where xT is the adjusted growth curve factor for the T-year return period, xruralT is the 'as rural' growth factor for the T-year return period and UAF is as given above. The effect of the adjustment is to modify the underlying 'as rural' growth curve progressively with the greatest modification occurring at low return periods. The rural and urban growth curves are assumed to converge at a return period of 1000 years. This represents the concept that, within a moderately urbanised catchment, the urbanisation will have proportionally the greatest impact on runoff for smaller storms, whereas for a very large flood flow, the soils are likely to be uniformly saturated (reducing the difference in runoff production between permeable and impermeable areas) and the sheer depth of rainfall will tend to dominate over differences in local runoff coefficients (see Fig. 18.3). The basic urban adjustment is relevant for the degree of urbanisation observed in the URBEXT descriptor for the year 2000. For analysis of past or projected

Fig. 18.3 FEH urban growth curve adjustment.

future changes, the URBEXT value can be adjusted according to a general national model of urban development that is encoded within the software packages that support the FEH, or by map-based analysis of urban locations.

The FEH rainfall-runoff model and ReFH methods described in Chapter 13 are generally recommended for catchments with a high proportion of urban land cover, and for extremely heavily urbanised catchments (e.g. URBEXT1990 > 0.5) the advice is to use sewer design methods such as the WallingfordProcedure discussed below. The revised flood hydrograph method ReFH was calibrated with only seven urban catchments and so the general guidance is not to use it for heavily urbanised catchments, though further research may change this position in the future.

0 0

Post a comment