Laserfiche WebLink
<br />Guidelines and Specifications for Flood Hazard Mapping Partners [April 2003] <br /> <br />Handbook of Hydrology (Maidment, 1993). The reviewing Mapping Partner may use the 68- <br />percent confidence interval, which is analogous to plus or minus one standard error for a normal <br />distribution, to judge the reasonableness of flood discharges derived from regression equations. <br />If the proposed flood discharges generally lie within the 68-percent confidence interval of the <br />gaged data, the reviewing Mapping Partner shall accept these flood discharges as reasonable for <br />the hydraulic analysis. If not, then options for obtaining more reasonable flood discharges shall <br />be provided. <br /> <br />The reviewing Mapping Partner shall use caution in reviewing 1-percent-annual-chance flood <br />discharges derived from regression equations that are significantly different from those derived <br />from gage data. When the regression estimates differ significantly from data from long-term <br />gaging stations and the elevation difference is significant, the regression estimate may be <br />adjusted based on the gaging station data. <br /> <br />C.2.3 <br /> <br />Hydrologic Analysis Based on a Rainfall-Runoff Model <br />[February 2002] <br /> <br />The reviewing Mapping Partner shall first verify that the rainfall-runoff model used by the <br />Mapping Partner performing the hydrologic analysis is included on the FEMA accepted models <br />list, which is posted on the FEMA Flood Hazard Mapping Web site at <br />http://www.fema.gov/fhm/en modl.shtm. The reviewing Mapping Partner shall compare the <br />proposed 1-percent-annual-chance flood discharges from the rainfall-runoff model to the flood <br />discharges from USGS regional regression equations (if they are applicable) and to flood <br />discharges at gaging stations in the vicinity. Procedures for developing estimates from gaging <br />station data and regression equations are discussed in Subsections C.2.1 and C.2.2. <br /> <br />The reviewing Mapping Partner shall plot the flood discharge estimates from these sources <br />against drainage area on logarithmic paper to determine if the proposed flood discharges are <br />reasonable. Plus or minus one standard error bars (68-percent confidence intervals) shall be <br />shown about the regression and gaging station estimates. The USGS regional flood reports <br />typically provide the standard error of prediction or estimate. The reviewing Mapping Partner <br />shall use the standard error of prediction, if available, because this is more indicative of the <br />predictive accuracy of the equations. <br /> <br />The reviewing Mapping Partner shall consider the proposed flood discharges from the rainfall- <br />runoff model reasonable if they are generally within one standard error of the regression and <br />gaging station estimates. If not, the reviewing Mapping Partner shall review the rainfall-runoff <br />model in greater detail to determine why there are significant differences. Some unique <br />characteristics of the watershed may explain these differences and justify the use of the proposed <br />rainfall-runoff model estimates, and the Mapping Partner that performed the hydrologic analysis <br />shall provide detailed information to explain these unique characteristics. <br /> <br />Even if the criteria for flood discharge reasonableness are satisfied, a review of the rainfall- <br />runoff model is advisable to determine that the model was applied appropriately. <br />Recommendations to use a reasonable flood discharge in the hydraulic model cannot be made if <br />the calculation of the flood discharges was incorrect and yielded reasonable flood discharges <br /> <br />C-14 <br /> <br />Section C.2 <br />