Research project SR/12/158 (Research action SR)
Long-term data sets of homogeneous quality are needed for climate studies. Therefore it is necessary to develop or improve techniques to homogenise data coming from different satellites, as biases exist between them due to their technical characteristics and the changing physical properties of the measured environments.
The objectives of this project are:
- to develop a technique to objectively derive bias fields for the various Sea Surface Temperature (SST) data sets used in meteorological, climate and oceanographic studies;
- to study the temporal and spatial variability of the bias fields between sensors;
- to study the possible marine physical processes that influence the SST and that may be at the origin of the observed differences between sensors;
- to establish a series of cloud statistics and to infer the effect that the cloud presence/persistence has in the SST.
Method
The Advanced Along-Track Scanning Radiometer (AATSR) onboard Envisat spacecraft will be used as a referenced data set. SST differences between SEVIRI and METOP-A and the reference AATSR sensor will be analysed by implementing DINEOF (Data INterpolating Empirical Orthogonal Functions). Two approaches will be tested:
each data set will be reconstructed separately, and then bias feilds between them will be calculated; and
the diference between daatsets will be directly reconstructed with DINEOF.
The complete bias fields will allow for the study of their temporal and spatial variability both on a large scale (North Atlantic and Antarctic) and at a regional scale (North Sea, Mediterranean Sea). Correlation studies will be implemented to assess the relation between the bias fields and different oceanographic variables.
Expected results
- A methodology to correct satellite inter-sensor differences, and that can be applied to any variable where this problematic appears.
- The technique developed within this proposal will be embedded in the CMS -MeteoFrance correction chain. If the embedding proves successful, it can potentially benefit end users at an international level, as the products generated at MeteoFrance are delivered through the MyOcean project to numerous users.
- A series of reports addressing the relation of the SST differences with properties such as the mixed layer depth and cloud statistics.
- The code of the basic technique used in this proposal (DINEOF) is freely available to download, and all developments made in the frame of this project will be also made available, assuring the usability of the approach by interested end-users.
- Dissemination of results: scientific publications in international peer-reviewed journals and presentation of results in international conferences.
