|
<br />Bottom width, side slopes, and main channel depths were estimated using the best available mapping and
<br />field observations, Except in reaches characterized by broad, relatively flat swales, peak flows could be
<br />expected to overtop the main channel depth, For this reason, secondary overflow channels were defined,
<br />These overflow channels, which use the same variables as the main channels, were assumed to have a
<br />bottom width equal to the top width of the main channel at its full depth, side slopes equal to 50: 1, and
<br />depths of 20 feet.
<br />
<br />Invert longitudinal slope and reach lengths were developed based on the mapping, The slope was developed
<br />by dividing the difference between the upper and lower elevations by the reach length,
<br />
<br />The Manning's 'n' roughness coefficient was developed using Equation RO-lO from the Criteria Manual:
<br />
<br />n = O,393S038R,o.16, where
<br />n = Manning's roughness coefficient
<br />S = friction slope (ftlft) - assumed to be equal to longitudinal slope
<br />R = hydraulic radius (ft) - assumed to be equal to channel depth
<br />
<br />For reaches defined by streets (where flow is assumed to be diverted and travel down a street), an 'n' value
<br />of 0,0200 was used, For reaches containing a secondary overflow channel, the 'n' value of the overflow
<br />channel was set equal to the 'n' value of the main channel.
<br />
<br />The variables used for the SWMM hydrologic routing model are shown in Table 6 of Appendix B,
<br />
<br />3.8 Regional Reservoirs
<br />
<br />Four reservoirs were included in the Ralston Creek watershed, These include Ralston Reservoir,
<br />Arvada/Blunn Reservoir, Leyden Lake, and Oberon Lake, Storage-discharge parameters, used in the
<br />UDSWMM hydrologic routing model were provided by the City of Arvada for Ralston and Arvada/Blunn
<br />Reservoirs and Leyden Lake, UDFCD provided stage-storage information for Oberon Lake, Stage-storage
<br />curves used in the models are provided in Table 7 of Appendix B,
<br />
<br />For the initial calibration model, Arvada/Blunn Reservoir, Leyden Lake, and Hays Lake, were removed as
<br />storage elements from the model. They were replaced with I-foot direct conveyance links,
<br />
<br />3.9 Peak Flows
<br />
<br />Peak flows were produced by analyzing the watershed for each rainfall scenario and each design frequency,
<br />In order for the model to be considered consistent with past published results, the peak flows at critical
<br />design points must be within 10% of those values reported in past studies, In order to make the values
<br />consistent, the following procedures were performed:
<br />
<br />1. For all reaches basinwide for which Manning's 'n' values greater than 0,07 were computed, a
<br />reduced value between 0,06 and 0,07 was substituted, For example, if a value of 0,096 or 0,106 was
<br />computed, it was replaced with 0,066, The assumption was that despite the results of Equation RO-
<br />10, a value higher than 0,07 was not consistent with channels in this basin, The Manning's 'n'
<br />
<br />coefficient was also judged to be the most justifiably altered physical parameter for calibration
<br />purposes,
<br />2, Leyden Creek, which was initially modeled as a broad, relatively flat floodplain, was replaced with a
<br />more-defined channel.
<br />3, The Manning's 'n' values of the Leyden Creek mainstem were replaced with a value of 0,040 below
<br />Leyden Lake, This was needed to speed up the response of the Leyden Creek watershed to better
<br />match past published results,
<br />4, The Manning's 'n' value for two upper reaches of Ralston Creek was increased to 0,070, This was
<br />needed to slow down the response of the Ralston Creek watershed to better match past results,
<br />
<br />After each of the above procedures was executed, peak flow results were within 10% of previously
<br />published values, The model was then considered to be calibrated, A separate model was then created
<br />which added in Arvada/Blunn Reservoir, Leyden Lake, and Hays Lake,
<br />
<br />3.10 Results
<br />
<br />The results of the models are summarized in Table 8 of Appendix B, which lists the peak flow rates at
<br />selected design points, Results are listed both including and excluding the above-mentioned reservoirs, In
<br />addition, a comparison to previous published results is given for the IOO-year and lO-year events for the
<br />model, which excluded the reservoirs, The previously published PIS study for the City of Arvada did not
<br />include the Arvada/Blunn Reservoir or Leyden Lake, Because inclusion of the reservoirs represents a
<br />changed condition from that previously modeled, it is inappropriate to compare results for this scenario, No
<br />previously modeled results are available for the 50-year and 500-year events,
<br />
<br />The new basin model was calibrated such that all 100-year values for both Ralston and Leyden Creek were
<br />within 10% of previously reported values at all design points when the Arvada/Blunn Reservoir and Leyden
<br />Lake were excluded, Because van Bibber Creek was not included in this study for anything more than
<br />contributions to Ralston Creek, it was not intended for it to be calibrated; however, it too was within 10% of
<br />the previously recorded values,
<br />
<br />The results, with comparisons to the FIS, are fully summarized in Table 8, Some example 100-year values
<br />from the new model are 5,262 cfs, 7,962 cfs and 11,134 cfs for Ralston Creek at locations below Ralston
<br />Reservoir, below Leyden Creek, and at the mouth, respectively, compared to FIS values of 5,000 cfs, 7,900
<br />cfs, and 11,500 cfs for the same points, On Leyden Creek, the 100-year results for the new model are 4,081
<br />cfs at Leyden Lake and 4,042 cfs at the mouth compared to PIS values of 3,800 cfs and 4,300 cfs at the
<br />I' I
<br />same ocatlOns,
<br />
<br />Because the flow values produced by this new model have been calibrated to within 10% of the previously
<br />published PIS results at all previously reported design points within the watershed, it is recommended that
<br />this new model be accepted as an equivalent to that previously used, Because the new reservoirs are
<br />contributing as flood-control elements, it is recommended that they be added to the model and included in a
<br />revision to the FIS and a LOMR representing revised conditions within the City of Arvada, It is also
<br />recommended that results from this model with the reservoirs included be used in the preparation of a Flood
<br />Hazard Area Delineation representing the flood potential for future developed conditions in the watershed,
<br />
<br />Inclusion of the reservoirs has the effect of lowering peak flowrates al all locations below Arvada/Blunn
<br />Reservoir on Ralston Creek and below Leyden Lake on Leyden Creek, The reduction is most notable just
<br />below the reservoirs with the difference gradually damping further downstream as impacts from lower
<br />
<br />6
<br />
|