G4.11

G4.11   Geographical sampling of reference in-situ data

Gap detailed description

Comparisons between satellite observations and NWP model and reanalysis fields often reveal biases that have a geographical structure. This has been observed both in previous studies (e.g. Lu et al 2011, Lu and Bell 2014) and in the assessment of the FY-3C sounders (MWTS-2, MWHS-2) for work package 4. Similar biases are observed for a number of different satellite instruments measuring at the same frequency, and so are likely to be due either to errors in the radiative transfer forward model used to transform the background fields into radiance space or to errors in the background fields themselves. Comparing reference in-situ data for atmospheric temperature and humidity to the model background fields could give us estimates of the uncertainty in the background fields at different locations and help us determine whether the geographical biases observed in O – B statistics are due to the model background or to other sources. However, the reference in-situ data is limited to a small number of locations and so this is only possible if the areas sampled cover the whole range of geographically-dependent biases that have been observed for satellite data.

Activities within GAIA-CLIM related to this gap

An assessment of observation minus NWP background statistics, sub-sampled at the GRUAN reference sites, is envisaged in work package 1 (task 1.5). This should help to develop an understanding of whether the GRUAN reference data are ‘geographically capable’ of assessing the geographic variability of NWP background errors. See task 1.5 documents for more details.

In addition, an assessment of departure statistics for measurements from the GUAN will be used to help assess the extent to which the background errors are homogeneous, and hence the validity of the assumption that NWP fields can be sampled at a limited number of GRUAN sites, and the resulting derived uncertainties assumed to apply globally.

Gap remedies

Remedy #1

Background departure analysis at GRUAN (and GUAN) locations

Specific remedy proposed

Assess background departures at GRUAN (and GUAN) locations, to assess the capability of GRUAN reference data to characterise background errors.

 Measurable outcome of success

Report on a quantitative assessment of background departures at GRUAN sites, as well as at GUAN sites.  The background departures from the GRUAN sites will be computed and evaluated in both geophysical and radiance space. 

Achievable outcomes

Technological / organizational viability: Medium.  This work forms aprt of the work planned for WP4 of GAIA-CLIM.

Indicative cost estimate: low (< €0.5 million)

Relevance

The solution proposed here is fully aligned with the requirement (to establish traceable uncertainties for NWP fields and radiances calculated from them).

Timebound

Remedy#2 forms part of the workplan for GAIA-CLIM, and will be complete by 2017.

Assessment in task 1.5 (see above).

Gap risks to non-resolution