enterprise Linux 7
yum install libldb.i686 libXext.i686
libXrender.i686 libXtst.i686 libgcc.i686
libstdc++.i686

Ubuntu 14.04
sudo apt-get install libc6-i386 libstdc++6:i386
libxtst6:i386 libxrender1:i386 libgcc1:i386
libxi6:i386

New Capabilities
A total of eighteen notable new capabilities have been added to the
program. Summary descriptions are given below.

Clark and Modclark Transform Parameters

The subbasin element includes the Clark and Modclark transform
methods, among others. Previous versions erased the parameter
values when switching from Clark to Modclark, or from Modclark to

2
 HEC-HMS Version 4.3 Release Notes

Clark methods. The parameters have exactly the same meaning in
both methods. The parameter values are now automatically
preserved when switching between these methods.

Clark Transform Variable Parameters

Unit hydrograph theory assumes a linear relationship between
precipitation and the runoff response. This assumption can lead to
errors in timing and peak magnitude when simulating events that
result from extremely large excess precipitation rates, such as the
Probable Maximum Precipitation. For the Clark transform method,
users now have the option of inputting tables relating time of
concentration and storage coefficient to the excess precipitation.
This allows for the user to recreate a non-linear runoff response
when excess precipitation rates vary. These tables are entered in
the paired data manager and then used in the Clark transform.

Lag and K Routing

The Lag & K method is a hydrologic storage routing method based
on a graphical routing technique that is extensively used by the
National Weather Service (NWS) in their forecast modeling tools.
The method is a special case of the Muskingum method where
channel storage is represented by the prism component alone with
no wedge storage, i.e. Muskingum X = 0. The lack of wedge
storage means that the method should only be used for slowly
varying flood waves. Like all hydrologic routing methods, it does
not account for complex flow conditions such as backwater effects
and hydraulic structures. The Lag term refers to the travel time for
a flood wave to move downstream. The K term accounts for
attenuation. Lag and K parameters can be entered as constants or
variable. For variable Lag and K, Lag is entered as an Inflow-Lag
paired data function and K is entered as an Outflow-Attenuation
paired data function. Evaluation of historical flood hydrographs
provides the basis for Lag and K relationships within a reach.

Muskingum Cunge Tabular Cross Section

The Muskingum Cunge channel routing method has always
included several conceptual cross section shapes: circle, trapezoid,
triangle, and rectangle. There has also been an option for a simple
8-point cross section shape. The new tabular shape has been
added so that complex cross section shapes can be represented as
hydraulic property tables. The underlying equations require several
properties as a function of flow depth: discharge, flow area, and top
width. The conceptual shapes and 8-point shape compute these
properties using basic hydraulic formulas or simple geometry
formulas. Conversely, the new tabular shape takes these

3
HEC-HMS Version 4.3 Release Notes


properties as hydraulic tables specified by the user. The tables can

be entered directly in the paired data manager as elevation-
discharge, elevation-area, and elevation-width tables. There is also
a special editor available from the Tools menu for entering all the
tables simultaneously. From a practical perspective, these tables
will usually be externally computed with the aid of a hydraulic
model.

Normal Depth Routing

A new channel routing method called Normal Depth uses the
modified Puls method for performing a routing, but presents the
user with parameters for normal depth calculations. The modified
Puls channel routing method uses a curve of storage versus
outflow plus a number of subreaches to perform routing
calculations. The normal depth routing approach calculates storage
versus outflow using Manning's equation for normal depth. This
assumes a uniform normal depth throughout the channel. The
normal depth approach calculates the number of subreaches by
dividing the travel time by the simulation time interval.

Reach Routing Initial Conditions

The reach element in the basin model includes a number of
different hydrologic routing methods. All of these methods have
used the same initial conditions. That is, the total inflow to the
reach at the simulation start time is used as the initial conditions
throughout the entire reach length. All of the routing methods have
been enhanced to offer two choices for the initial conditions. It is
still possible to use the total inflow at the start time. It is now also
possible to specify the outflow from the reach at the start time. For
this second choice, the initial conditions are set so that the outflow
from the reach matches the specified value with a smooth transition
to the total inflow at the upstream end of the reach.

Automatic Initial Reach Discharge in Forecasts

The forecast alternative already includes special tools for
automatically computing subbasin initial baseflow and reservoir
initial pool storage using observed gage data. A new tool is now
available to automatically compute reach initial outflow using
observed gage data. The new tool is specially designed to work in
conjunction with the new outflow initial condition for all routing
methods. An observed flow gage can be selected for each reach
element. Optional properties can be specified to screen the
observed data for validity before using it during a forecast. The
observed data is processed and automatically used as the initial

4
 HEC-HMS Version 4.3 Release Notes

conditions in the reach. An override option can be used at the
discretion of the user if automatic processing does not produce
good results.

