Laserfiche WebLink
<br />absent than present. The USGS surveys of the Harvard Gulch drainage basin <br />from maps and aerial photographs found the effective imperviousness to be <br />between 27 and 31 percent. It is quite likely that these statistics, based on <br />detailed observation, would be assumed to be correct by a hydrologist <br />developing design criteria. Based upon the calibrated results, however, an <br />effective imperviousness of 40 percent was determined to be most accurate. <br />While this discrepancy may not seem to be large, the effects are significant. <br />Graph 10 shows what the predicted hydrograph looks like with 30 percent <br />imperviousness compared to the observed hydrograph for the August 14th storm. <br />The graph illustrates how relatively small changes in imperviousness can have a <br />large impact on results (and associated error). <br />While it is true that the UDFCD Drainage Criteria Manual recommends <br />imperviousness values of between 36 and 43 percent for medium density <br />residential neighborhoods, these values are based upon a marginally correlating <br />generalized rating curve generated from only seven data points (Urban Drainage <br />Criteria Manual, 1984). I don't believe it was the intent of the UDFCD to imply <br />that these values should be used in place of more accurate surveys of a specific <br />site. What this implies is that the CUHP program has a low bias. The percent <br />imperviousness for Harvard Gulch, based upon the calibration, may be artificially <br />inflated. When using CUHP, this investigation found the program to have a <br />propensity to underestimate rather than overestimate. Use of the CUHP program <br />without calibration may produce drastically underestimated peak flow values. <br />With the advent of today's urban sprawl, especially in the Denver <br />metropolitan area, hydrologic engineers are most likely to find they are <br />developing storm water systems in outlying areas. In these cases, hydrologists <br />are already faced with the task of anticipating future growth. Considering the low <br />bias of CUHP, design engineers would be well advised to generously <br />overestimate effective imperviousness with their forecast modeling. <br /> <br />, <br />, <br /> <br />Application and Evaluation of CUHP <br /> <br />Page 25 of 52 <br />