STORMWATER HYDROLOGY AND QUALITY - Low Impact Development and Sustainable Stormwater Management - page 33

runoff across a natural landscape is a process of interconnected microchannels
conveying runoff between small depressions in the land surface, and does not
approximate the concept of overland flow utilized in many hydrologic models. In
that analogy, a friction factor is assumed to be applied to the vegetated surface,
and the runoff is assumed to be a sheet of uniformly deep flow moving across
the landscape. In reality, the movement of surface runoff on a natural landscape
is far too complex to describe with a single equation.
Some investigators have also suggested that the process of rainfall runoff
from natural land surfaces varies both spatially and temporally as the distance
from perennial channels increases. Described as partial area hydrology [5], this
concept proposes that the initial runoff volume is generated from land surfaces
within the floodplain or within the drainage network where wetter soils exist.
As precipitation increases during a storm, the runoff area expands in an ever-
expanding pattern along waterways and is by no means uniform throughout the
watershed. The problem with this analogy is that every watershed would need to
be described in terms of the spatial variability of hydrologic soil groups ,andthe
runoff produced would increase as the area of runoff expanded during rainfall.
No computer model of this mechanism has yet been formulated, but the concept
is important when considering how some portions of a watershed may contribute
greater pollutant loads to runoff.
During rainfall, water is retained in many soils for extended periods and as
surface conditions change, may ultimately be removed by deep roots or even
evaporation. Agricultural productivity has been monitored for decades by mod-
eling the soil moisture content as a guide to drought conditions. However, the
fraction of rainfall that ultimately reaches the zone of saturation, or water table,
is fairly constant from year to year in any given physiographic region.
Since we can measure the aquifer recharge as base flow only when it is
finally discharged to a surface stream, it is difficult is gain direct insight into the
intermediate subsurface movement of rainfall as “groundwater.” We can develop
networks of wells and measure the changing depth in the gradually fluctuating
water table, and we can withdraw water from a single well and measure the
water table change in nearby wells, thereby measuring implicitly the movement
through the aquifers [6]. This tells us a great deal about the permeability or
transmissivity of the rock itself, but does not adequately measure the movement of
water through the three-dimensional saturated matrix of rocks. Several computer
models have been developed [7] that simulate this movement by considering the
rock to exist in layers, each with different permeability, then calculating the water
movement from a unit volume in six directions. Although this type of simulation
has produced some good approximations, especially on larger scales, the overall
movement of infiltrating rain through the subsurface remains a complex process,
with gravity as the fundamental energy of movement.
In theory, a continuous recording stream gage situated at or near the mouth
of a river or stream will measure all of the water draining out of the watershed.
It is possible to separate the volume of flow measured into the two contributing