Hargreaves Shortwave Radiation

A new shortwave radiation method has been added to the
meteorologic model. The method uses only maximum and
minimum temperature data along with latitude and longitude to
compute shortwave radiation according to the methodology of
Hargreaves and Samani, 1985, Eq. 2. The adjustment coefficient,
kRs, is set to a default value of 0.17 and can be adjusted by the
user. The new method can be used with gridded and non-gridded
subbasins.

Hargreaves Evapotranspiration
A new evapotranspiration method has been added to the
meteorologic model. The method uses average daily temperature
data combined with solar radiation to compute evapotranspiration
according to the methodology of Hargreaves and Samani, 1985,
Eq. 1. The coefficient is set to a default value of 0.0135 and can be
adjusted by the user. When the Hargreaves evapotranspiration
method is combined with the Hargreaves radiation method the
evapotranspiration calculation takes the form of Hargreaves and
Samani, 1985, Eq. 8. The user can control the coefficient through
the Hargreaves shortwave and evapotranspiration methods. The
new method can be used with gridded and non-gridded subbasins.

Hamon Evapotranspiration
A new evapotranspiration method has been added to the
meteorologic model. The method uses average daily temperature
data combined with latitude and longitude to compute
evapotranspiration according to the methodology of Hamon, 1963,
Eq. 2. The Hamon coefficient is set to a default value of 0.0065
and can be adjusted by the user. If the simulation time interval is
sub-daily the daily evapotranspiration amount is distributed
between sunrise and sunset according to a sinusoidal pattern. The
new method can be used with gridded and non-gridded subbasins.

Optimization Trial
The optimization trial is a flexible tool for automatically estimating
model parameters when observed flow is available. An objective
function is used to compute model performance by comparing the
simulated and observed flow. A search method is used to
automatically adjust model parameters in order to minimize the

5
HEC-HMS Version 4.3 Release Notes


objective function. Up until now, two search methods have been

available. The "univariate gradient" method was a modernization of
the original HEC-1 algorithm. The "Nelder Mead" method used a
simplex algorithm.
The univariate gradient method has been eliminated. The method
typically offered poor performance in estimating parameters. It
often showed very slow convergence and frequently failed to
identify the global minimum. It almost always underperformed
when compared to the Nelder Mead method.
The Nelder Mead method has been replaced by the "simplex"
method. The replacement is functionally the same since it is based
on the original work by researchers John Nelder and Roger Mead
in 1965. The simplex method uses an improved algorithm that is
more efficient. Old projects will be automatically upgraded to use
the replacement method.
A new "univariate" method has been added. This new method
works with only a single parameter. While the simplex method is
powerful, it requires a minimum of two parameters in order to
operate. The univariate method is now available for cases where
only one parameter needs to be estimated.
Both the simplex and univariate search methods can now be used
for minimization and maximization problems. In a minimization
problem, the model parameters are automatically adjusted in order
to find the smallest possible value of the objective function. The
root mean square error is a common choice for the objective
function in a minimization problem. A small value of the root mean
square usually means that the simulated and observed results are
close together over much of the simulation. All past versions of the
program only solved minimization problems. In a maximization
problem, the model parameters are automatically adjusted in order
to find the largest possible value of the objective function. The
Nash Sutcliffe coefficient or total runoff volume are common
choices for the objective function in a maximization problem.
Allowing for maximization problems is a new feature with this
version. One example application of maximization is using the
HMR52 precipitation method and allowing the optimization search
method to find storm parameters that result in the largest possible
runoff volume.

Markov Chain Monte Carlo in the Optimization Trial

A new search method called "Markov Chain Monte Carlo" has been
added to the optimization trial. This method results in a robust
assessment of parameter uncertainty as part of the optimization

6
 HEC-HMS Version 4.3 Release Notes

and effectively converts the deterministic modeling components into
a probabilistic flow simulation system. The search process begins
with a Latin Hypercube Sample (LHS) which generates an initial
number of parameter sets that cover the entire parameter space
box as defined by user-specified lower and upper bounds for each
adjustable model parameter. The LHS initialization serves as the
basis for a population of Markov Chain directed random walks
which run in parallel and characterize the flow simulation. A data
likelihood function is associated with each evaluated parameter set.
Currently, the squared deviation likelihood is the only option
available. The definition for this classical likelihood includes the
simple root mean square error (RMSE) function from traditional
optimization. During the Markov Chain directed random walks,
proposed jumps are evaluated for acceptance or rejection using the
Metropolis acceptance probability rule. Given that a quantitative
convergence diagnostic can prematurely assess equilibrium, we
recommend assessing equilibrium using a weight-of-evidence
approach that involves user interaction. To support this approach,
plots are provided for (1) the likelihood function (RMSE), (2) the
quantitative convergence diagnostic (Gelman and Rubin) for each
parameter, and (3) traces of all chains for each parameter. When
all of these point to convergence, then most likely a thinned history
of subsequent draws will be from the equilibrium target distribution
and can be used for inference. The post burn-in monitoring period
draws empirically estimate the target posterior distribution and
provide the basis to estimate the probability of simulated output for
model design alternatives. For example, a robust estimate of the
peak flow with only a 1% chance of being exceeded.

