12
G.A. Schultz and E.T. Engman
1.4.4 Temporal Resolution
RS data are acquired with a given resolution in time. Also here the resolution
varies very much from sensor to sensor and satellite system. Ground-based
weather radar data can be acquired every 5 minutes, geostationary satellites provide data every half hour and some polar orbiting satellites provide data as seldom
as every 16 days (e.g. Landsat) or longer with some of the narrow swath SARs.
The hydrologist considering use of remote sensing data has to choose data that
match the needs of his analysis. In some cases of dynamic processes and small
basins, the data may be needed daily of more often. In other cases for less dynamic processes and large basins, data on two week or longer may be satisfactory.
An example of this might be snow melt runoff prediction in a large drainage basin.
Then again, there are some needs that have little or no temporal criteria. Examples
of this would be delineating stream channels where maps do not exist or land use
which changes very slowly.
Appendix 20.1 lists the currently available and future satellite systems that are
of interest to hydrologists and water managers. Appendix 20.2 is a similar table
that lists the specific sensor and orbit characteristics that are of great interest to
hydrologists and water managers.
1.5 Remote Sensing and Hydrology
It should be kept in mind, that RS data are not only used for monitoring of hydrological state variables, but also as the basis for parameter estimation of hydrological models. Remote sensing, particularly from various satellites in various spectral
bands, can provide information on catchment characteristics (e.g. landcover, landuse, slope, vegetation), from which the parameters of hydrological models can be
gathered. Particularly in combination with other spatial information, such as digital elevation models, digital terrain models, digital soil maps. RS will allow the
spatial estimation of hydrological model parameters, e.g. the maximum soil water
storage capacity in a river basin.
Another important facet of remote sensing is the fact, that such data can be acquired in remote areas, where no measurements are feasible or can be carried out
only under very difficult circumstances which cause high costs. For these measurements particularly airplanes and satellites are suitable. Furthermore satellite
remote sensing allows coverage of the whole globe, which is highly relevant in the
development of global coupled atmospheric-hydrological models for weather and
flood forecasting as well as for long-term analysis and forecast of climate conditions. This property makes RS data particularly valuable for all activities within
the framework of the world climate research program (e;g. GEWEX, CLIV AR,
ACSYS etc.).
Looking at the historical development of hydrological modeling one comes easily to the conclusion, that all hydrological models are data limited. Models are
generally not built in the way which would be scientifically most sound, but rather
according to data availability. A significant example of this deficiency of existing
models is the fact, that rainfall-runoff models as well as water balance models are
G.A. Schultz and E.T. Engman
1.4.4 Temporal Resolution
RS data are acquired with a given resolution in time. Also here the resolution
varies very much from sensor to sensor and satellite system. Ground-based
weather radar data can be acquired every 5 minutes, geostationary satellites provide data every half hour and some polar orbiting satellites provide data as seldom
as every 16 days (e.g. Landsat) or longer with some of the narrow swath SARs.
The hydrologist considering use of remote sensing data has to choose data that
match the needs of his analysis. In some cases of dynamic processes and small
basins, the data may be needed daily of more often. In other cases for less dynamic processes and large basins, data on two week or longer may be satisfactory.
An example of this might be snow melt runoff prediction in a large drainage basin.
Then again, there are some needs that have little or no temporal criteria. Examples
of this would be delineating stream channels where maps do not exist or land use
which changes very slowly.
Appendix 20.1 lists the currently available and future satellite systems that are
of interest to hydrologists and water managers. Appendix 20.2 is a similar table
that lists the specific sensor and orbit characteristics that are of great interest to
hydrologists and water managers.
1.5 Remote Sensing and Hydrology
It should be kept in mind, that RS data are not only used for monitoring of hydrological state variables, but also as the basis for parameter estimation of hydrological models. Remote sensing, particularly from various satellites in various spectral
bands, can provide information on catchment characteristics (e.g. landcover, landuse, slope, vegetation), from which the parameters of hydrological models can be
gathered. Particularly in combination with other spatial information, such as digital elevation models, digital terrain models, digital soil maps. RS will allow the
spatial estimation of hydrological model parameters, e.g. the maximum soil water
storage capacity in a river basin.
Another important facet of remote sensing is the fact, that such data can be acquired in remote areas, where no measurements are feasible or can be carried out
only under very difficult circumstances which cause high costs. For these measurements particularly airplanes and satellites are suitable. Furthermore satellite
remote sensing allows coverage of the whole globe, which is highly relevant in the
development of global coupled atmospheric-hydrological models for weather and
flood forecasting as well as for long-term analysis and forecast of climate conditions. This property makes RS data particularly valuable for all activities within
the framework of the world climate research program (e;g. GEWEX, CLIV AR,
ACSYS etc.).
Looking at the historical development of hydrological modeling one comes easily to the conclusion, that all hydrological models are data limited. Models are
generally not built in the way which would be scientifically most sound, but rather
according to data availability. A significant example of this deficiency of existing
models is the fact, that rainfall-runoff models as well as water balance models are