United States Office of Water EPA-823-ROO-012
Environmental Protection 4305 July 2000
rxEPA BASINS Technical Note 6
Estimating Hydrology and Hydraulic
Parameters for HSPF
-------
CONTENTS
Introduction 1
Air Temperature (ATEMP) Parameters 2
Snow (SNOW) Parameters 3
Pervious Land Hydrology (PWATER) Parameters 7
Impervious Land Hydrology (IWATER) Parameters 18
Flow Routing (HYDR and ADCALC) Parameters 20
References 24
Parameter and Value Range Summary Tables 26
-------
BASINS Technical Note 6
Estimating Hydrology and Hydraulic Parameters for HSPF
July, 2000
Introduction
This technical note provides BASINS users with guidance in how to estimate the input parameters
in the ATEMP, SNOW, PWATER, IWATER, HYDR, and ADCALC sections of the
Hydrological Simulation Program Fortran (HSPF) watershed model. For each input parameter,
this guidance includes a parameter definition, the units used in HSPF, and how the input value
may be determined (e.g. initialize with reported values, estimate, measure, and/or calibrate).
Where possible, the note discusses how to estimate initial values using the data and tools included
with BASINS. Also discussed, where appropriate, is the physical basis of each parameter and the
corresponding algorithms as described in the HSPF Users Manual (Bicknell, et al, 1997) and
earlier literature sources. In addition to the guidance provided below, model users are directed to
other sources, including the ARM Model User Manual (Donigian and Davis, 1978), as well as
studies of agricultural BMP representation with HSPF (Donigian and Crawford, 1973;Donigian et
al, 1983; and Gasman, 1989).
Summary tables are attached that provide 'typical' and 'possible' (i.e. maximum 'expected')
ranges for the parameters discussed below, based on both the parameter guidance and experience
with HSPF over the past two decades on watersheds across the U. S. and abroad (Donigian,
1998). The overarching principal in parameter estimation should be that the estimated
values must be realistic, i.e. make 'physical' sense, and must reflect conditions on the
watershed. If the values estimated by the model user and/or derived from the guidance below, do
not agree with the value ranges in the summary table, the user should question and re-examine the
estimation procedures. The estimated values may still be appropriate, but the user needs to
confirm that the parameter values reflect unusual conditions on the watershed.
Another source of parameter information is the HSPF Parameter Database (HSPFParm) (US
EPA, 1999). Developed by AQUA TERRA Consultants, under contract to the EPA, HSPFParm
consists of parameter values from previous applications of HSPF across North America
assimilated into a single database, and with a customized graphical user interface for viewing and
exporting the data. The pilot HSPFParm Database contains parameter values for model
applications in over 40 watersheds in 14 states. The parameter values, contained in the database,
characterize a broad variety of physical settings, land use practices, and water quality constituents.
The database has been provided with a simplified interactive interface that enables modelers to
access and explore the HSPF parameter values developed and calibrated in various watersheds
across the United States, and to assess the relevance of the parameters to their own watershed
setting.
Page 1 of 32
-------
The parameter guidance below is listed in order of the parameter tables required by each module
section (i.e. ATEMP, SNOW, PWATER, IWATER, HYDR, and ADCALC) in the HSPF UCI
(users control input) file, and the parameters are grouped as required in each UCI table.
Page 2 of 32
-------
Air Temperature (ATEMP) Parameters
The ATEMP section variables are used by both the PERLND and IMPLND modules. This
section is not required for basic hydrology unless SNOW is being simulated.
ATEMP-DAT Table:
ELD AT Elevation difference (feet), (measure). ELD AT is the difference in elevation between
the air temperature gage and the mean elevation of the associated pervious land
segment (PLS). ELD AT is equal to the PLS elevation minus the gage elevation and
can be either positive (PLS higher) or negative (gage elevation higher). ELDAT is
used to adjust the gage air temperature to the PLS using a lapse rate; see Section
4.2(1). 1 in the HSPF User Manual (Bicknell et al., 1997) for additional information.
Use of BASINS Data/Tools:
Weather station elevation is available in BASINS from the Elev_ft field in the WDM
Weather Data Stations theme attribute table. To get the mean watershed elevation,
run a Watershed Topographic Report. The mean elevation is available in the
Elevation Report.
AIRTMP Initial air temperature (degrees F), (measure/ estimate). Air temperature at start of
simulation period.
Use of BASINS DatafTooh:
Open the WDM file in WDM Utility (WDMUtil) and select the hourly air temperature
time series (ATEM) for the weather station to be used in the simulation. Specify both
the current start and end dates as the model simulation start date and select an hourly
time step. Use the List/Edit Time Series function to display the air temperature for the
starting hour of the simulation.
Page 3 of 32
-------
Snow (SNOW) Parameters
The SNOW section variables are used in both the PERLND and IMPLND modules. Note: flag
CSNOFG in PWATER and IWATER must be selected for snow melt to be taken into account.
ICE-FLAG Table:
The ICE-FLAG table contains only the flag to simulate ice formation in the snow pack. A value
of 0 means ice formation is not simulated; 1 means that frozen water or frozen ground at the
bottom of the snowpack is computed daily based on air temperature and inches of melt in the
snowpack. This option is appropriate in regions where frozen ground conditions are observed.
SNOW-PARM1 Table:
LAT Latitude (degrees), (measure). LAT is positive for the northern hemisphere,
negative for the southern hemisphere. LAT is used in the calculation of snow
surface albedo to identify summer and winter time periods.
Use of BASINS Data/Tools:
The Lat_dd field in the WDM Weather Data Stations theme attribute table
contains the latitude of the station in decimal degrees.
MELEV Mean elevation (feet), (measure). MELEV is used to compute the convective
heat flux from the atmosphere to the snowpack.
Use of BASINS Data/Tools:
Generate a Watershed Topographic Report. The mean elevation for the watershed
is located in the Elevation Report.
SHADE Fraction of the land segment shaded from solar radiation by trees or slope
(unitless), (measure/estimate). SHADE controls short wave solar radiation that
reaches the snowpack. SHADE is also used in the calculation of long wave
radiation, based on Stefan's law of black body radiation (Bicknell et al, 1997).
Use of BASINS DatafTooh:
Estimate as the fraction of the watershed with coniferous forest (see FOREST in
PWATER).
SNOWCF Factor by which recorded precipitation is multiplied during snow events to account
for poor gage catch efficiency (unitless), (estimate). Snow catch in a gage is
affected by wind speed, instrument characteristics (e.g. snow shields), and gage
exposure/location. SNOWCF is normally 1.0 or greater. Crawford (1999) states
that physically realistic values of SNOWCF are in the range of 1.0 to 1.5. Larson
and Peck, 1974 present data relating snow catch factor (SCF - a SNOWCF
analogue) to windspeed. For unshielded gages, they show a parabolic curve for
Page 4 of 32
-------
SCF with a value of 1.0 at 0.0 mph, 1.3 at 5 mph, and 1.8 at 10.0 mph. They also
show a plot of the ratio of gage catch to true catch (the inverse variable) as
roughly straight line plots, both starting from 1.0 at 0.0 mph. The shielded gage is
roughly 0.75 at 10 mph, and the unshielded is roughly 0.55 at 10 mph. Only 200
out of 3500 weather stations in the U.S. are shielded; shielding of weather stations
began in the mid 1940's (Doty, 2000).
COVIND Maximum snowpack depth (water equivalent) at which the entire land segment is
covered with snow (inches), (estimate). COVIND is a function of topography and
climatic conditions. The ARM Model User Manual (Donigian and Davis, 1978, p.
65, MPACK variable) says typical values are in the range of 1.0 to 6.0 inches. For
mostly flat topography and where snow is a common occurrence, values will be
near the low end of this range since complete coverage of the segment will result
from low snowfall amounts; whereas in mountainous watersheds the COVIND
values will be near the high end of the range reflecting the need for greater
snowfall to achieve complete coverage. Note that COVIND is in water equivalent,
so that a value of 1.0 inches corresponds to approximately 10 inches of snow.