Depth Area Analysis

The depth area analysis was designed to make it easy to compute
flow-frequency results throughout a large basin model. The
analysis included the selection of an existing simulation run where
the included meteorologic model used the frequency precipitation
method. Then the analysis added one or more points where results
would be computed, automatically linking the drainage area at the
selected points to the storm area in the frequency storm.
The depth area analysis no longer includes the selection of a
simulation run. Instead, a basin model and meteorologic model are
selected directly. A start time, end time, and interval are also
chosen. Then points are added for computing results matched to
the drainage area. It is still required that the meteorologic model
use the frequency precipitation method. Management of
components and results is now much easier and more consistent

7
HEC-HMS Version 4.3 Release Notes


with other simulations such as optimization trials and forecast

alternatives.

Uncertainty Analysis Specified Sampling

The uncertainty analysis has included three different methods for
sampling a selected parameter in the basin model: simple
distribution, monthly distribution, and regression with additive error.
A new method called "specified sampling" is now available. This
new method uses a parameter table created in the paired data
manager. The table is configured for a specific parameter type
within the basin model, and a sequence of parameter values
manually entered. The table can then be selected for specified
sampling in the uncertainty analysis. The sampling process will
sequentially use the values in the table, repeating as many times as
necessary to complete the specified total number of samples.
Specified sampling is designed to be used in two different ways.
First, it can be used when a parameter can take on a limited
number of different values with equal probability. Dam breach
parameters are a good example. The scientific literature provides a
small number of equations for calculating dam breach parameters.
Each of the equations produces a slightly different parameter
estimate. There is no consensus on which equation is "best" so
they may all be treated as equally likely. The user could use all of
the equations external to HEC-HMS and then prepare a table for
specified sampling. Second, it can be used with results from an
optimization trial using the Markov Chain Monte Carlo search
method. The results from such a trial typically include hundreds of
equally likely parameter sets. The trial needs to be configured first
and successfully computed. Then a special command on the Tools
menu can be used to automatically create parameter tables in the
paired data manager. The tables are created in a way which
maintains the correlation inherent in the trial results. All of the
parameters analyzed in the trial should be included in the
uncertainty analysis with specified sampling.

Observed Stage and Pool Elevation

All elements in the basin model include the option for observed flow
and observed stage. Additionally, reservoir elements include the
option for observed pool elevation. Element result summary tables
have included a special section when observed flow is used. This
special section provides summary information about the observed
flow during the simulation and provides certain comparison
statistics relative to the computed flow. Similar special sections are

8
 HEC-HMS Version 4.3 Release Notes

now included in results summary table when observed stage or
observed pool elevation are selected.

Forecast Alternative Results

The forecast alternative stores the same time-series results for all
of the basin model elements as would be found from a simulation
run. When the special forecast option for blending is used,
additional time-series results specific to forecasts are computed
and available for viewing. Activating the blending option at an
element replaces computed outflow with observed outflow from the
start time up to the forecast time. The observed outflow then
transitions to the computed flow from the forecast time to the end
time. Previously, an element using blending provided a result
graph that showed the combination of observed flow (before the
forecast time) and computed flow (after the forecast time). The
result graph has been enhanced to show both the computed and
observed outflow from the start time to the forecast time. The
outflow from the forecast time to the end time is the same.
The forecast alternative can be used in its own right to produce
stream flow forecasts. It is also specially designed to integrate with
the Corps Water Management System (CWMS) and the Real Time
Modeling System (HEC-RTS) to produce integrated forecasts in
conjunction with meteorologic, hydraulic, reservoir, and economic
software. A key feature in this integration is the use of data
variants. Data variants are used by CWMS/HEC-RTS to connect
results from other software to the forecast alternative, and to
connect results from the forecast alternative to other software.
Data variants can only be created and managed when using HEC-
HMS within CWMS/HEC-RTS. New enhancements now make it
possible to compute a forecast alternative configured with data
variants directly from the HEC-HMS interface. Additional
enhancements make it possible to view the primary forecast
alternative results as well as results produced using data variants.

Uncertainty Analysis Results

The uncertainty analysis does not store all time-series results at all
basin model elements. Instead, time-series results are only stored
when specifically selected at elements. Special results have been
available when choosing outflow at any type of element. Special
results have also been available when choosing pool elevation at a
reservoir element. It is now possible to choose additional time-
series results. Incremental precipitation can now be selected for
subbasin elements.

9
HEC-HMS Version 4.3 Release Notes


Selecting Elements in the Basin Map

The basin map provides an intuitive spatial display of the elements
in the basin model. An element can be selected by clicking on it in
the map. An element can also be selected by name from the Edit
menu using the "Select By Name" command. It is now possible to
select all of the elements upstream or downstream of an element in
the basin model. The element should first be selected by clicking
on it in the map. The "Select Upstream" command on the Edit
menu can be used to automatically select all elements upstream of
initial selection. The "Invert Selection" command on the Edit menu
can be used to unselect all of the currently selected elements, and
select all of the currently unselected elements in a single action.

