1
RENUMA A REGIONAL SCALE NUTRIENT LOADING MODEL
FOR MANAGEMENT Dennis P. Swaney1, Robert W.
Howarth1 and Beth Boyer2 1Cornell University,
Ithaca, NY USA 2SUNY College of Environmental
Science and Forestry, Syracuse, NY USA (email
dps1_at_cornell.edu)
Uncertainty/Sensitivity Analysis Mode
In uncertainty analysis, the user selects which
variables Should be random in the stochastic
worksheet, and specifies their probability
distribution and the desired number of
runsrandom values are generated
The pulldown menu permits the user to choose
between alternate simulation or other options
 ABSTRACT We are developing ReNuMA, an
event-based, regional watershed-scale nutrient
loading model with biogeochemical dynamics that
is readily accessible to watershed and coastal
zone managers. The model, written in Visual
Basic, uses Microsoft Excel as a platform. The
spreadsheet format facilitates model
dissemination to users, and visualization of
model output of simulations of water, N and P
dynamics. The modeling package is based on
hydrologic dynamics of the GWLF model (Haith and
Shoemaker, 1987), and quantifies solid-phase as
well as dissolved nutrient transport processes
when complete, the model will incorporate new
biogeochemical dynamics, including explicit
consideration of atmospheric deposition,
denitrification, and phosphorus saturation. Novel
aspects of the analysis include consideration of
the effect of alternative land-use
trajectories, i.e. categories of land use
change over the period of the simulation, rather
than fixed snapshots of landuse, and model
parameter fitting using the Excel solver feature.
Maximum event loads, of particular significance
to the coastal zone, can be estimated as well as
seasonal and annual loads.
Model outputs can be viewed in the
output worksheets
Parameters can be changed in the parameter
worksheets (yellow)
The user next chooses the type of simulation, and
the model is run repeatedly for each random value
of the combination of parameters selectedAs this
takes some time, the current iteration number is
displayed in a worksheet along with a message
that the analysis is in progress
Standard Simulation Mode Simulating effects of
hypothetical changes in land use and climate
change
A standard mode of using the model is to evaluate
the response of a watershed to hypothetical
scenarios of landuse or climate changeIn this
case, an increase of 2 degrees C and 20 increase
in precipitation over 10 years are considered,
together with a doubling of area of urban and
agricultural landuses
User selects whether to fit streamflow, sediment
or nutrient data. Objective function is sum of
squares of deviations from monthly values.
Model Calibration Mode
When the simulation is complete, the user can
observe the results in the uncertainty
worksheet, in which the mean and standard
deviations of all model variables are reported
for each year simulated
The calibration mode uses the standard Excel
addin package, Solver, to determine best-fit
values of user-selected parameters to
user-specified monthly data.
But both landuse and climatic change affected
nitrogen flux

Climatic change appeared to be more important
than landuse change to ET and streamflow in this
simulation
Solver iterates the selected parameters until it
finds the best solution, then
displays a message And reports the
best-fit value