SNOW-PARM2 Table:
RDCSN Density of new snow relative to water when air temperature is at or below 0
degrees F (unitless ratio), (estimate, then calibrate). The expected values are
from 0.10 to 0.20. RDCSN can be calibrated by comparing field snow density
data from snow courses with model snowpack (PACK) and density results
(RDENPF). RDCSN is automatically adjusted in the model when the air
temperature is above 0 degrees F. The snow density/ temperature functional
relationship is described by Bicknell et al., (1997).
TSNOW Wet bulb air temperature below which precipitation occurs as snow under
saturated conditions (degrees F), (estimate, then calibrate). Values of 31 to 33
degrees F are often used (Donigian and Davis, 1978). When the air temperature
exceeds the value of TSNOW, precipitation is modeled as rain, not snow.
TSNOW can vary between 30 and 40 degrees F; the initial estimate should be at or
near 32 degrees F. Crawford (1999) states that TSNOW is the most obvious
parameter to change to increase or reduce snow accumulation. However,
changing TSNOW is effective only when significant snowfall occurs at or near 32
degrees F. If typical winter air temperatures are below 30 degrees F then TSNOW
will have little or no effect.
SNOEVP Factor to adjust evaporation (sublimation) from the snowpack (unitless),
(estimate, then calibrate). Values near 0.1 are expected (Donigian and Davis,
1978, p. 65, EVAPSN variable). Evaporation from the snowpack occurs only
Page 5 of 32
-------
when the vapor pressure of the air is less than that at the snow surface (Bicknell et
al, 1997). Evaporation occurs only from the frozen content of the snowpack and
is adjusted based on SNOEVP, wind movement, and the fraction of the land
segment covered by the snowpack. Snow evaporation is not large in most
watersheds, but can be important where windy, low humidity conditions are
common (Crawford, 1999).
CCFACT Factor to adjust the rate of heat transfer from the atmosphere to the snowpack,
due to condensation and convection, to match field conditions (unitless),
(estimate, then calibrate). CCFACT is a function of climatic conditions. Typical
values are near 1.0, although a range of 0.5 to 2.0 has been observed (Donigian
and Davis, 1978, p. 64, CCFAC variable). CCFACT is used in conjunction with
wind movement and air temperature to compute heat transfer from the snowpack
to the land surface.
MWATER Maximum liquid water holding capacity in the snowpack (in/in), (estimate, then
calibrate). MWATER is a function of the mass of ice layers; the size, shape, and
spacing of snow crystals; and the degree of channelization and honeycombing of
the snowpack to allow liquid water accumulation. Experimental values range from
0.01 to 0.05, with 0.03 a common value (Donigian and Davis, 1978, p. 64, WC).
MGMELT Maximum rate of snowmelt by ground heat (in/day), (estimate, then calibrate).
MGMELT is the rate of melt when the snowpack temperature is at 32 degrees F.
A standard value is 0.01 in/day. Areas with deep frost penetration and/or frozen
ground have MGMELT values approaching zero (Donigian and Davis, 1978, p.
64, DGM).
SNOW-INIT1 Table:
Pack-snow Initial quantity of snow (water equivalent) in the snowpack (inches), (estimate). If
the simulation starts at the beginning of the water year (October 1) Pack-snow is
usually set to zero, except in arctic climates.
Pack-ice Initial quantity of ice (water equivalent) in the snowpack (inches), (estimate). If
the simulation starts at the beginning of the water year (October 1) Pack-ice is
usually set to zero, except in arctic climates.
Pack-watr Initial quantity of liquid water in the snowpack (inches), (estimate). If the
simulation starts at the beginning of the water year (October 1) Pack-watr is
usually set to zero, except in arctic climates.
Page 6 of 32
-------
RDENPF Initial density of the frozen contents (Pack-snow and Pack-ice) of the snowpack
relative to water (unitless ratio), (estimate). If the simulation starts at the
beginning of the water year (October 1) RDENPF is usually set to 0.01 (minimum
value), except in arctic climates; otherwise see RDCSN.
DULL Initial index to the dullness of the snowpack surface, from which the initial albedo
is estimated (unitless), (estimate). If the simulation starts at the beginning of the
water year (October 1) DULL is usually set to zero (for perfectly reflectable
snow), except in arctic climates. DULL ranges from zero to 800 and is an
empirical index. It's increased by one for each hour in which new snow does not
fall.
PAKTMP Initial mean temperature of the frozen contents of the snowpack (degrees F),
(estimate). If the simulation starts at the beginning of the water year (October 1)
PAKTMP is usually set to 32 degrees F, except in arctic climates.
SNOW-INIT2 Table:
COVINX Initial snowpack depth (water equivalent) required for the entire land segment to
be covered with snow (inches), (estimate). If the simulation starts at the beginning
of the water year (October 1) COVINX is usually set to its default value of 0.01,
except in arctic climates; otherwise see COVIND, SNOW-PARM1.
XLNMLT Initial increment to ice storage in the snowpack (inches), (estimate). XLNMLT
represents an equivalent heat deficit that must be overcome before snowmelt is
released from the pack; otherwise some portion of the potential melt will freeze
and become pack ice. For most simulations, XLNMLT can usually be set to zero,
except possibly in arctic climates, because the values will be recalculated based on
current (usually hourly) air temperatures. XLNMLT is used only if ICE-FLAG is
set to 1.
SKYCLR Initial fraction of sky assumed to be clear (unitless), (estimate). Unless a storm is
in progress at the start of simulation period, set SKYCLR to 1.0 (no clouds).
Use of BASINS DatafTooh:
Open the WDM file with the WDM Utility (WDMUtil) and select the CLOU time
series for the weather station of interest. Identify the value of CLOU for the
starting date and time in the model simulation. Set SKYCLR to 1-CLOU.
Page 7 of 32
-------
Pervious Land Hydrology (PWATER) Parameters
PWAT-PARM1 Table:
The PWAT-PARM1 table includes flags to indicate the selected simulation algorithm option, or
the selection of monthly variability versus constant values for selected parameters. Where flags
indicate monthly variability, the corresponding monthly values must be provided in Monthly Input
Parameters (see below following the PWAT PARM4 Table section). That section also provides
guidance on which parameters are normally specified as monthly values.
CSNOFG Flag to use snow simulation data; must be checked (CSNOFG=1) if SNOW is
simulated.
RTOPFG Flag to select overland flow routing method; choose either the method used in
predecessor models (HSPX, ARM, and NFS) or the alternative method as
described in the HSPF User Manual. Recommendation: Set RTOPFG=1; This
method, used in the predecessor models is more commonly used, and has been
subjected to more widespread application.
UZFG Flag to select upper zone inflow computation method; choose either the method
used in predecessor models (HSPX, ARM, and NFS) or the more exact numerical
solution to the integral of inflow to upper zone, i.e the alternative method.
Recommendation: Set UZFG=1; This method, used in the predecessor models, is
more commonly used, and has been subjected to more widespread application.
VCSFG Flag to select constant or monthly-variable interception storage capacity, CEPSC.
Monthly value can be varied to represent seasonal changes in foliage cover;
monthly values are commonly used for agricultural, and sometimes deciduous
forest land areas.
VUZFG Flag to select constant or monthly-variable upper zone nominal soil moisture
storage, UZSN. Monthly values are commonly used for agricultural areas to
reflect the timing of cropping and tillage practices.
VMNFG Flag to select constant or monthly-variable Mannings n for overland flow plane,
NSUR. Monthly values are commonly used for agricultural, and sometimes
deciduous forest land areas.
VIFWFG Flag to select constant or monthly-variable interflow inflow parameter, INTFW.
Monthly values are not often used.
VIRCFG Flag to select constant or monthly varied interflow recession parameter, IRC.
Monthly values are not often used.
Page 8 of 32
-------
VLEFG Flag to select constant or monthly varied lower zone ET parameter, LZETP.
Monthly values are commonly used for agricultural, and sometimes deciduous
forest land areas.
PWAT-PARM2 Table:
FOREST Fraction of land covered by forest (unitless) (measure/estimate). FOREST is the
fraction of the land segment which is covered by forest which will continue to
transpire in winter (i.e. coniferous). This is only relevant if snow is being
considered (i.e., CSNOFG=1 in PWATER-PARM1).