Enhancement and Revisions

Occasionally it becomes necessary to revise how an existing
feature operates in order to create more capability or correct a
problem. Several changes of this nature have occurred since the
last software release. A complete description of each
enhancement or revision is given below. In general, revisions
result in changes in the computed results when compared to
previous program versions.

Subbasin, Reach, and Reservoir GIS Data

Previous program versions included the option to add geospatial
data to subbasin, reach, and reservoir elements. Subbasins and
reservoirs could be represented with polygons while reaches could
be represented with polylines. The spatial data was stored in three
Esri shapefiles with a separate file for each of the three element
types. A special naming convention was used to implicitly link the
Esri shapefiles to a basin model. The geospatial data was primarily
used for drawing in the basin map. An important exception was
that the geospatial data was required for subbasins when using the
HMR52 precipitation method.
The previous method of storing geospatial data has been
eliminated and replaced with a new method using the Spatial Lite
file format. Each basin model now has a Spatial Lite file to
accompany the basin model ASCII file. The Spatial Lite file is
automatically managed by the program and holds all of the
geospatial data for elements in the basin model.
Geospatial data can be added to the elements in the basin model
using a special georeferencing feature. The georeferencing feature
is capable of reading from Esri shapefiles and many other common

10
 HEC-HMS Version 4.3 Release Notes

file formats for storing polygons and polylines. In this way,
geospatial data can be added to existing elements in the basin
model. There is no upgrade of spatial data stored according to the
old conventions. Users should use the new georeferencing feature
to transfer spatial data from the Esri shapefiles to the new Spatial
Lite file.
Geospatial data continues to be required for subbasins when the
HMR52 precipitation method is used. The old spatial data must be
imported before the simulations using the HMR52 precipitation
method can be computed successfully.

Grid Cell Files and Grid Regions

Grid cell files have been associated with gridded precipitation and
the ModClark transform. The grid cell file contains a section for
each subbasin in the basin model. Each section is actually a list of
the grid cells for the subbasin with key properties including the
coordinates of each cell, area within the subbasin, and travel index.
The grid cell file has almost always been created by the companion
product HEC-GeoHMS.
Grid regions are a new project component designed to make it
easier to define the grid cells for a basin model. Each grid region
includes a coordinate system, cell size, and extent. This
information is implicit when using grid cell files created by HEC-
GeoHMS. Future program versions will allow for the creation of
grid regions based on the Standard Hydrologic Grid (SHG),
Hydrologic Rainfall Analysis Project (HRAP), and Universal
Transverse Mercator (UTM), all of which require explicit
specification of the coordinate system, cell size, and extent.
Old projects with a grid cell file are automatically upgraded when
opened with the new version. The grid cell file is converted to a
grid region that allows for backward compatibility. The grid region
created during upgrade is automatically configured in all of the
basin models that used the grid cell file. Any simulations using the
basin model will continue to operate just as they did in previous
versions.
New basin models that will utilize gridded simulation methods will
require a grid region. The best approach is to create a grid region
and link it to the grid cell file created by HEC-GeoHMS.

Muskingum Cunge 8-Point Cross Section

Previous versions required the elevations in the 8-point cross
section to be strictly greater than zero. This effectively prevented

11
HEC-HMS Version 4.3 Release Notes


use in coastal watersheds that otherwise met the usage criteria for
Muskingum Cunge routing. Negative elevations are now allowed.
However, a warning message is issued because in most
circumstances a negative elevation is incorrect.

Muskingum Cunge and Kinematic Wave Index Flow

The Muskingum Cunge and kinematic wave reach routing methods
are similar in that both subdivide the reach into a number of equal-
length segments. These segments are often called "dx" because
the need for them arises when developing a numerical solution
scheme for the underlying equations. An important part of
determining the length of each dx is the consideration of the wave
celerity as expressed in the courant condition. All previous
versions automatically considered the courant condition without
input from the user. This was possible because the basin model
completed a simulation from upstream to downstream.
Consequently, the entire upstream hydrograph was available for a
reach using either of these two routing methods. An index flow
could be calculated automatically, from which the wave celerity
could be calculated, and the courant condition satisfied.
Major changes have been made in how the basin model completes
a simulation. These changes are foundational to powerful new
features that will be added in future program versions. Briefly, the
basin model no longer completes a simulation from upstream to
downstream. The practical effect of this change is that the
upstream hydrograph is no longer available for each reach. This
complicates the use of Muskingum Cunge and kinematic wave
routing because the index flow can no longer be determined
automatically. The user must now manually enter either an index
flow or an index wave celerity. If the index flow is entered, the
wave celerity is automatically calculated from the index flow using
hydraulic formulas related to the cross section shape. The entered
or computed wave celerity is then used to satisfy the courant
condition when calculating the dx length. In the case of Muskingum
Cunge routing, the user also has the option of explicitly specifying
the "dx" and "dt" in which case the index flow or index celerity is not
needed. Any existing basin model that uses Muskingum Cunge or
kinematic wave routing will need to have either index flow or index
celerity added to each reach element. This change in the
specification of the index flow may change computed results by
small amounts.

