Towards a Mediterranean GEWEX
365
research Area". An important sixth component is essential but will not further be
discussed here: The joint outreach to the wider scientific and user's communities.
The results of the communities addressed here certainly are of considerable
importance for hydrological development strategies and the problem of
desertification in the Mediterranean area furthered by the United Nations
Convention to Combat Desertification. A logical consequence therefor would be that
this part of the information is directly fed into the EEA/FMA project DIS/MED
(Desertification Information System to support National Action Programmes in the
Mediterranean).
5 The Potential of Observations Made From Space
Remote sensing methods are limited to a few measurable quantities which are as
stated before of a completely different nature than the information which models and
in situ measurements produce. They deal with reflection, scattering, absorption and
emission of electromagnetic waves, while in situ state variables such as pressure,
temperature, constituent concentrations, material and energy fluxes are measured.
From the radiance, polarization and signal return time measurements made in space
in regular intervals nevertheless a wealth of information about the environment is
inferred. In Table 1 quantities are listed which are necessary to determine changes
and study processes immanent to the system. These are or can be affected by global
changes and are used in mathematical descriptions as for empirical diagnostics of
the system. To the assessment of a number of them remote sensing can contribute
substantially. By analysis of long term data series these quantities also can be used
as describers or "indicators" for changes. But if one wants to know what precisely
changed and why, more detailed information is required. In the past few years
methods have been developed to assess this information from remote sensing with
a minimum of collateral data.
As an example let us look at the easily to construct Normalized Difference
Vegetation Index (NOVI) that indicates changes of the "greenness" of the landscape,
such as the mean annual vegetation indices. To the variability from year to year a
number of causes can contribute: Change of precipitation, change of groundwater
table, variable length of the vegetation period, vegetation stress due to high
temperatures, changes of the density ofthe vegetation cover, variability of biomass
production, a change of crops, irrigation, or fires. To entangle these causes and to
obtain a clearer view at the processes additional information has to be aquired.
Trends of the vegetation index can be brought in relation to desertification. Land-use
changes can be analysed with the aid of spectral retlectances and, ifhigh resolution
images (such as from aircraft) are available, of the planting pattern. If the radiative
properties of the components covering the surface are known, it is possible to
determine by "spectral decomposition" their fractional coverage. Spectral
retlectances combined with a measure of surface albedo will inform us, if the net
solar tlux at the surface changed. The effective surface temperature adds the
information about the net infrared flux and by this allows to estimate the total net
365
research Area". An important sixth component is essential but will not further be
discussed here: The joint outreach to the wider scientific and user's communities.
The results of the communities addressed here certainly are of considerable
importance for hydrological development strategies and the problem of
desertification in the Mediterranean area furthered by the United Nations
Convention to Combat Desertification. A logical consequence therefor would be that
this part of the information is directly fed into the EEA/FMA project DIS/MED
(Desertification Information System to support National Action Programmes in the
Mediterranean).
5 The Potential of Observations Made From Space
Remote sensing methods are limited to a few measurable quantities which are as
stated before of a completely different nature than the information which models and
in situ measurements produce. They deal with reflection, scattering, absorption and
emission of electromagnetic waves, while in situ state variables such as pressure,
temperature, constituent concentrations, material and energy fluxes are measured.
From the radiance, polarization and signal return time measurements made in space
in regular intervals nevertheless a wealth of information about the environment is
inferred. In Table 1 quantities are listed which are necessary to determine changes
and study processes immanent to the system. These are or can be affected by global
changes and are used in mathematical descriptions as for empirical diagnostics of
the system. To the assessment of a number of them remote sensing can contribute
substantially. By analysis of long term data series these quantities also can be used
as describers or "indicators" for changes. But if one wants to know what precisely
changed and why, more detailed information is required. In the past few years
methods have been developed to assess this information from remote sensing with
a minimum of collateral data.
As an example let us look at the easily to construct Normalized Difference
Vegetation Index (NOVI) that indicates changes of the "greenness" of the landscape,
such as the mean annual vegetation indices. To the variability from year to year a
number of causes can contribute: Change of precipitation, change of groundwater
table, variable length of the vegetation period, vegetation stress due to high
temperatures, changes of the density ofthe vegetation cover, variability of biomass
production, a change of crops, irrigation, or fires. To entangle these causes and to
obtain a clearer view at the processes additional information has to be aquired.
Trends of the vegetation index can be brought in relation to desertification. Land-use
changes can be analysed with the aid of spectral retlectances and, ifhigh resolution
images (such as from aircraft) are available, of the planting pattern. If the radiative
properties of the components covering the surface are known, it is possible to
determine by "spectral decomposition" their fractional coverage. Spectral
retlectances combined with a measure of surface albedo will inform us, if the net
solar tlux at the surface changed. The effective surface temperature adds the
information about the net infrared flux and by this allows to estimate the total net