Use of BASINS Data/Tools:
Run a Land Use Distribution Report on the watershed(s). Determine the acreage
for EVERGREEN FOREST LAND - 42. Estimate the fraction of MIXED
FOREST LAND - 43 acreage which is coniferous. Add this to the EVERGREEN
FOREST LAND - 42 acreage and divide the sum by the Forest land subtotal
acreage. This is the value of FOREST for the Forested land use in that watershed.
Estimate from field survey for all other land use types.
LZSN Lower zone nominal soil moisture storage (inches), (estimate, then calibrate}.
LZSN is related to both precipitation patterns and soil characteristics in the region.
The ARM Model User Manual (Donigian and Davis, 1978, p. 56, LZSN variable)
includes a mapping of calibrated LZSN values across the country based on almost
60 applications of earlier models derived from the Stanford-based hydrology
algorithms. LaRoche et al (1996) shows values of 5 inches to 14 inches, which is
consistent with the 'possible' range of 2 inches to 15 inches shown in the Summary
Table. Viessman, et al, 1989, provide initial estimates for LZSN in the Stanford
Watershed Model (SWM-IV, predecessor model to HSPF) as one-quarter of the
mean annual rainfall plus four inches for arid and semiarid regions, or one-eighth
annual mean rainfall plus 4 inches for coastal, humid, or subhumid climates. These
formulae tend to give values somewhat higher than are typically seen as final
calibrated values; since LZSN will be adjusted through calibration, initial estimates
obtained through these formulae may be reasonable starting values.
INFILT Index to mean soil infiltration rate (in/hr); (estimate, then calibrate). In HSPF,
INFILT is the parameter that effectively controls the overall division of the
available moisture from precipitation (after interception) into surface and
subsurface flow and storage components. Thus, high values of INFILT will
produce more water in the lower zone and groundwater, and result in higher
baseflow to the stream; low values of INFILT will produce more upper zone and
interflow storage water, and thus result in greater direct overland flow and
interflow. LaRoche et al (1996) shows a range of INFILT values used from 0.004
Page 9 of 32
-------
in/hr to 0.23 in/hr, consistent with the 'typical' range of 0.01 to 0.25 in/hr in the
Summary Table. Fontaine and Jacomino (1997) show sediment and sediment
associated transport to be sensitive to the INFILT parameter since it controls the
amount of direct overland flow transporting the sediment. Since INFILT is not a
maximum rate nor an infiltration capacity term, it's values are normally much less
than published infiltration rates, percolation rates (from soil percolation tests), or
permeability rates from the literature. In any case, initial values are adjusted in the
calibration process.
INFILT is primarily a function of soil characteristics, and value ranges have been
related to SCS hydrologic soil groups (Donigian and Davis, 1978, p.61, variable
INFIL) as follows:
SCS Hvdrologic INFILT Estimate
Soil Group (in/hr) (mm/hr) Runoff Potential
A 0.4-1.0 10.0-25.0 Low
B 0.1-0.4 2.5-10.0 Moderate
C 0.05-0.1 1.25-2.5 Moderate to High
D 0.01-0.05 0.25-1.25 High
An alternate estimation method that has not been validated., is derived from the
premise that the combination of infiltration and interflow in HSPF represents the
infiltration commonly modeled in the literature (e.g. Viessman et al, 1989, Chapter
4). With this assumption, the value of 2.0*INFILT*INTFW should approximate
the average measured soil infiltration rate at saturation, or mean permeability.
Use of BASINS DatafTooh:
Use of Soil Hydrologic Group/INFILT Table:
Use the identify tool on the State Soil theme to identify the map unit identification
numbers (Muid's) for each soil layer overlapping the watershed. Open the Soil
Component Data table and perform a query for all records for each map unit (e.g.
([Muid] = "PA044") or ([Muid] = "PA052") ) in the watershed. Highlight the
Hydgrp (soil hydrologic group) field and select summarize from the Table menu.
Use the summary, together with the above table to estimate an average INFILT
value for the watershed.
Use of Alternative 2.0*INFILT*INTFW Method:
The State Soil (STATSGO) data layer contains data on soil permeability, defined
in the 1996 National Soil Survey Handbook (Soil Survey Staff, 1996), as the rate
of water movement through completely saturated soils. Run the BASINS State
Soil Characteristic Report and select mean estimate, area-weighted, surface layer,
for permeability to get the value of 2*INFILT*INTFW. Note that although this
Page 10 of 32
-------
method has not been validated, it may produce reasonable starting values for
adjustment through calibration.
LSUR Length of assumed overland flow plane (ft) (estimate/measure). LSUR
approximates the average length of travel for water to reach the stream reach, or
any drainage path such as small streams, swales, ditches, etc. that quickly deliver
the water to the stream or waterbody. LSUR is often assumed to vary with slope
such that flat slopes have larger LSUR values and vice versa; typical values range
from 200 feet to 500 feet for slopes ranging from 15% to 1 %. It is also often
estimated from topographic data by dividing the watershed area by twice the
length of all streams, gullies, ditches, etc that move the water to the stream. That
is, a representative straight-line reach with length, L, bisecting a representative
square areal segment of the watershed, will produce two overland flow planes of
width l/2 L. However, LSUR values derived from topographic data are often too
large (i.e. overestimated) when the data is of insufficient resolution to display the
many small streams and drainage ways. Users should make sure that values
calculated from GIS or topographic data are consistent with the ranges shown in
the Summary Table.
Use of BASINS Data/Tools:
Since RF3 data is not of sufficient resolution, use of this BASINS data layer is not
recommended for estimating LSUR.
SLSUR Average slope of assumed overland flow path (unitless) (estimate/measure).
Average SLSUR values for each land use being simulated can often be estimated
directly with GIS capabilities. Graphical techniques include imposing a grid
pattern on the watershed and calculating slope values for each grid point for each
land use.
Use of BASINS DatafTools:
Within the BASINS GIS, identify DEM polygons, along the length of the reach(es)
being modeled, which happen to be bisected by the reach. Adjacent, uphill DEM
grid cells, then, contain the elevation one grid cell away (i.e. 300 meters,
measuring centerpoint to centerpoint). The approximate overland flow path slope
at that point in the reach is then the difference in elevation between the bisected
and the adjacent/uphill DEM grid cells, divided by the 300 meter (984 feet) width
of a grid cell. Make multiple estimates along the length of the reach(es) and use
these measurements to guide estimation of this parameter value.
KVARY Groundwater recession flow parameter used to describe non-linear groundwater
recession rate (/inches) (initialize with reported values, then calibrate as needed}.
KVARY is usually one of the last PWATER parameters to be adjusted; it is used
when the observed groundwater recession demonstrates a seasonal variability with
Page 11 of 32
-------
a faster recession (i.e. higher slope and lower AGWRC values) during wet periods,
and the opposite during dry periods. LaRoche, et al, 1996 reported an extremely
high 'optimized' value of 0.66 mm"1 or (17 in"1) (much higher than any other
applications) while Chen, et al, 1995 reported a calibrated value of 0.14 mm"1 (or
3.6 in"1). Value ranges are shown in the Summary Table. Users should start with a
value of 0.0 for KVARY, and then adjust (i.e. increase) if seasonal variations are
evident. Plotting daily flows with a logarithmic scale helps to elucidate the slope
of the flow recession.
AGWRC Groundwater recession rate, or ratio of current groundwater discharge to that
from 24 hours earlier (when KVARY is zero) (/day) (estimate, then calibrate).
The overall watershed recession rate is a complex function of watershed
conditions, including climate, topography, soils, and land use. Hydrograph
separation techniques (see any hydrology or water resources textbook) can be used
to estimate the recession rate from observed daily flow data (such as plotting on a
logarithmic scale, as noted above); estimated values will likely need to be adjusted
through calibration. Value ranges are shown in the Summary Table. LaRoche, et
al, 1996 reported an optimized value of 0.99; Chen, et al, 1995 reported values
that varied with land use type, ranging from 0.971 for grassland and clearings to
0.996 for high density forest; Fontaine and Jacomino, 1997 reported a calibrated
value of 0.99. This experience reflects normal practice of using higher values for
forests than open, grassland, cropland and urban areas.
