6
Precipitation
C. G. Collier
Telford Institute of Environmental Systems
Department of Civil and Environmental Engineering
University of Salford, United Kingdom
6.1 Introduction
Precipitation is the central component of the hydrological cycle, and as such is of
primary importance in hydrology. It shows large and frequent spatial variations and
usually exhibits rapid temporal variations. Table 6.1 illustrates the spectrum of
significant precipitation events that can occur and emphasizes the important change
of approach which is forced upon the analyst as he crosses the rain-day time averaging threshold. For time periods less than one day precipitation distributions are often
analysed in terms of the 'cloud-scale' motions that cause them. For time periods
greater than one day it becomes increasingly difficult, and ultimately impossible, to
do this. Up to about 3 days, synoptic scale systems can provide a framework for
analysis, although this can be difficult as such systems often move and develop
significantly over this period.
The scales over which precipitation occurs are so large that methods of measurement are wide ranging, and usually complementary. Point measurements are insufficiently representative of sub-catchment scales, and hence it is difficult to understand
and model hydrological processes using such data. This point is emphasised in Colour
Plate 6.A which shows a rainfall situation for which point measurement from raingauges miss most of the rain as measured by ground-based radar. The basic measurement requirement of hydrology is for areal measurements albeit over sometimes very
small, as well as very large, areas.
The hydrological requirement for measurements of precipitation is discussed by
Collier (1996a) in terms of the maximum, minimum and most usual values for
resolution, frequency of observations and accuracy (freedom from bias). Precision is
also expressed qualitatively as the standard deviation of results from repeated trials
under identified conditions. Hence, a measurement from a raingauge is spatially
precise, but may be imprecise in time and quantity. On the other hand radar and
satellite measurements are usually precise in time, less so in space, and much more
imprecise in quantity. It is therefore important to regard all these types of measurement as complementary. In this chapter we discuss, in particular, radar and satellite
measurements of precipitation. The reader is referred to other texts for detailed
discussion of point measurements of rainfall and snowfall (see for example Collier,
1996b).
G. A. Schultz et al. (eds.), Remote Sensing in Hydrology and Water Management
© Springer-Verlag Berlin Heidelberg 2000
Precipitation
C. G. Collier
Telford Institute of Environmental Systems
Department of Civil and Environmental Engineering
University of Salford, United Kingdom
6.1 Introduction
Precipitation is the central component of the hydrological cycle, and as such is of
primary importance in hydrology. It shows large and frequent spatial variations and
usually exhibits rapid temporal variations. Table 6.1 illustrates the spectrum of
significant precipitation events that can occur and emphasizes the important change
of approach which is forced upon the analyst as he crosses the rain-day time averaging threshold. For time periods less than one day precipitation distributions are often
analysed in terms of the 'cloud-scale' motions that cause them. For time periods
greater than one day it becomes increasingly difficult, and ultimately impossible, to
do this. Up to about 3 days, synoptic scale systems can provide a framework for
analysis, although this can be difficult as such systems often move and develop
significantly over this period.
The scales over which precipitation occurs are so large that methods of measurement are wide ranging, and usually complementary. Point measurements are insufficiently representative of sub-catchment scales, and hence it is difficult to understand
and model hydrological processes using such data. This point is emphasised in Colour
Plate 6.A which shows a rainfall situation for which point measurement from raingauges miss most of the rain as measured by ground-based radar. The basic measurement requirement of hydrology is for areal measurements albeit over sometimes very
small, as well as very large, areas.
The hydrological requirement for measurements of precipitation is discussed by
Collier (1996a) in terms of the maximum, minimum and most usual values for
resolution, frequency of observations and accuracy (freedom from bias). Precision is
also expressed qualitatively as the standard deviation of results from repeated trials
under identified conditions. Hence, a measurement from a raingauge is spatially
precise, but may be imprecise in time and quantity. On the other hand radar and
satellite measurements are usually precise in time, less so in space, and much more
imprecise in quantity. It is therefore important to regard all these types of measurement as complementary. In this chapter we discuss, in particular, radar and satellite
measurements of precipitation. The reader is referred to other texts for detailed
discussion of point measurements of rainfall and snowfall (see for example Collier,
1996b).
G. A. Schultz et al. (eds.), Remote Sensing in Hydrology and Water Management
© Springer-Verlag Berlin Heidelberg 2000