GCOS Implementation Plan - WMO

Implementation Plan for the Global Observing System for Climate in Support of the UNFCCC
(2010 Update)
Action T39
Action: Develop set of active fire and FRP products from the global suite of operational
geostationary satellites.
Who: Through operators of geostationary systems, via CGMS, GSICS, and GOFC-GOLD.
Time-Frame: Continuous.
Performance Indicator: Availability of products.
Annual Cost Implications: 1-10M US$ (Mainly by Annex-I Parties).
6.3. Terrestrial Domain – Data Management and Reanalysis
Terrestrial Reanalysis
As in other fields of science, and most prominently in the atmospheric sciences, data assimilation and
reanalysis techniques have taken on a prominent role in offering a practical way to ensure the
consistency between the various products that are being generated. This is achieved by explicitly
taking into account the degree of uncertainty associated with models (which are always simplified
representations of reality), a priori estimates, and input data (which are always observed or measured
with a finite accuracy). The techniques developed by atmospheric scientists and other geophysicists
are now being applied to the analysis of satellite data over land (including land interactions with the
atmosphere) and permit the retrieval of products such as the components of the surface carbon cycle
or albedo, in such a way that the accuracy of the resulting product is documented as part of the
procedure. This approach helps determine objectively the respective contributions of both models and
data, the optimal ways to improve the observing system to increase the accuracy of the results, and
naturally generates products that are much easier to incorporate in larger models (e.g., climate
models). In the near future, it is likely that multiple land surface products (e.g., albedo and FAPAR)
will be jointly retrieved through such procedures, thereby ensuring that they are all mutually
consistent. The reprocessing of existing satellite remote-sensing archives with these advanced tools
should thus be encouraged, especially in the context of (or perhaps in advance of) large-scale
reanalysis exercises over multiple decades. It is recommended that a review of the state of the art in
land surface albedo (LSA) estimation from space measurements be made in coordination with
AOPC/TOPC to begin the reanalysis of existing datasets by including the following tasks:
• Benchmark existing products (continuation and extension of the initial effort to compare Meteosat,
MODIS, and MISR LSA products), both in space and time, and propose ways and means to
merge such products to generate truly global products with adequate coverage in Polar Regions.
• Investigate the compatibility between albedo products derived from bi-directional reflectance
observations acquired by sensors on geostationary and polar-orbiting platforms.
• Evaluate the factors affecting the quality of LSA products and, in particular, their dependency on
related atmospheric products (e.g., clouds and aerosols).
• Investigate the drawbacks, limitations, and obstacles that have prevented the effective use of
these albedo products in General Circulation Models (GCMs) and recommend ways to address
these issues.
• Promote sensitivity studies and other appropriate projects aimed at documenting the role and
impact of LSA products in climate models, with a view to establishing precise requirements for the
characteristics of this product.
This approach could also very usefully be applied in the case of the soil moisture ECV. A number of
land assimilation projects have assessed soil moisture on a global scale, such as the Global Soil
Wetness Project (http://www.iges.org/gswp/). These projects have produced significant apparent
differences between model outputs but show similar variations from year to year in model projections.
Therefore it appears that there are systematic differences in the projections that are computer-model
dependent. Reanalysis of in situ and satellite data integrated into Earth-system dynamic models
could improve our understanding of soil moisture observations and model projections.
129