12
 HEC-HMS Version 4.3 Release Notes


Kinematic Wave Transform

The kinematic wave transform in the subbasin is typically used to
represent urban hydrology. This application is different from flood
wave routing in reach channels, but the fundamental equations are
the same. The program uses common algorithms for both reaches
and subbasins. This means that in general the changes described
above for reaches also apply for subbasins. However, there are
some differences in subbasins that make it possible to avoid
requiring the entry of an index flow. The main difference is that
runoff in the subbasin is caused by precipitation. Especially in
urban areas where infiltration rates are typically small, there is a
strong correlation between the maximum precipitation rate and the
maximum runoff flow. The maximum precipitation rate for the
simulation is automatically identified and used with the area of each
plane to compute an estimated peak flow. This estimated peak
flow is used to compute the segment length dx for each plane and
each channel. As a final control, the user still has the option of
specifying the minimum number of spatial steps.

Unit Hydrograph Volume Check

A volume check has been added to all unit hydrograph methods in
the subbasin: Clark, Snyder, SCS, s-graph, and user specified. A
note is now displayed with the computed volume of runoff in the
unit hydrograph. If the volume is significantly different from 1 mm
(1 inch) then a warning will be displayed. The note or warning does
not stop continued simulation for each subbasin using a unit
hydrograph, but is provided for information purposes.

Reservoir Culvert Outlet Structure

The culvert outlet was originally added to the reservoir element for
HEC-HMS Version 3.0 (2005). The algorithms for the culvert outlet
were adapted from the HEC-RAS program. Since that time, the
original algorithms used in HEC-RAS have been enhanced based
on several unique test cases that became available. The
enhancements primarily affect the transition from submerged flow
to pressurized flow. The resulting differences in flow rate through
the culvert are relatively small on a percentage basis. The results
both below and above the transition zone are almost identical. This
update in the algorithms may change computed results for some
previous projects.

Reservoir Piping Breach

The reservoir element includes the option for a dam breach, either
by overtopping failure or by piping failure. An error was discovered

13
HEC-HMS Version 4.3 Release Notes


in the piping failure under some circumstances. Parameter

combinations with the piping elevation above the midpoint of the
top and bottom elevations caused the incorrect results. Results
remained correct if the piping elevation was at or below the
midpoint. The piping breach has been corrected so that all possible
piping elevations produce correct results. This correction may
change computed results for some parameter combinations.

Sign Convention in Latitude and Longitude

Version 4.1 and earlier versions only used latitude and longitude for
the inverse distance precipitation method. The standard
convention for specifying longitude required positive values east of
the Greenwich meridian and negative values west of the meridian.
Specifying latitude required positive values north of the Equator and
negative values south of the Equator. The User’s Manual stated
that negative values could be excluded if all subbasins were in the
same hemisphere. Version 4.1 added new capabilities for
shortwave radiation, longwave radiation, and evapotranspiration
that required latitude and longitude. In general these new methods
require strict adherence to the sign convention. The User’s Manual
has been updated to always discuss sign convention as a strict
requirement.

Bristow Campbell Radiation

It was discovered that the Bristow Campbell shortwave radiation
method was not correctly using the specified longitude. The
method is intended to internally calculate a time zone offset
between the subbasin longitude and the central longitude of the
time zone containing the subbasin. It was found that the time zone
of the user’s computer was used instead of the time zone
containing the subbasin. The internal calculation has been
changed to correctly determine the time zone offset. This
correction may change computed results for previous projects.

FAO56 Radiation
It was discovered that the FAO56 shortwave and longwave
radiation methods were not correctly using the specified longitude.
The methods are intended to internally calculate the number of
degrees west of the zero longitude line. This difference in degrees
should be used in calculations. It was found that the longitude of
the subbasin was used in error instead of calculating the difference.
The calculations have now been corrected to use the difference in
degrees between the subbasin and zero longitude. This correction
may change computed results for previous projects.

