FLOOD08374 - Laserfiche WebLink

Home Browse Search

-13- . combined in a predictive model. Watershed boundaries were first drawn on the topographic quadrangles following generally accepted procedures which involve the isolation of all surface drainage above a stream gaging point and require the crossing of all contour lines at right angles. Although these techniques are something of an art, they are generally quite reproducible in fluvial terrain. In the glacial terrain of south- eastern New England, however, greater uncertainty in the precise location of drainage divides arises from two factors: the presence of broad areas having low relief which is imperfectly portrayed by the fixed contour interval of ten feet; and the occurrence of wetland areas which drain both into and out of a watershed, and therefore lie on the divide. Such inaccuracies might be expected to cancel somewhat as they accumulate, but errors as large as a few percent may exist in various areal data. Delineation of the drainage network was a matter of concern since several possible elements of the predictive model would be dependent on it. The glacial history of the study region has left a strong imprint on the drainage composition. Lakes, small ponds and wetland areas are abundant, and the drainage is deranged in many places, being neither closely related to the topography nor in a state of even approximate equilibrium. The majority of channels are permanent and some of the drainage is artificial. Because of these factors, any standard methods of quantitative measurement can only be approximate. One would expect all potential elements of the drainage network to be active during peak flow conditions. Therefore, it might be most appropriate to identify all such channels through detailed topographic analysis of contour crenula- tions and interpretation of stereo photographs. However, the great a- mount of labor involved in accomplishing this task for a total watershed . .