Geoscience Reference
In-Depth Information
Quantifying the land state and fluxes and understanding soil moisture-temperature and
soil moisture-precipitation couplings allow a better representation of hydrological pro-
cesses in climate models and significantly help to reduce uncertainties in future climate
scenarios, in particular regarding changes in climate variability and extreme events, and
ecosystem/agricultural impacts (Seneviratne et al. 2010 ). This understanding is also cru-
cially important for improving short-range numerical weather prediction (NWP) capabil-
ities, in particularly regarding prediction of convective precipitation (Sherwood 1999 ;
Adams et al. 2011 , and references therein).
Hydrological observations are prone to errors and are discrete in space and time with
the result that the information provided by these observations has gaps. Figure 2 shows an
example of gaps in satellite observations. It is desirable to fill gaps in the observed
information using additional information and computational techniques. Algorithms or
models to fill in information gaps should organize, summarize and propagate the infor-
mation from observations in an objective and consistent way. A simple approach such as
linear interpolation could be a reasonably accurate ''model'', when observations are dense
enough. However, linear interpolation may not be consistent with our advanced under-
standing of how the land surface behaves. A more realistic approach would be to fill in the
gaps using a land surface model (LSM). While observations give an instantaneous view of
the land surface, LSMs provide continuous estimates, based on physical laws that are
derived from historical observations. These models are not perfect, and gaps in their
structure, parametrization or initialization can be filled in with observations.
Fig. 2 Plot representing retrieved soil moisture data from the Soil Moisture Ocean Salinity (SMOS, Kerr
et al. 2010 ) mission for August 3, 2012 (top left panel), August 10, 2012 (top right panel), August 17, 2012
(bottom left panel), and August 25, 2012 (bottom right panel), based on the observational geometry from
ascending orbits from SMOS (units of m 3 m -3 ). Blue denotes relatively wet values; red denotes relatively
dry values. The uncoloured (i.e. grey) areas over land represent gaps between the satellite orbits.
Noteworthy are the sparse SMOS observations over Scandinavia, where retrievals from remotely sensed
observations are particularly difficult, when the land is covered with snow, ice, forest, water bodies or rocks
Search WWH ::




Custom Search