Geoscience Reference
In-Depth Information
3
Data for Rainfall-Runoff Modelling
It may seem strange to end a review of modelling with an observation that future progress is
very strongly linked to the acquisition of new data and to new experimental work but that,
in our opinion, is the state of the science.
George Hornberger and Beth Boyer, 1995
Ultimately, the success of a hydrological model depends critically on the data available to set it up
and drive it. In the first two chapters of this topic there have been several references to the fact that
hydrology is limited as a science by data availability and measurement techniques. In some areas of data
collection relevant to rainfall-runoff modelling, techniques have improved in recent years. We now have
a much better idea of spatial rainfall variations due to the development of rainfall radar; improvements in
transducers and data loggers have led to more reliable and more continuous measurements of water levels,
water tables and soil moisture; ultrasonic devices have led to faster (if not necessarily more accurate)
measurements of flow velocities in streams and estimation of discharges; there are techniques available
for the direct estimation of evapotranspiration rates; and remote-sensing techniques have led to a range of
spatial datasets being available for use in modelling. This topic is not about data collection for hydrological
purposes and does not cover measurement techniques in detail but the rest of this chapter considers the
issues associated with the main types of data that are available to the rainfall-runoff modelling process.
3.1 Rainfall Data
Rainfall-runoff modelling still depends heavily on the records from point raingauges, both recording
raingauges giving estimates of rainfall intensities at time steps of one hour or better, and daily raingauges.
In large catchments, models using a daily time step may be perfectly adequate for application purposes: the
spatial variation in inputs is generally more important than the temporal variation. In small catchments, a
daily time step may be longer than the storm response time of the catchment and finer time resolution may
be required for adequate modelling of the dynamics of the response and the hydrograph peak. Recording