PWAT-PARM3 Table:
PETMAX Temperature below which ET will be reduced to 50% of that in the input time
series (deg F), unless it=s been reduced to a lesser value from adjustments made in
the SNOW routine (where ET is reduced based on the percent areal snow
coverage and fraction of coniferous forest). PETMAX represents a temperature
threshold where plant transpiration, which is part of ET, is reduced due to low
temperatures (initialize with reported values, then calibrate as needed). It is only
used if SNOW is being simulated because it requires air temperature as input (also
a requirement of the SNOW module), and the required low temperatures will
usually only occur in areas of frequent snowfall. Use the default of 40°F as an
initial value, which can be adjusted a few degrees if required.
PETMIN Temperature at and below which ET will be zero (deg F). PETMIN represents the
temperature threshold where plant transpiration is effectively suspended, i.e. set to
zero, due to temperatures approaching freezing (initialize with reported values,
then calibrate as needed). Like PETMAX, this parameter is used only if SNOW
is being simulated because it requires air temperature as input (also a requirement
of the SNOW module), and the required low temperatures will usually only occur
Page 12 of 32
-------
in areas of frequent snowfall. Use the default of 35°F as an initial value, which can
be adjusted a few degrees if required.
INFEXP Exponent that determines how much a deviation from nominal lower zone storage
affects the infiltration rate (HSPF Manual, p. 60) (initialize with reported values,
then calibrate as needed). Variations of the Stanford approach have used a
POWER variable for this parameter; various values of POWER are included in
Donigian and Davis (1978, p. 58). However, the vast majority of HSPF
applications have used the default value of 2.0 for this exponent. Use the default
value of 2.0, and adjust only if supported by local data and conditions.
INFILD Ratio of maximum and mean soil infiltration capacities (initialize with reported
value). In the Stanford approach, this parameter has always been set to 2.0, so
that the maximum infiltration rate is twice the mean (i.e. input) value; when HSPF
was developed, the INFILD parameter was included to allow investigation of this
assumption. However, there has been very little research to support using a value
other than 2.0. Use the default value of 2.0, and adjust only if supported by local
data and conditions.
Use of BASINS DatafTools:
Run the BASINS State Soil Characteristic Report and select mean estimate, area-
weighted, surface layer, for permeability. The report lists the mean and maximum
permeability statistics by subwatershed. Use these values in conjunction with the
guidance provided for INFILT.
DEEPFR The fraction of infiltrating water which is lost to deep aquifers (i.e. inactive
groundwater), with the remaining fraction (i.e. 1-DEEPFR) assigned to active
groundwater storage that contributes baseflow to the stream (estimate, then
calibrate}. It is also used to represent any other losses that may not be measured
at the flow gage used for calibration, such as flow around or under the gage site.
This accounts for one of only three major losses from the PWATER water balance
(i.e. in addition to ET, and lateral and stream outflows). Watershed areas at high
elevations, or in the upland portion of the watershed, are likely to lose more water
to deep groundwater (i.e. groundwater that does not discharge within the area of
the watershed), than areas at lower elevations or closer to the gage (see discussion
and figures in Freeze and Cherry, 1979, section 6.1). DEEPFR should be set to
0.0 initially or estimated based on groundwater studies, and then calibrated, in
conjunction with adjustments to ET parameters, to achieve a satisfactory annual
water balance.
BASETP ET by riparian vegetation as active groundwater enters streambed; specified as a
fraction of potential ET, which is fulfilled only as outflow exists (estimate, then
calibrate). Typical and possible value ranges are shown in the Summary Table. If
Page 13 of 32
-------
significant riparian vegetation is present in the watershed then non-zero values of
BASETP should be used. Adjustments to BASETP will be visible in changes in
the low-flow simulation, and will effect the annual water balance. If riparian
vegetation is significant, start with a BASETP value of 0.03 and adjust to obtain a
reasonable low-flow simulation in conjunction with a satisfactory annual water
balance.
AGWETP Fraction of model segment (i.e. pervious land segment) that is subject to direct
evaporation from groundwater storage, e.g. wetlands or marsh areas, where the
groundwater surface is at or near the land surface, or in areas with phreatophytic
vegetation drawing directly from groundwater. This is represented in the model as
the fraction of remaining potential ET (i.e. after base ET, interception ET, and
upper zone ET are satisfied), that can be met from active groundwater storage
(estimate, then calibrate). If wetlands are represented as a separate PLS
(pervious land segment), then AGWETP should be 0.0 for all other land uses, and
a high value (0.3 to 0.7) should be used for the wetlands PLS. If wetlands are not
separated out as a PLS, identify the fraction of the model segment that meets the
conditions of wetlands/marshes or phreatophytic vegetation and use that fraction
for an initial value of AGWETP. Like BASETP, adjustments to AGWETP will be
visible in changes in the low-flow simulation, and will effect the annual water
balance. Follow above guidance for an initial value of AGWETP, and then adjust
to obtain a reasonable low-flow simulation in conjunction with a satisfactory
annual water balance.
PWAT PARM4 Table:
CEPSC Amount of rainfall, in inches, which is retained by vegetation, never reaches the
land surface, and is eventually evaporated (estimate, then calibrate). Typical
guidance for CEPSC for selected land surfaces is provided in Donigian and Davis
(1978, p. 54, variable EPXM) as follows:
Land Cover Maximum Interception (in)
Grassland 0.10
Cropland 0.10-0.25
Forest Cover, light 0.15
Forest Cover, heavy 0.20
Donigian et al (1983) provide more detail guidance for agricultural conditions,
including residue cover for agricultural BMPs. As part of an annual water
balance, Viessman, et al. 1989 note that 10-20% of precipitation during growing
season is intercepted and as much as 25% of total annual precipitation is
intercepted under dense closed forest stands; crops and grasses exhibit a wide
Page 14 of 32
-------
range of interception rates - between 7% and 60% of total rainfall. Users should
compare the annual interception evaporation (CEPE) with the total rainfall
available (PREC in the WDM file), and then adjust the CEPSC values accordingly.
(See Monthly Input Values below).
UZSN Nominal upper zone soil moisture storage (inches) (estimate, then calibrate).
UZSN is related to land surface characteristics, topography, and LZSN. For
agricultural conditions, tillage and other practices, UZSN may change over the
course of the growing season. Increasing UZSN value increases the amount of
water retained in the upper zone and available for ET, and thereby decreases the
dynamic behavior of the surface and reduces direct overland flow; decreasing
UZSN has the opposite effect. Donigian and Davis (1978, p. 54) provide initial
estimates for UZSN as 0.06 of LZSN, for steep slopes, limited vegetation, low
depression storage; 0.08 LZSN for moderate slopes, moderate vegetation, and
moderate depression storage; 0.14 LZSN for heavy vegetal or forest cover, soils
subject to cracking, high depression storage, very mild slopes. Donigian et al.,
(1983) include detailed guidance for UZSN for agricultural conditions. LaRoche
shows values ranging from 0.016 in to 0.75 in. Fontaine and Jacomino showed
average daily stream flow was relatively insensitive to this value but sediment and
sediment associated contaminant outflow was sensitive; this is consistent with
experience with UZSN having an impact on direct overland flow, but little impact
on the annual water balance (except for extremely small watersheds with no
baseflow). Typical and possible value ranges are shown in the Summary Table.