14
 HEC-HMS Version 4.3 Release Notes


Frequency Storm
The frequency storm precipitation method no longer allows an
unspecified or blank storm area. The storm area is used to
compute the correct depth-area reduction factor for the storm.
Previous versions allowed the storm area to be left blank, and then
automatically recomputed the storm for each subbasin using the
subbasin area as the storm area. This automatic behavior during a
simulation run is no longer allowed. A storm area must be specified
as part of the storm properties. There are still many applications
where it is necessary to compute results for various size storms
over the watershed. This application can be satisfied using the
depth-area analysis. The depth-area analysis allows the user to
select one or many locations in the basin model (potentially every
subbasin) and then the area in the frequency storm is automatically
adjusted to compute the correct storm at each selected location.
The conversion between annual duration and partial duration has
also been simplified. Choices are now provided for converting
annual duration input precipitation depths to partial duration output,
and for converting partial duration input to annual duration output.
These choices are provided for specific exceedance probabilities.
No conversion should be selected if the input data is annual
duration and an annual duration output is required. Additionally, no
conversion should be selected if the input data is partial duration
and a partial duration output is required. Finally, conversion is
never used for exceedance probabilities less frequent than 10%.

HMR52 and Standard Project Storm Precipitation

The HMR52 and Standard Project Storm precipitation methods
share certain characteristics. Both methods start from a
precipitation depth and a temporal pattern then build a time-series
of precipitation depths. The algorithms in both methods cannot be
applied to simulation time intervals less than five minutes. Previous
versions either stopped a simulation with a shorter time interval, or
experienced an uncontrolled error. Both methods have now been
enhanced to operate at a 1-minute time interval. This is
implemented by assuming a 5-minute time interval and then
dividing each of these intervals equally to produce a 1-minute
output.

Penman Monteith Evapotranspiration

The Penman Monteith evapotranspiration method is based on the
energy balance concept and consequently requires a number of
atmospheric variables: air pressure, windspeed, temperature, and
either dew point temperature or relative humidity. The equations

15
HEC-HMS Version 4.3 Release Notes


underlying the method are designed to use atmospheric data

recorded at an interval of 1 hour or less. The original equations are
used in HEC-HMS for all simulations with time intervals from 1
minute up to 1 hour.
The FAO56 report presents an alternate set of equations known as
reduced set that can be used with atmospheric data at intervals
greater than 1-hour, with daily interval being the most common.
Some atmospheric variables are not used at all with the reduced
set equations; informed assumptions are substituted instead. The
reduced set equations are now used in HEC-HMS for all
simulations with time intervals from 2 hours up to 1 day.
The computed results from Penman Monteith will be closest to the
intentions of the reduced set assumptions when the Hargreaves
shortwave radiation and FAO56 longwave radiation methods are
selected. For time intervals between 2 hours and 12 hours, the
equations are applied at daily interval and interpolated to the actual
interval.

Simulation Time Window

All simulations include a start time, end time, and time interval.
This information has always been checked to make sure it is valid.
For example, the end time must be after the start time, and the
difference between the start and end times must be a natural
number of intervals (no partial intervals). The time intervals that
may be used come from a list. The available intervals have been
carefully chosen to facilitate both data storage and automatic
interpolation.
Changes in how the basin model completes a simulation now
require the start time and time interval to be consistent.
Consistency is now enforced when checking the other time window
information. The start time must be a natural number of time
intervals after midnight. For example, a time interval of 6 hours can
be used with a start time of 00:00, 06:00, 12:00, or 18:00. This
means that some old simulations using inconsistent start time and
time interval will no longer be able to compute. This can be fixed
by adjusting the start time to be a natural number of intervals after
midnight.

Simulation Save States

Save states has been an optional feature of simulation runs. The
save states represent all of the information necessary to exactly
restore a basin model to a moment in time during a simulation run.
This includes, for example, subbasin soil moisture and excess

16
 HEC-HMS Version 4.3 Release Notes

precipitation in the process of moving over the land surface. Save
states could be created at the end of simulation run or at the end of
any time interval during the run. The save states could be used
later to initialize a new simulation run. It was required that the
basin model be the same in both runs. It was also required that the
simulation time of the save states exactly match the start time of
the run using them for initialization.
Save states can now be used with both simulation runs and
forecast alternatives, and are also much more flexible. Save states
can now be created by both runs and forecasts. The save states
can be created at the end of a run or forecast, at the forecast time
during a forecast, or at the end of any time interval during the run or
forecast. Once a save state has been created, it can be used with
any other simulation run or forecast alternative regardless of how it
was created. There are no limits due to the basin model used or
the simulation time represented by the information in the save
states. If the basin model used to create the save states does not
match the basin model in the simulation using the states, a warning
message will be shown and the data in the states still used. If the
simulation time used to create the save states does not match the
start time of the simulation using the states, a warning message will
be shown and the data in the states still used. If a basin model
element cannot be found in the save states, then the values from
the basin model will be used.
In summary, save states can still be used to take a very long
simulation and break it into a sequence of simulation runs. For this
use, the saves states are created at the end of a run and used to
start the next run. It is now also possible to treat the save states
like an optional replacement for the initial conditions specified in the
basin model either for simulation runs or forecast alternatives.

Problems Repaired
Several errors were found in the program and repaired. Complete
descriptions of each error are given below.

