IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Postdoctoral position available

Michel Weinfeld weinfeld at lix.polytechnique.fr
Wed Feb 21 12:59:19 EST 1996


      Use and development of new procedures (geostatistic,
    *** neural network techniques ***) for data reduction
                  and structural analysis.


This announcement is transmitted by Michel Weinfeld on behalf of N.Scott
(LMD) to the NN community.

The candidate is not required to be specialised in climate studies, but
of course some interest on the subject is welcome. As can be seen in
the announcement, someone specialist in data reduction and analysis
using Neural Network techniques might be the right person to fulfill
the requirements.

Please forward this call to whoever could be directly or indirectly
interested. Thanks in advance.

Within the frame of the "Human Capital and Mobility Programme", the ARA
group has been selected as Host Institute on the following research topics:
"Climate Studies froms Space Observations: Evolution, Variability"

Our objective is to elaborate an enhanced analysis of the real Earth
climate driving forces: our approach relies upon a strong coupling between
observations (satellite or ground based) and simulations: radiative
transfer (forward and inverse) models, climate models.

Detection of possible climate changes and an improved understanding of the
Earth's climate system require accurate estimates of many atmospheric
thermodynamic properties: vertical temperature and moisture profiles,
atmospheric stability index, cloud characteristics (amount, top pressure
and types), liquid water and precipitations,  surface temperatures and
characteristics (ice, snow, deserts, etc...), radiation fluxes and cloud
forcing, surface wind speeds, etc... . With operational space observation
like the NOAA/TIROS-N series, started in 1979 and still in operation for
the coming years, or the DMSP (Defense Meteorological Satellite Programme)
series, started in 1987, an unique opportunity is offered to study the most
relevant climate variables, globally and over long time series. Another
unique opportunity comes from the ability of global climate models to
describe the full variability of the atmosphere ant its detailed
interaction with the surface of the globe.

This research goes through:
+> the elaboration of a comprehensive long term validated no gap
dataset of atmospheric and surface variables (temperature, water vapor,
clouds, surface characteristics, ...) associated with flux components
(spectral, vertical, surface) and radiative cooling or heating rates. A
three year benchmark period (July1987-June1989) has already been processed
within the frame of the NOAA/NASA Pathfinder programme and the CEC
Environment programme 1993-1994. Plans are to process from 1979 to current
times within the coming 18 months;
+> the analysis of such a dataset for a better understanding of the
forcing due to clouds and greenhouse effect (water vapor);
+> the elaboration, validation and improvement of diagnostics tools
towards, iteratively, improve the parameterization schemes in the GCM's and
improve the retrieval algorithms;
+> the use and development of new procedures (geostatistic,
*** neural network techniques ***) for data reduction and structural analysis.

Computing facilities: computations are performed on the CRAY C-98 of the
IDRIS/CNRS. Other in-house facilities exist.

Period: from June 1996 to December 1997

Salary  according to the CNRS rules (EEC contract)

Interested person should send a CV and statement of research interests
including references by March 5, 1996 to:
        N.A. SCOTT (see address above):
        Phone +33 1 69 33 45 32
                Fax +33 1 69 33 30 05
        E-mail: nas at jungle.polytechnique.fr

More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net