Environmental Engineering Reference
In-Depth Information
runoff that moves through the channel system to a given point (in practice,
often a stream gauge at which streamflow measurements are made).
Runoff is a key index of the availability of freshwater, a quantity that is
essential for human life. Although on average only about 5 liters per day of
water are necessary for adult survival, water use in industrialized countries
is much higher—around 200 liters/day in the United States and 150 liters
in Europe. On the other hand, the average for Africa is only about 10 liters,
or double the minimum for survival. Total water consumption for agricul-
ture globally is about 10 times that for municipal use (Shiklomanov, 1999).
Groundwater, an important source of water in many parts of the world,
makes up an additional estimated 25 percent of total global water with-
drawals (International Water Management Institute, 2007). Groundwater
recharge, although difficult to estimate except on a local basis, is roughly
related to the excess of precipitation over evapotranspiration, and so very ap-
proximately can be taken as a fraction of annual runoff. Hence, understand-
ing how runoff will change in a future climate is the key to understanding
how water availability for human use might change.
In climate models, runoff is represented by land surface models, which
have the primary purpose of partitioning net radiation at the land surface into
latent, sensible, and ground heat flux. A secondary purpose (which is linked
to modeling of the surface energy balance, because evapotranspiration, or
equivalently latent heat, is common to the energy and water balances) is to
partition precipitation into infiltration and runoff. Runoff is also produced
in most models by parameterizations of subsurface hydrologic processes,
albeit crudely in most models. Although many models do not consider the
transformation of runoff to streamflow, a few do. On long-term (average
annual) balance, runoff is roughly equal to streamflow, ignoring channel
processes (such as groundwater interactions), which in most cases have
modest effect.
Past attempts to understand the sensitivity of runoff and streamflow to
a changing climate have followed two general pathways. The most com-
mon is to use river basin hydrological models, which typically are forced
with precipitation and other surface atmospheric variables, and to prescribe
differences in the forcings that reflect the effects of climate change. This ap-
proach, sometimes termed downscaling, has been applied in one of three
ways. The first is the so-called delta method, which adjusts precipitation and
temperature by factors or shifts that reflect climate model-estimated differ-
ences between current and future climate. An example of this approach is
the study by Hamlet and Lettenmaier (1999) of the effects of climate change
on the water resources of the Columbia River Basin. The second approach