Observed Snow Water Equivalent

It is an option to add observed snow water equivalent to a subbasin
element. When included, the observed snow water equivalent is
used to calculate the residual snow water equivalent (observed
minus subbasin average). A unit conversion error was found in the
calculation, even though the units of the observed and subbasin
average values were correct. The calculation of the residual snow
water equivalent is now correct.

17
HEC-HMS Version 4.3 Release Notes


Outflow Curve Reservoir Routing

The outflow curve routing method in the reservoir element uses
storage as the state variable. There are several options for
specifying the initial value of the state variable, and the variable is
updated during the simulation based on inflow and a solution of the
Modified Puls algorithm with one routing step. There is an option in
the simulation run for “Save States.” Once saved, the states can
be used to initialize a new simulation run. It was found that the
outflow curve routing method did not initialize correctly when using
states at the beginning of the simulation run. This error has been
corrected.

Reservoir Evaporation
The outlet structures routing method in the reservoir allows various
ways of specifying how water leaves the storage. One of those
ways is by evaporation. The amount of evaporation can be
specified as a depth for each month of the year. It was discovered
that the evaporation for each time interval was calculated correctly,
but there was a unit conversion error when storing the time-series
of calculated evaporation. This means that the mass balance for
the reservoir storage was correct, but appeared wrong when
reviewing the stored time-series of evaporation. The unit
conversion error has been corrected and the stored time-series of
evaporation is now correct.

Forecast Alternative Initial Baseflow

The forecast alternative includes the option of automatically
computing subbasin initial baseflow using an observed flow gage.
It was discovered that under some circumstances, renamed flow
gages were not updated in the forecast alternative. Observed flow
gage selections in the forecast alternative are now always updated
when the gage is renamed.

Missing File When Opening a Project

As part of opening a project, the project is checked for missing file
references. Missing files may have been manually moved or
deleted since first being used in the project. When missing files are
identified, a window is presented to the user for an opportunity to
find the missing files and update them in the project. It was
discovered that files used in some components of the project were
not checked or where incorrectly displayed in the window. All
project components are now checked and missing files correctly
displayed in the window.

18
 HEC-HMS Version 4.3 Release Notes


Missing Precipitation Data

The meteorologic model checks for missing values in all of the
precipitation data prior to beginning calculations. Users have the
option of replacing missing data with a zero value, but it remains
important to correctly determine and report in an error message the
number of missing values. It was found that if a precipitation gage
was missing a value at exactly the start time of the simulation, it
would be reported in an error message. This is incorrect because
precipitation data is only required beginning with the first time
interval after the simulation start time. The check for missing
precipitation values has been corrected.

Subbasin Canopy with Kinematic Wave Transform

It was discovered that the program interface would become
unresponsive (frozen) when attempting to use any canopy method
in combination with the kinematic wave transform method. This
has been corrected. It is now possible to use a canopy method
with the kinematic wave transform method.

Global Editors and Column Reordering

It was discovered that all of the global editors in the basin model
allowed the user to reorder the columns. While this flexibility
allowed the user to customize the look of the global editors, it also
caused all of the parameter data to be saved incorrectly. Fully
supporting column reordering would have required a large amount
of work. This issued was fixed by locking the columns and
eliminating reordering.

Gage Weights Precipitation

It was discovered that a simulation could fail if the gaged weights
for a subbasin did not sum to 1.0. No message was displayed
when the error occurred and the simulation attempted to continue.
This condition is now checked and an appropriate message
displayed before stopping the simulation.

HMR52 Precipitation
It was discovered that under some circumstances the calculated
precipitation was displayed with values in millimeters but labels in
inches. However, the calculated values were correctly used in
subsequent calculations for infiltration and surface runoff. This
error was limited to basin models using the ModClark transform
method, and limited to the incorrect units label. All other transform
methods produced a precipitation display with values and labels

19
HEC-HMS Version 4.3 Release Notes


that matched correctly. The precipitation display has been

corrected so that the units match the values for ModClark in
addition to all other transform methods.

Copy Project
It was discovered that some project components were “lost” when
making a copy of a project. All components of the project were
reviewed to be sure they are included during the copy process.
The copy now correctly includes all components of the original
project.

Temperature Index Snowmelt Default Parameters

The temperature index snowmelt method includes several optional
parameters: rain rate limit, ATI-meltrate coefficient, cold limit, ATI-
coldrate coefficient, and water capacity. If an optional parameter is
left blank in the Component Editor, then a default value will be used
instead. It was discovered that the optional parameters were not
checked correctly at the beginning of a simulation. Further, it could
be ambiguous to the user exactly what default value was used.
The parameter checking has been made consistent and the clarity
of the messages improved.

XML Reports
It was discovered that subbasin components (canopy, surface, loss
rate, transform, and baseflow) were not placed below the subbasin
at the correct nesting level in the basin model XML file. It was also
discovered that the reservoir peak storage was not labeled correctly
in the simulation run XML file. For both reports, it was found that
the decimal separator used inside the XML files was not correct if
the program settings were configured to use a comma as the
decimal separator. Both XML reports have been corrected of all
three issues.