NSUR Manning's n for overland flow plane (estimate). Mannings n values for overland
flow are considerably higher than the more common published values for flow
through a channel, where values range from a low of about 0.011 for smooth
concrete, to as high as 0.050-0.1 for flow through unmaintained channels (Hwang
and Hita, 1987). Donigian and Davis (1978, p. 61, variable NN) and Donigian et
al (1983) have tabulated the following values for different land surface conditions:
Smooth packed surface 0.05
Normal roads and parking lots 0.10
Disturbed land surfaces 0.15 - 0.25
Moderate turf/pasture 0.20-0.30
Heavy turf, forest litter 0.30 - 0.45
Agricultural Conditions
Conventional Tillage 0.15-0.25
Smooth fallow 0.15-0.20
Rough fallow, cultivated 0.20 - 0.30
Crop residues 0.25 - 0.35
Meadow, heavy turf 0.30-0.40
Page 15 of 32
-------
For agricultural conditions, monthly values are often used to reflect the seasonal
changes in land surfaces conditions depending on cropping and tillage practices.
Additional tabulations of Manning's n values for different types of surface cover
can be found in: Weltz, et al, 1992; Engman, 1986; and Mays, 1999. Manning's n
values are not often calibrated since they have a relatively small impact on both
peak flows and volumes as long as they are within the normal ranges shown above.
Also, calibration requires data on just overland flow from very small watersheds,
which is not normally available except at research plots and possibly urban sites.
INTFW Coefficient that determines the amount of water which enters the ground from
surface detention storage and becomes interflow, as opposed to direct overland
flow and upper zone storage (estimate, then calibrate). Interflow can have an
important influence on storm hydrographs, particularly when vertical percolation is
retarded by a shallow, less permeable soil layer. INTFW affects the timing of
runoff by effecting the division of water between interflow and surface processes.
Increasing INTFW increases the amount of interflow and decreases direct overland
flow, thereby reducing peak flows while maintaining the same volume. Thus it
affects the shape of the hydrograph, by shifting and delaying the flow to later in
time. Likewise, decreasing INTFW has the opposite effect. Base flow is not
affected by INTFW. Rather, once total storm volumes are calibrated, INTFW can
be used to raise or lower the peaks to better match the observed hydrograph.
Typical and possible value ranges are shown in the Summary Table.
IRC Interflow recession coefficient (estimate, then calibrate). IRC is analogous to the
groundwater recession parameter, AGWRC, i.e. it is the ratio of the current daily
interflow discharge to the interflow discharge on the previous day. Whereas
INTFW affects the volume of interflow, IRC affects the rate at which interflow is
discharged from storage. Thus it also affects the hydrograph shape in the 'falling'
or recession region of the curve between the peak storm flow and baseflow. The
maximum value range is 0.3 - 0.85, with lower values on steeper slopes; values
near the high end of the range will make interflow behave more like baseflow,
while low values will make interflow behave more like overland flow. IRC should
be adjusted based on whether simulated storm peaks recede faster/slower than
measured, once AGWRC has been calibrated. Typical and possible value ranges
are shown in the Summary Table.
Page 16 of 32
-------
LZETP Index to lower zone evapotranspiration (unitless) (estimate, then calibrate).
LZETP is a coefficient to define the ET opportunity; it affects evapotranspiration
from the lower zone which represents the primary soil moisture storage and root
zone of the soil profile. LZETP behaves much like a 'crop coefficient' with values
mostly in the range of 0.2 to 0.7; as such it is primarily a function of vegetation;
Typical and possible value ranges are shown in the Summary Table, and the
following ranges for different vegetation are expected for the 'maximum' value
during the year:
Forest 0.6 - 0.8
Grassland 0.4-0.6
Row crops 0.5 - 0.7
Barren 0.1 -0.4
Wetlands 0.6-0.9
Page 17 of 32
-------
Monthly Input Parameter Tables:
In general, monthly variation in selected parameters, such as CEPSC and LZETP should be
included with the initial parameter estimates. However, adjustments to the monthly values should
be addressed only after annual flow volumes are matched well with monitored data. All monthly
values can be adjusted to calibrate for seasonal variations.
MON-INTERCEP Table:
Monthly values for interception storage. Monthly values can be developed based on the data
presented in the discussion in PWAT-PARM4/CEPSC and the Summary Tables.
MON-UZSN Table:
Monthly values for upper zone storage. For agricultural areas under conventional tillage, lower
values are used to reflect seedbed preparation in the spring with values increasing during the
growing season until harvest and fall tillage. See PWAT-PARM4/UZSN discussion and Summary
Tables for guidance.
MON-MANNING Table:
Monthly values for Mannings n for the overland flow plane. Monthly values can be used to
represent seasonal variability in ground cover including crop and litter residue. See discussion in
PWAT-PARM4/NSUR for Manning's n as a function of agricultural conditions.
MON-INTERFLW Table:
Monthly values for interflow parameter (INTFW) are not often used.
MON-IRC Table:
Monthly values for interflow recession parameter are not often used.
MON-LZETPARM Table:
Monthly values for LZETP for evapotranspiration from the lower zone can be developed using an
expected maximum value from the PWAT-PARM4/LZETP discussion and the range of values
presented in the Summary Tables. Monthly variable values should be used to reflect the
seasonality of evapotranspiration, in response to changes in density of vegetation, depth of root
zone, and stage of plant growth.
Page 18 of 32
-------
PWAT-STATE1 Table:
CEPS, SURS, IFWS, UZS, LZS, AGWS, are initial values for storage of water in interception,
surface ponding, interflow, the upper zone, lower zone, and active groundwater, respectively, and
GWVS is the initial index to groundwater slope. All these storages pertain to the first interval of
the simulation period.
The surface related storages (i.e. CEPS, SURS, IFWS) are highly dynamic, and will reach a
dynamic equilibrium within a few days, at most. These state variables can be left blank, or set to
0.0 unless an individual storm is being simulated. The soil storages (i.e. UZS, LZS, and AGWS,
and the GWVS) are much less dynamic, so their beginning values can impact the simulation for a
period of months to a few years. If possible, users should allow as long a startup time period as
possible (i.e. set the simulation period to begin prior to the period you=ll use for comparison
against monitoring data or other use); as noted each of these storages should reach a dynamic
equilibrium within a few years of simulation. UZS and LZS should be set equal to UZSN and
LZSN respectively, unless it is known that the starting date is during a particularly wet or dry
period; starting values can be increased or decreased if wet or dry conditions were evident prior
to the simulation period. AGWS is a bit more problematic. If far too high or too low, baseflow
will be excessive or skewed low for several months or years, depending on AGWRC and
KVARY. Improper values of GWVS can also cause simulation accuracy problems again for
lengths of time depending on values of AGWRC and KVARY. However, since when KVARY is
set to 0.0 seasonal recession is not represented and GWVS is not calculated. To avoid problems,
then, AGWS should be set to 1.0 inch and GWVS to 0.0 for initial simulation runs.
If the simulation period is limited in duration, you can check and reset these state variables to
values observed for the same period in subsequent years with similar climatic conditions.
However, if major calibration changes are made to the parameters controlling these storages (e.g.
UZSN, LZSN, INFILT), then the initial conditions should be checked and adjusted during the
calibration process. The values for AGWS and GWVS should be checked and adjusted as noted
above, which assuming a yearly cycle of groundwater storage, can be compared to values during
similar seasons in the simulation period. If the initial simulated baseflow (before the first
significant rainfall) is much different from the initial observed streamflow, then further adjustments
can be made to raise or lower the flow rates.
Page 19 of 32
-------
Impervious Land Hydrology (IWATER) Parameters
IWAT-PARM1 Table:
The IWAT-PARM1 table includes a number of flag variables to indicate either the selection of a
simulation algorithm option, or whether the parameter will be treated as a constant or be varied
monthly. As with PWAT-PARM1, where flags indicate monthly variability, corresponding
monthly values must be provided in Monthly Input Parameter tables (see below following IWAT-
PARM3 section).
CSNOFG Flag to use snow simulation data; must be checked (CSNOFG=1) if SNOW
module is run.
RTOPFG Flag to select overland flow routing method. If RTOPFG=0, a new routing
algorithm is used. RTOPFG=1 results in the use of the method used by
predecessor models (HSPX, ARM, and NFS). Recommendation: setRTOPFG=l;
this method is more commonly used and has been subjected to more widespread
application.
VRSFG Flag to select constant or monthly-variable retention storage capacity, RETSC.
Monthly values are not often used.
