EIS Arapahoe & Roosevelt National Forest, Pawnee National Grassland

Home Browse Search

ATTACHMENT 1 > DRAFT Methods for estimating water yields for Forest Plan Carl Chambers June 4, 1997 This paper outlines methods for estimating water yield for Forest Plan revisions. It contains assumptioiis made for data input into the WRENSS model. While WRENSS is still the state-of- art for this water yield modeling, it was designed as a site specific model, and the assumptions are designed to generalize it for use in (non-site specitic) Forest Plan water yield modeling. Comments have been provided on modellinp methods by Chuck Troendle and Jim Maxwell and their comments have been incorporated into this document. There are several versions of the WRENSS model. I suggest the use of the latest version (ver. 1.1) of the model developed by Robert Swanston, formerly of the Canadian Forestry Service (the Canadian model). While modifications and updates have been provided to the RMS version of the model to retlect the advancing state of the science, data enu-y and editing is much more difficult if you are not intimately familiar with the software. The Canadian model has an easy to use interface. Some modifications to the original WRENSS procedure have also been made to the Canadian Model. However, documentation provided with the model indicates that if Wind Speed is set to , "zero", number of days with no snow is set to "zero", lapse rate is set to "one", a11d S110W ? scouring is set to "yes", the model will provide the same results as the original WRENSS - procedure. Inputs for wind speed, numbers of days with no snow, and snow scour are all designed to modify the original procedure with regard to snow loss and redistribution. Chuck Troendle pointed out that while there is nothing wrong conceptually with adjusting snow scour factors, the WRENSS procedure inherently includes snow scour conditions that were present for the data sets used to develop the model, and that before these variables are added to the model, the original nomographs or their digital versions would need to be moditied to zero out the effects of the assumed conditions. He recommends setting the model to duplicate the original procedure. FVS (Forest Vegetation Simulation) is the model used to provide tree growth and yield information for FORPLAN. It is the standard model used nationally by the Forest Service. It can provide the vegetation information needed for input into WRENSS, including maximum basal area, pre- and post-treatment basal area, and tree height. I propose to use the results of FVS to provide that input for WRENSS. Using FVS results ensures that we are working from the same basic set of assumptions as other resources in Forest Plan analysis. I suggest that we only model the effects of clearcutting and estimate the effects of partial cutting by proportioning the water yield from clearcutting by the percent of BA removed by the partial cut (e.g. a partial cut removing 30% of the basal area will produce 30% of the water yield of a clear cut.), as suggested by Chuck Troendle. The original WRENSS procedure indicated that the relationship between basal area removed and water yield was not 1:1 (e.g. a partial cut removing / 4