Import Meteorologic Model

There is a feature on the File menu for importing a meteorologic
model to the current project. The feature includes the option of
importing time-series gages, paired data, and grid data used by the
meteorologic model. It was discovered that grid data was not
imported correctly. The import feature for meteorologic models has
been corrected.

20
 HEC-HMS Version 4.3 Release Notes


Forecast Alternative Zone Configurations

The forecast alternative uses zone configurations to manage
adjustments for subbasin parameters in loss rate, transform, and
baseflow components. A zone configuration can also be used to
manage adjustments for reach parameters in the routing
component. Each zone configuration can be applied to subbasins
or reaches, but not both. It was discovered that some aspects of
routing parameter management were not correctly performed with
respect to the zone configuration selected for reaches. This could
lead to subbasins being displayed instead of reaches. Zone
configuration and parameter adjustment management has now
been corrected for subbasins and reaches.

Kinematic Wave Transform

It was discovered that kinematic wave transform parameters may
not be loaded correctly when opening a basin model by clicking on
it in the Watershed Explorer. While parameters entered through
the program were always processed correctly, there could be a
problem when manually editing the basin model file outside the
program. Processing has been improved so that even if
parameters are incorrectly modified in the basin model file, the
program can recover and correct the problem.

Meteorologic Model Parameter Tracking

The program is designed to automatically track all parameters and
other data used in the meteorologic model. This tracking is used to
determine if the last computed results are current or expired. It was
discovered that some types of data in the meteorologic model were
not tracked correctly. The tracking has been improved to cover all
parameters and all other data.

Documentation
The Hydrologic Modeling System HEC-HMS: Quick Start Guide
(Version 4.3 August 2018) provides a brief description of the
program for new users. It describes the different parts of the
interface and the basic steps necessary to obtain simulation results.
A tutorial takes the user through the creation of a new project and
shows how to obtain results. The guide has been updated to reflect
changes in the interface.
The Hydrologic Modeling System HEC-HMS: User's Manual
(Version 4.3 August 2018) contains extensive information on
installing and using the program. Details on the use of each of the
features and capabilities in the program are included. The manual

21
HEC-HMS Version 4.3 Release Notes


has been updated with information describing new features added

to the program for this Version 4.2 release.
The Hydrologic Modeling System HEC-HMS: Validation Guide
(Version 4.3 August 2018) contains information on the procedures
used to certify the software for release. The manual describes the
tests that have been established and the procedures used for
determining the correct result for each test. An accompanying data
kit includes all of the project data necessary to replicate the tests
performed at HEC prior to certifying a new release for distribution.
The Hydrologic Modeling System HEC-HMS: Applications Guide
(March 2015) illustrates how to apply the program in a variety of
different studies. Data requirements, calibration procedures, and
typical results presentation styles are provided for studies ranging
from flood risk management to sediment yield.
The Hydrologic Modeling System HEC-HMS: Technical Reference
Manual (March 2000) continues to accurately describe the
mathematical models included in the program. New simulation
capabilities have been added to the program and are not included
in the manual. The manual is currently undergoing a major revision
to expand documentation of existing mathematical models and fully
describe the newly added models.

Support Policy
Technical support for program users within the U.S. Army Corps of
Engineers is provided through an annual subscription service.
Subscribing offices can expect full support from HEC staff in the
routine application of the program. Users are strongly urged to
consult with HEC staff on the technical feasibility of using the
program before beginning a project with unique requirements such
as grid cell hydrology, snow melt, continuous simulation, sediment
transport, or water quality. Special guidance is also available for
the development of models for flow forecasting. Extended support
for large or complex projects can be arranged under a separate
reimbursable project agreement.
Support cannot be provided to users outside the U.S. Army Corps
of Engineers. Several companies and organizations offer varying
levels of support, some through a fee-for-service support similar to
the support provided to subscribing Corps offices. Such service
agreements are between the user and the vendor and do not
include HEC staff. Vendors can be located through internet
searches.

22
 HEC-HMS Version 4.3 Release Notes

Reporting of suspected program errors is unrestricted and we will
reply to all correspondence concerning such errors. We are
continuously working to improve the program and possible bugs
should always be reported. Reports should include a written
description of the steps that lead to the problem and the effects that
result from it. If we cannot reproduce the reported problem, we
may ask you to send a copy of your project.
Request support or report program errors through the following
channels:

• Visit our web site at http://www.hec.usace.army.mil.

• Send email to hec.hms@usace.army.mil on the internet.

• Fax 530.756.8250 any time.

• Call +1.530.756.1104, 7:30 am to 4:30 pm Pacific Time

Monday through Friday.
The postal address for the development team:
U.S. Army Corps of Engineers
Institute for Water Resources
Hydrologic Engineering Center
609 2nd Street
Davis, CA 95616
United States

23