VNNFG Flag to select constant or monthly-variable Manning's n for overland flow plane,
NSUR. Monthly values are not often used.
RTLIFG Flag to determine if lateral surface inflow to the impervious land segment will be
subject to retention storage (RTLIFG=1). This flag only has an impact if the
another land segment drains to the impervious land segment; otherwise lateral
surface inflow is nonexistent. This feature is not commonly used in most HSPF
applications.
IWAT-PARM2 Table:
LSTJR Length of assumed overland flow plane (feet), (measure/estimate). See PWAT-
PARM2/ LSUR discussion. For impervious areas, LSUR reflects the overland
flow length on directly connected, or effective impervious area (EIA), and is
usually in the range of 50 to 150 feet, although longer lengths may apply in
commercial or industrial regions of large metropolitan areas. Impervious surfaces
that drain to pervious land, rather than to a reach, are considered part of the
pervious land segment and not part of the EIA.
SLSTJR Average slope of the assumed overland flow path (unitless), (measure/estimate).
See PWAT-PARM2 / SLSUR discussion.
Page 20 of 32
-------
NSUR Manning's n for overland flow plane (estimate). See PWAT-PARM4 / NSUR
discussion. Recommendation: set NSUR within the range of 0.05 to 0.10 for
paved roads and parking lots.
RETSC Retention (interception) storage of the impervious surface (inches) (estimate).
RETSC is the impervious equivalent to the interception storage variable (CEPSC)
used for pervious land segments. RETSC is the depth of water that collects on the
impervious surface before any runoff occurs. A study of five urban watersheds in
the Puget Sound region conducted by the U.S. Geological Survey (Dinicola, 1990)
found that a value of 0.10 for RETSC was appropriate. If parking lots and
rooftops are designed for detention storage, larger values up to 0.5 inches may be
reasonable.
IWAT-PARM3 Table:
The following two parameters are used only if SNOW is being simulated.
PETMAX Temperature below which ET will be reduced by 50% of that in the input time
series (degree F), (estimate, then calibrate). See PWAT-PARM3 /PETMAX
discussion.
PETMIN Temperature at and below which ET will be set to zero (degree F), (estimate, then
calibrate). See PWAT-PARM3 /PETMIN discussion.
Page 21 of 32
-------
Monthly Input Parameter Tables:
MON-RETN Table:
Monthly values for retention storage. Monthly values can be varied to represent seasonal changes
in surface retention storage due to litter accumulation or sediment deposition on the impervious
surface. Monthly values are not often used.
MON-MANNING Table:
Monthly values for Manning's n for the overland flow plane. As described above for MON-
RETN, monthly values can be changed to represent seasonal changes on the surface of the
impervious area. Monthly values are not often used.
IWAT-STATE1 Table:
RETS and SURS are initial values for storage of water in retention and surface ponding,
respectively. Both of these storages pertain to the first day of the simulation period. RETS and
SURS are highly dynamic and are only non-zero if the simulation starts during or just following a
storm event. They can be left blank or set to zero unless an individual storm is being simulated.
Page 22 of 32
-------
Flow Routine (HYDR andADCALC) Parameters
HSPF computes streamflow through a stream reach or reservoir based on two assumptions: (1)
there is a fixed relationship between depth, volume, and discharge, and (2) discharge is a function
of volume (see ODGTFG discussion for exception). This means that flow reversals and
backwater effects in an upstream reach are not simulated. Routing is computed using storage
routing or kinematic wave routing. Momentum is not considered in the routing computations.
HYDR-PARM1 Table:
The HYDR-PARM1 table includes flags to indicate which auxiliary variables to compute and
include in the RCHRES printout as well as flags to select routing options based on either volume,
time, or both. The auxiliary variables can be output to the WDM and statistically analyzed like
other time series.
VCONFG Flag to select constant or monthly-variable factors to adjust the outflow from the
FTABLE discharge column, i.e. use the MON-CONVF table. Monthly values are
not often used.
AUX1FG Flag to compute stream channel depth, stage, surface area, average depth
(volume/surface area), top width (surface area/length), and hydraulic radius.
AUX1FG must be set to 1 if precipitation and evaporation fluxes are calculated for
the stream reach, or water quality is simulated for the reach.
AUX2FG Flag to compute average cross section area (volume/length) and average velocity
(discharge/average cross sectional area). AUX2FG must be set to 1 if oxygen in
the stream reach is simulated. If AUX2FG=1 then AUX1FG must equal 1.
AUX3FG Flag to compute bed shear velocity and bed shear stress. These values are used to
calculate sediment deposition and scour (inorganic and organic) for a stream reach.
If AUX3FG=1 then AUX2FG and AUX1FG must equal 1.
ODFVFG For use when the stream reach outflow demand is a function of volume. The value
is the column in the appropriate FTABLE that contains stream reach discharge
values. A maximum of five discharge exits can be specified for a single stream
reach. For a stream reach with a single exit, the appropriate value of ODFVFG(1)
is usually 4, since the FTABLE includes columns for depth, surface area, and
volume in columns 1, 2, and 3, respectively. With two exits (e.g. natural exit plus
a side diversion) ODFVFG(1)=4 and ODFVFG(2)=5.
If the value of ODFVFG is less than zero, then the absolute value is the column in
the COLIND time series which designates which FTABLE column to use. The
COLIND time series is used to vary the FTABLE column used in the computation
Page 23 of 32
-------
of discharge on a seasonal or daily basis. For example, if ODFVFG=-1 then the
model checks the appropriate COLIND value for that time step and if the
COLIND value equals 6 then column 6 in the FTABLE is used in the outflow
calculations for that particular stream reach. In addition, a fractional value for
COLIND can be used to interpolate between columns in the FTABLE, e.g. a value
of 4.3 can be used to interpolate between columns 4 and 5, with 30% of the
difference between columns 4 and 5 being added to column 4. This option can be
used to simulate different rule curves and to vary from one set to another.
ODGTFG Flag to specify that the stream reach outflow demand is a function of time. The
value is the number corresponding to the OUTDGT subscript number specified in
the External Sources Block. The OUTDGT number links the reach outflow
demand to a flow time series data set in the WDM. A maximum of five discharge
exits can be specified for a single stream reach. For a stream reach with a single
exit, the appropriate value of ODGTFG(l) is usually 1 (the corresponding
operation number of OUTDGT in External Sources will then be 1). With two
exits (e.g. natural exit plus a side diversion) ODFVFG(1)=1 and ODFVFG(2)=2,
corresponding to subscripts 1 and 2 of OUTDGT.
FUNCT Flag to combine ODFVFG and ODGTFG functions, if appropriate. For each of a
maximum of 5 possible exits the value of FUNCT can be 1 (select the smaller of
the ODFVFG and ODGTFG values), 2 (larger of ODFVFG and ODGTFG), or 3
(sum of ODFVFG and ODGTFG).
HYDR-PARM2 Table:
FTBDSN
FTABNO
LEN
When the FTABLE is stored in the WDM file, this is the Data Set Number (DSN)
in the WDM file for that reach's FTABLE. When the FTABLE is stored in the
FTABLES block of the UCI file, FTBDSN is set to zero.
When FTBDSN is zero, this is the Id number of the FTABLE as included in the
FTABLES block of the UCI file. When FTBDSN > 0, this is the indicator to
identify a particular FTABLE data set in the FTBDSN DSN.
Length of the stream reach (miles), (measure). Length is used in the computation
of auxiliary 1, 2, and 3 parameter values.
Use of BASINS Data/Tools:
This is populated automatically by BASINS during model initialization.
Page 24 of 32
-------
DELTH Change in elevation from the upstream end of the stream reach to the downstream
end (feet), (measure). DELTH is used if reaeration is computed using the
Tsivoglou-Wallace equation in the OXRX Block or if sandload transport capacity
is computed using either the Toffaleti or Colby method in the SEDTRN Block.
Use of BASINS DatafTooh:
This is populated automatically by BASINS during model initialization for RF1
reaches. The RF3 data coverage, however, does not contain the top and bottom
elevation data held in the RF1 (i.e. Ptopele, Pbotele). Instead use the identify tool
on the DEM coverage to obtain the elevation at the top and bottom of each reach
segment. DELTH is the reach top elevation minus the reach bottom elevation
contained in the RF1.
