408
I. Papadakis and G.A. Schultz
The application of the technique requires that various parameter values and the
mathematical approach have to be specified prior to computations:
time resolution of the images (given by sensor system, may be chosen as multiples thereof)
spatial and time resolution for area precipitation computations
cloud top temperature threshold value relevant for precipitation occurrence
mathematical relationship between satellite information and area precipitation
The choice of the above factors depends on the requirements of the user.
Based on the technique of Arkin the determination of the Fractional Cloud
Cover Index (FCCI) may be carried out in the following two consecutive steps,
nanlely:
Step I: Identification of the cloud cell areas which are colder than a certain
threshold cloud top temperature value, covering the river basin under
consideration either totally or partially in the thermal infrared spectral
band.
Step 2: Addition of all pixels for every cloud cell as obtained in Step 1 within the
river basin under consideration.
The result of this procedure consists of a discrete Fractional Cloud Cover Index
(FCCI) for each image having values between 0 and I representing the fraction of
area (or relative area) of the total river basin under consideration, for which a
certain information (Cloud top temperature) relevant for precipitation may be
obtained in one or more spectral bands of a satellite system. The relative area is
expressed by the number of relevant satellite image pixels divided by the total
number of satellite image pixels within the river basin.
The AFCC (Accumulated Fractional Cloud Cover) which is of interest for the
hydrological computations (rainfall estimation) is simply the sum of the FCCI's for
a given time interval (e.g. a day or a month).
In order to estimate the rainfall in a time interval, an empirical relationship of
best fit may be determined using the least-squares method. This means that different numerical values depending upon the radiance, e.g. the threshold cloud top
temperature (from the thermal infrared spectral band) have to be varied such that
an optimum agreement between observed rainfall and spectral information in the
same time interval is achieved. The mathematical function P = f(AFCC) is derived
specifically for different climatic regions. For the test catchment in West Africa the
equation is shown in Sect. 18.3.2.1.
I. Papadakis and G.A. Schultz
The application of the technique requires that various parameter values and the
mathematical approach have to be specified prior to computations:
time resolution of the images (given by sensor system, may be chosen as multiples thereof)
spatial and time resolution for area precipitation computations
cloud top temperature threshold value relevant for precipitation occurrence
mathematical relationship between satellite information and area precipitation
The choice of the above factors depends on the requirements of the user.
Based on the technique of Arkin the determination of the Fractional Cloud
Cover Index (FCCI) may be carried out in the following two consecutive steps,
nanlely:
Step I: Identification of the cloud cell areas which are colder than a certain
threshold cloud top temperature value, covering the river basin under
consideration either totally or partially in the thermal infrared spectral
band.
Step 2: Addition of all pixels for every cloud cell as obtained in Step 1 within the
river basin under consideration.
The result of this procedure consists of a discrete Fractional Cloud Cover Index
(FCCI) for each image having values between 0 and I representing the fraction of
area (or relative area) of the total river basin under consideration, for which a
certain information (Cloud top temperature) relevant for precipitation may be
obtained in one or more spectral bands of a satellite system. The relative area is
expressed by the number of relevant satellite image pixels divided by the total
number of satellite image pixels within the river basin.
The AFCC (Accumulated Fractional Cloud Cover) which is of interest for the
hydrological computations (rainfall estimation) is simply the sum of the FCCI's for
a given time interval (e.g. a day or a month).
In order to estimate the rainfall in a time interval, an empirical relationship of
best fit may be determined using the least-squares method. This means that different numerical values depending upon the radiance, e.g. the threshold cloud top
temperature (from the thermal infrared spectral band) have to be varied such that
an optimum agreement between observed rainfall and spectral information in the
same time interval is achieved. The mathematical function P = f(AFCC) is derived
specifically for different climatic regions. For the test catchment in West Africa the
equation is shown in Sect. 18.3.2.1.