Operational Requirements
The primary application of this erosion prediction technology
will be by field personnel who will be using the procedure
as a tool
to assist land managers in making soil conservation decisions.
Major factors important to these users are:
(a) computational time,
(b) ease of use,
(c)
applicability to the broad range of conditions typically encountered
in field programs,
(d) robustness,
(e) validity,
and
(f)
ease of explanation to client.
Computational Time
Developers of the procedure are to strive for computational efficiency
and to have the procedure operate as quickly as possible.
The procedure is to compute the frequency
distribution of annual soil loss values
for the profile version at the rate of
one management practice per minute
and one practice per two
minutes for
the watershed version running
a single overland flow profile
and a single concentrated flow channel.
The rate can be proportionally slower for more complex systems.
Also,
not more than 30 minutes per farm of actual user time
(computer time can be longer)
is to be required in the office
to prepare and assemble needed information before going to the field.
Once in the field,
no more time can be used to collect and assemble input information
than would be required for the USLE
when the profile version is used.
The criteria to be used by the developers for judging the acceptability
of an internal simplification in the procedure
are:
(a) does the planning or assessment decision change --
if not,
use the procedure that requires the least resources -- and
do computed values for the primary output variable change
more than 10 percent -- if they do,
is the change of consequence?
Ease of Use
The procedure shall be easy to use,
especially for the infrequent user,
by accepting simple inputs that are commonly available
and understood by personnel in the local field office.
It should require little structured training or
support.
Also,
it shall be flexible and accept inputs
on
increasing detailed
and complex levels if the user
determines that more
detail is needed or that default values need changing.
The user shall not have to directly manipulate
any mathematical equations to use the procedure;
all mathematical manipulations shall be done by a computer program.
The procedure is to be designed so that a maximum amount of computations can be
made once-and-for-all and
stored for repeated use.
Likewise,
the procedure shall be constructed so that
data files specific to a given local area
(a county or subcounty unit)
can be prepared and stored locally
so that the field office user only has to search and
retrieve minimal data with each application.
In so far as possible,
the procedure shall use data
and data files used in other NRCS,
BLM, and FS applications,
and it should be compatible with
Geographic Information Systems (GIS).
The output should display in an easily understood
form the consequence of alternative management options.
Applicable to Broad Range of Conditions
The procedure,
within the limits defined by this document,
must apply to all sheet-rill erosion problems
that local field office personnel encounter.
In particular,
it must apply to all conditions covered by the USLE
plus additional ones.
Similarly,
this requirement also applies to all concentrated
flow erosion
situations,
but the procedure is not
for
hydraulic design of waterways.
Robustness
The procedure must tolerate
out-of-range
input data
and
combinations of inputs that might cause problems.
The procedure should use asymptotic, "well behaved" functions
to avoid extremely incorrect values
and the procedure unexpectedly
"blowing up."
See the Appendix 7 for an example of a function that adds
robustness to model.
It must tolerate applications
for which it is not intended.
However,
the procedure must alert the user to these excesses,
alert for loss of accuracy when inputs are
over simplified such as for slope shape,
and check for incorrect data entries.
The procedure should alert the user
to the possibility of
obtaining
"additional" information with more detailed inputs.
Validity
The procedure must be sufficiently accurate to lead to the
planning and assessment decision
that would be made in the large majority of cases
when
full information is available.
However,
more than accuracy is to be considered in
establishing the validity of the procedure.
The procedure is to be validated, and the
validation process and its results
are to be documented.
The prediction procedure is expected to be composed
of a number of modules.
Each major module is to be individually validated,
and the procedure is to be validated as a package.
Validation Criteria
Validation is to be based on the procedure meeting all of the following criteria.
(a) The model is valid if it serves its intended purpose
as defined by these specific User Requirements.
(b) The model is based on scientific principles
and represents a reasonable expression of current scientific understanding of
erosion processes that
can be used in an applied procedure.
(c) The procedure gives expected responses that appear reasonable.
For example,
the output varies qualitatively
with ground cover
(or any variable or combination of variables)
in the way that is commonly accepted by erosion experts.
(d) The model gives results that are
more useful for agency program objectives
than those given by the USLE
and applies to situations not appropriate for the USLE.
These situations include deposition in furrows,
especially as influenced by plant residue
in the furrows;
nonuniform distribution of cover between ridges and furrows;
the acceleration of rill erosion above a critical steepness;
the variation in slope length, slope steepness,
ground cover,
and contouring relationships with climate,
soil,
topography,
and land use;
erosion by surface irrigation;
deposition on concave slopes;
and concentrated flow erosion.
(e) The model provides a reasonable representation of data covering
the range of conditions of the
"key"
situations described above.
(f)
Judgements on the "goodness of fit"
of the estimates from the procedure to observed data are to be based on the data sets as a whole
and not on a few specific and isolated data sets.
Quantitative measures of the "goodness of fits" will be calculated and presented,
but a specific quantitative level of accuracy figure is
not being required
because of the great variation in the experimental data that will be used in validation.
However,
the results are to be at least as good
with respect to observed data
and known relationships as those from the USLE.
(g)
The model is able to
"stand up"
in public hearings of management plans
and assessments.
Ease of Explanation
The procedure is to be based on a set
of principles and concepts that can be explained by
local field personnel to the client.
The procedure is to be developed so
that the user can easily demonstrate how
the major factors of climate,
soil, topography,
and land use affect erosion.
Any Comments Would Be Appreciated
Disclaimer and Copyright Information