STCOR
KS
Stage correction to convert RCHRES depth (DEP) to an equivalent stage or
elevation (feet), (measure). STAGE (elevation) = DEP (depth) + STCOR. This
parameter is relevant only if you want to compare reach depth reported as stage to
modeled reach stage.
Weighting factor for hydraulic routing (unitless), (initialize with reported value).
KS is a weighting factor applied in the computation of the stream reach outflow.
The outflow for any given time step is KS times the outflow at the beginning of the
time step plus the complement (1-KS) times the outflow at the end of the time
step. Increasing KS from 0.0 to 1.0 increases the likelihood of model instability;
the value 0.5 will produce the most accurate results. Recommendation: set
KS=0.50.
DB50
Median diameter of the bed sediment (inches), (estimate/measure). DB50 is used
to calculate (1) the bed shear stress if the stream reach is a lake, and (2) the rate of
sand transport if the Toffaleti or Colby method is used. Note: DB50 is not
connected with the sand particle diameter (D) input in the SAND-PM table in the
SEDTRN Block.
Page 25 of 32
-------
Monthly Input Parameter Table:
MON-CONVF Table:
Monthly values for the FTABLE discharge adjustment factors are not often used. Monthly values
can be varied to represent seasonal changes in stream channel and flood plain volume-based
discharge due to vegetation growth or other alterations.
HYDR-INIT Table:
VOL is the initial volume of water in the stream channel. For small streams VOL can be set to
zero to represent a dry channel. Rivers and lakes should start with an initial volume appropriate
for the time of year the simulation begins, or should be based on stream volume predicted by the
model for previous runs for long time periods leading up to the current simulation starting period.
COLIND is the initial value for the FTABLE discharge exit column and should be set to the same
value as ODFVFG for the exit (usually 4). OUTDGT is the initial value for the time series
outflow demand, if ODGTFG is set to one, and has units of cubic feet per second.
ADCALC-DATA Table:
ADCALC calculates values for variables which are necessary to simulate longitudinal advection of
dissolved or entrained constituents. These values are dependent on the volume and outflow
values computed by the hydraulics section (HYDR) of the model.
CRRAT Ratio of the maximum velocity to the mean velocity in the stream channel cross
section under typical flow conditions (unitless), (estimate/measure). CRRAT
must be 1.0 or greater, where a value of 1.0 corresponds to completely uniform
velocity (plug flow) across the stream channel. CRRAT is used to determine the
relative volumes of water stored in the stream reach versus that leaving the reach,
in a given time step. If CRRAT is greater than the volume:outflow ratio then the
outflow is assumed to be in part, made up of water that entered the reach in that
same interval. The inflow constituent concentration, then alters the outflow
constituent concentration.
VOL Volume of water in the stream reach at start of simulation (acre-feet) (estimate).
VOL is only used if section FIYDR is inactive, and thus FIYDR-INIT/VOL is not
used.
Page 26 of 32
-------
REFERENCES
Bicknell, B.R., J.C. Imhoff, J.L. Kittle Jr., A.S. Donigian, Jr. and R.C. Johanson. 1997.
Hydrological Simulation Program — FORTRAN, User's Manual for Version 11.
EPA/600/R-97/080. U.S. EPA, National Exposure Research Laboratory, Athens, GA.
Gasman, E. 1989. Effects of Agricultural Practices on Water Quality as Related to Adjustments
of HSPF Parameters, A Literature Review. ICPRB Report No. 89-6. Interstate
Commission on the Potomac River Basin in cooperation with the Maryland Department of
the Environment, Baltimore, MD. 255 p.
Chen, Y.D., S.C. McCutcheon, R.F. Carsel, A.S. Donigian, J.R. Cannell, J.P. Craig. 1995.
Validation of HSPF for the water balance simulation of the Upper Grande Ronde
watershed, Oregon, USA. Man's Influence on Freshwater Ecosystems and Water Use
(Proceedings of a Boulder Symposium), July, 1995. IAHS Publ. No. 230, 1995.
Crawford, N.H. 1999. Hydrologic Journal - Snowmelt Calibration. Hydrocomp, Inc.
'www.hydrocomp.com'.
Dinicola, R.S. 1990. Characterization and Simulation of Rainfall-Runoff Relations for
Headwater Basins in Western King and Snohomish Counties, Washington. U.S.
Geological Survey. Water-Resources Investigations Report 89-4052. Tacoma, WA. 52
pp.
Donigian, A.S., Jr. 1998. Personal communication, 1998.
Donigian, A.S., Jr. and N.H. Crawford. 1976. Modeling Pesticides and Nutrients on Agricultural
Lands. Environmental Research Laboratory. Environmental Protection Agency. Athens,
GA. EPA 600/2-76-043. 263pp.
Donigian, A.S., Jr. and H.H. Davis, Jr. 1978. User's Manual for Agricultural Runoff
Management (ARM) Model, U.S. Environmental Protection Agency, EPA- 600/3-78-080.
Donigian, A.S. Jr., Baker, D.A. Haith and M.F. Walter. 1983. HSPF Parameter Adjustments to
Evaluate the Effects of Agricultural Best Management Practices, EPA Contract No.
68-03-2895, U.S. EPA Environmental Research Laboratory, Athens, GA,
(PB-83-247171).
Doty, Stephen, 2000. Ex-oficio, American Association of State Climatologist (AASC). National
Climatic Data Center. Asheville, NC. Personal communication, May, 2000.
Engman, Edwin T., 1986. Roughness Coefficients for Routing Surface Runoff. Journal of
Irrigation and Drainage Engineering, Vol 112, No.l, Feb., 1986.
Page 27 of 32
-------
Fontaine, T.A., and V.M.F. Jacomino. 1997. Sensitivity Analysis of Simulated Contaminated
Sediment Transport, Journal of the American Water Resources Association. Vol. 33, No.
2, April, 1997.
Freeze, R. A. and J. A. Cherry. 1979. Groundwater. Prentice Hall, Englewood Cliffs, NJ, 1979.
Hwang, N.H.C. and C.E. Hita. 1987. Fundamentals of Hydraulic Engineering Systems. 2nd
Edition. Prentice-Hall, Inc. Houston, 1987.
Laroche, A., J. Gallichaud, R. Lagace, and A. Pesant. 1996. Simulating Atrazine Transport with
HSPF in an Agricultural Watershed. ASCE Journal of Environmental Engineering, July,
1996.
L. Larson and Peck, E.L., 1974. Accuracy of Precipitation Measurements for Hydrologic
Modeling. Water Resources Research: 857-862.
Mays, Larry W., 1999. Hydraulic Design Handbook. McGraw-Hill, New York, 1999.
Soil Survey Staff, 1996. Natural Resources Conservation Service, National Soil Survey
Handbook, title 430-VI (Washington, D.C., U.S. Government Printing Office, September
1996) available on the web at: http://www.statlab.iastate.edu/soils/nssh/.
US EPA, 1999. HSPFParm: An Interactive Database of HSPF Model Parameters, Version 1.0.
EPA-823-R-99-004. U.S. Environmental Protection Agency, Office of Water,
Washington, DC. Available from the BASINS web site,
http://www.epa.gov/ost/basins/support.htm.
Viessman, W., G.L. Lewis, and J.W. Knapp. 1989. Introduction to Hydrology. Third Edition.
Harper and Row, New York. 1989.
Weltz, Mark A., A. B. Arslan, L.J. Lane, 1992. Hydraulic Roughness Coefficients for Native
Rangelands. Journal of Irrigation and Drainage Engineering, Vol 118, No. 5, Sept/Oct,
1992.
Page 28 of 32
-------
HSPF ATEMP AND SNOW PARAMETERS AND VALUE RANGES
NAME
DEFINITION
UNITS
RANGE OF VALUES
TYPICAL
MIN
MAX
POSSIBLE
MIN
MAX
FUNCTION OF...
