POSTDOC – Hyperspectral & sol (IRD + IRSTEA)

Offre d'emploi : développeur pour le projet Smartgrappe
septembre 11, 2013
Offre d'emploi – CDD Développement Informatique
septembre 12, 2013
Show all

POSTDOC – Hyperspectral & sol (IRD + IRSTEA)

L’UMR LISAH et l’UMR ITAP recherchent un Post-doc pour étudier l’impact de la résolution spectrale pour la cartographie des propriétés des sols.
12 mois à partir de décembre 2013
POST -DOC : « Impact of the spectral resolution of Hyperspectral Vis-NIR data on soil properties mapping »
Context of the research
This research will contribute to the French HYPXIM hyperspectral space sensor definition. The postdoctoral position is hosted at INRA (UMR LISAH) in close collaboration with IRTSEA (UMR ITAP).
Job description
HYPXIM is a new French hyperspectral space sensor (0.4-2.5μm), planned to be launched in 2019, which will meet the needs of a wide community of users in the world, including in soil science. One of the HYPXIM missions will be the quantification of physico-chemical and bio-chemistry soil properties and the study of their roles in the hydrological cycle and climate change. The choice of the spectral resolution is still under discussion (between 6 to 15nm). The aim of this post-doctoral research is to examine the possibilities of this new hyperspectral sensor to extract topsoil properties (Texture, Calcium Carbonate, iron…) under varying spectral resolutions.
Different tasks would have to be conducted including the following:

  • Simulation of HYPXIM hyperspectral data at several spectral resolutions (6, 10 and 15nm), using Lab Vis-NIR spectra.
  • Analysis of soil properties prediction performances at each spectral resolution, using PLSR models and HYPXIM data simulated from Lab spectra.
  • Simulation of HYPXIM hyperspectral data with several spectral resolutions (at 6, 10 and 15nm), using HySpex airborne Vis-NIR spectra.
  • Development of methods for soil properties estimation, adapted to each spectral resolution. (based on regression models or spectral absorption features study).
  • Mapping of soil properties from HYPXIM data simulated from HySpex data and Analysis of uncertainties affecting predictions

The successful candidate will be expected to attend an international conference in 2014 and to publish his results in international peer-reviewed journals.

Avalaible data :

  • Vis-NIR HYSPEX airborne data (acquired in 2013 in Mediterranean area in France).
  • HYPXIM space data at 4, 8, 20 and 30m of spatial resolution, simulated from Vis-NIR HYSPEX airborne data.
  • Around 200 soil samples with Lab Vis-NIR spectra (ASD FieldSpect III) and physico-chimical analysis (texture, CaCO3, iron, pH, CEC, organic matter…).

The candidate must have skills in remote sensing and/or soil sciences, scientific programmation (Matlab, R or IDL), with notions of statistics and/or chemometrics.

Timing :
Duration : 12 months, Beginning : before end of 2013
Candidates should submit, by email, a cover letter with a statement of research interests and a CV to Cécile Gomez (UMR LISAH, Montpellier, France)