Geoscience Reference
In-Depth Information
Chapter 1
Basics of Climatological and Meteorological
Observations for GIS Applications
Weather and climate data are spatially distributed. Geographical information
technologies can therefore provide a useful and relevant working environment for
the distribution, integration, visualization, and analysis of these data. However,
compared to other scientific areas, the application of geographical information
system (GIS) tools was for a long time a clumsy process within meteorology and
climatology, and especially within most national meteorological services (NMS);
because of the shortcomings of GIS related to the underlying data model and
missing interfaces to standard meteorological tools (e.g. weather forecast model).
While the GIS data models are highly static based, meteorological data models have
a need for a strong dynamical component with causal dependencies in the
space/time domain (see for example [CHR 02]). Nativi
et al.
[NAT 04] describe the
differences between both underlying data models and advocate models that are
supported by so-called interoperability services. In addition to these differences in
the data models, there are significant differences in the spatial modeling approaches.
In general, GIS environments have implemented the geo-statistical modeling tools
that are based on one temporal realization only, whereas meteorological data offer
the temporal sample in addition to the spatial sample, which results in different
spatial modeling approaches [SZE 04]. However, within the last few years efforts
for integration of meteorological data models in GI environments were quite
successful, and well-established GIS web-mapping standards and spatial
infrastructures have gained increasing importance in meteorology and climatology.
Thus parallel efforts and development currently appear to be resolved [SHI 05].
Information to be derived from climate variability analyses is strongly dependent
not only on the spatiotemporal density, but also on the quality of the available data.
Today it is a well-established fact in climatology that the climate signal from
measurements, beside the statistical noise, is by inhomogenities. Therefore, a
primary step of climate studies is to analyze the input data used with respect to