COMMENT
ATEMP - DAT
ELDAT
AIRTMP
Weather station/ watershed elevation diff.
Initial air temperature
feet
deg. F
-1000
30.0
1000
70.0
none
0.0
none
90.0
Topography, station location
Climate
Only used in air temperature data calculations
Only used in air temperature data calculations
SNOW - PARM1
LAT
MELEV
SHADE
SNOWCF
COVIND
Latitude of watershed segment
Mean elevation of watershed segment
Fraction shaded from solar radiation
Snow gage catch correction factor
Snowfall required to fully cover surface
degrees
feet
none
none
inches
30.0
50.0
0.1
1.1
1.0
50.0
3000
0.5
1.5
3.0
-90.0
0.0
0.0
1.0
0.1
90.0
7000
0.8
2.0
10.0
Location
Topography
Forest cover, topography
Gage type, characteristics, location
Topography, climate
Positive for northern hemisphere
Used in convective heat flux equation
Controls radiation to and from the snowpack
Calibrate to snow depth observations
Higher for mountainous watersheds
SNOW - PARM2
RDCSN
TSNOW
SNOEVP
CCFACT
MWATER
MGMELT
Density of new snow
Temperature at which precip becomes snow
Snow evaporation factor
Condensation/convection melt factor
Liquid water storage capacity in snowpack
Ground heat daily melt rate
none
deg. F
none
none
in/in
in/day
0.10
31.0
0.10
1.0
0.01
0.01
0.20
33.0
0.15
2.0
0.05
0.03
0.05
30.0
0.0
0.5
0.005
0.0
0.30
40.0
0.5
8.0
0.2
0.1
Climate, air temperature
Climate, topography
Climate, topography
Climate
Climate
Climate, geology
Adjust with field snow density data, if available
Precip. is snow when temperature below TSNOW
Only important in windy, low humidity conditions
Calibrate to change rate/timing of snowmelt
Adjust to change timing of snowmelt
Usually small under frozen ground conditions
Page 29 of 32
-------
HSPF HYDROLOGY PARAMETERS AND VALUE RANGES
NAME
DEFINITION
UNITS
RANGE OF VALUES
TYPICAL
MIN
MAX
POSSIBLE
MIN
MAX
FUNCTION OF...
COMMENT
PWAT - PARM2
FOREST
LZSN
INFILT
LSUR
SLSUR
KVARY
AGWRC
Fraction forest cover
Lower Zone Nominal Soil Moisture Storage
Index to Infiltration Capacity
Length of overland flow
Slope of overland flow plane
Variable groundwater recession
Base groundwater recession
none
inches
in/hr
feet
ft/ft
1 /inches
none
0.0
3.0
0.01
200
0.01
0.0
0.92
0.50
8.0
0.25
500
0.15
3.0
0.99
0.0
2.0
0.001
100
0.001
0.0
0.85
0.95
15.0
0.50
700
0.30
5.0
0.999
Forest cover
Soils, climate
Soils, land use
Topography
Topography
Baseflow recession variation
Baseflow recession
Only impact when SNOW is active
Calibration
Calibration, divides surface and subsurface flow
Estimate from high resolution topo maps or CIS
Estimate from high resolution topo maps or GIS
Used when recession rate varies with GW levels
Calibration
PWAT - PARM3
PETMAX
PETMIN
INFEXP
INFILD
DEEPFR
BASETP
AGWETP
Temp below which ET is reduced
Temp below which ET is set to zero
Exponent in infiltration equation
Ratio of max/mean infiltration capacities
Fraction of GW inflow to deep recharge
Fraction of remaining ET from baseflow
Fraction of remaining ET from active GW
deg. F
deg. F
none
none
none
none
none
35.0
30.0
2.0
2.0
0.0
0.0
0.0
45.0
35.0
2.0
2.0
0.20
0.05
0.05
32.0
30.0
1.0
1.0
0.0
0.0
0.0
48.0
40.0
3.0
3.0
0.50
0.20
0.20
Climate, vegetation
Climate, vegetation
Soils variability
Soils variability
Geology, GW recharge
Riparian vegetation
Marsh/wetlands extent
Reduces ET near freezing, when SNOW is active
Reduces ET near freezing, when SNOW is active
Usually default to 2.0
Usually default to 2.0
Accounts for subsurface losses
Direct ET from riparian vegetation
Direct ET from shallow GW
PWAT - PARM4
CEPSC
UZSN
NSUR
INTFW
IRC
LZETP
Interception storage capacity
Upper zone nominal soil moisture storage
Manning's n (roughness) for overland flow
Interflow inflow parameter
Interflow recession parameter
Lower zone ET parameter
inches
inches
none
none
none
none
0.03
0.10
0.15
1.0
0.5
0.2
0.20
1.0
0.35
3.0
0.7
0.7
0.01
0.05
0.05
1.0
0.3
0.1
0.40
2.0
0.50
10.0
0.85
0.9
Vegetation type/density, land use
Surface soil conditions, land use
Surface conditions, residue, etc.
Soils, topography, land use
Soils, topography, land use
Vegetation type/density, root depth
Monthly values usually used
Accounts for near surface retention
Monthly values often used for croplands
Calibration, based on hydrograph separation
Often start with a value of 0.7, and then adjust
Calibration
Page 30 of 32
-------
HSPF IMPERVIOUS HYDROLOGY PARAMETERS AND VALUE RANGES
NAME
DEFINITION
UNITS
RANGE OF VALUES
TYPICAL
MIN
MAX
POSSIBLE
MIN
MAX
FUNCTION OF...
COMMENT
IWAT - PARM2
LSUR
SLSUR
NSUR
RETSC
Length of overland flow
Slope of overland flow plane
Manning's n (roughness) for overland flow
Retention storage capacity
feet
ft/ft
none
inches
50
0.01
0.03
0.03
150
0.05
0.10
0.10
50
0.001
0.01
0.01
250
0.15
0.15
0.30
Topography, drainage system
Topography, drainage
Impervious surface conditions
Impervious surface conditions
Estimate from maps, GIS, or field survey
Estimate from maps, GIS, or field survey
Typical range is 0.05 to 0.10 for roads/parking lots
Typical range is 0.03 to 0.10 for roads/parking lots
IWAT - PARM3
PETMAX
PETMIN
Temp below which ET is reduced by half
Temp below which ET is set to zero
deg. F
deg. F
35.0
30.0
45.0
35.0
32.0
30.0
48.0
40.0
Climate, vegetation
Climate, vegetation
Reduces ET near freezing, when SNOW is active
Reduces ET near freezing, when SNOW is active
Page 31 of 32
-------
HSPF HYDRAULIC PARAMETERS AND VALUE RANGES
NAME
DEFINITION
UNITS
RANGE OF VALUES
TYPICAL
MIN
MAX
POSSIBLE
MIN
MAX
FUNCTION OF...
COMMENT
HYDR - PARM2
FTBDSN
FTABNO
LEN
DELTH
STCOR
KS
DB50
WDM data set number for FTABLE
FTABLE number in UCI file
Stream reach (RCHRES) length
Stream reach length change in elevation
Stage correction factor
Routing weighting factor
Bed sediment diameter
none
none
miles
feet
feet
none
inches
none
none
0.1
10
0.0
0.0
0.01
none
none
1.0
100
none
0.5
0.02
1
1
0.01
0.1
0.0
0.0
0.001
999
999
100
1000
none
0.99
1.00
WDM File
RCHRES block/ reach numbering
Topography, stream morphology
Topography, stream morphology
Topography
Channel slope, flow obstructions
Channel bed properties
Used only if FTABLE is in WDM file
Used only if FTABLE is in UCI file
Used only in computing auxiliary parameters
Used only for water quality and sediment
Dependent on elevation datum used
Use KS = 0.5
Used only in sediment calculations
ADCALC - DATA
CRRAT
VOL
Ratio of maximum to mean flow velocity
Initial stream channel water volume
none
acre-feet
1.5
0.0
2.0
none
1.0
0.0
3.5
none
Climate, vegetation
Season, channel geometry, climate
Only used with water quality
Initial volume in reach channel
Page 32 of 32
------- |