General information
Offer title : Engineer (M/F) in GIS/remote sensing/geostatistics Google Earth Engine/R (H/F)
Reference : UMR5554-CHRHEL-002
Number of position : 1
Workplace : MONTPELLIER
Date of publication : 15 September 2025
Type of Contract : IT in FTC
Contract Period : 6 months
Expected date of employment : 3 November 2025
Proportion of work : Full Time
Remuneration : Starting at €2,521.95 gross monthly, adjustable based on experience
Desired level of education : BAC+3/4
Experience required : Indifferent
BAP : A - Life, Earth and Environmental Sciences
Emploi type : Geo-Natural and Human-Made Environments Studies Engineer
Missions
The candidate will aim at producing a GIS fire polygon database in France as part of the PREDISPOSE project over the period 1984-2024 (Landsat TM) and 1974-1984 (Landsat MSS). The candidate will use the freeware BAMTS available in the google earth engine platform, using the semi automated and fully automated BAMTS versions, including commission and omission errors following standard remote sening protocols (gini index, OE, CO, OA). Fire dating will be processed with the freely available MODIS MCD14ML and VIIRS, concomitantly with the French national fire inventory database BDIFF. The final deliverable will be composed of a GIS database of fires>5ha, with an attribute table stating the burned area, date and duration. Over the 2001-2024 period, hourly fire spread will be processed with geostatistical tools (R language). Most of the codes will be delivered by the hos laboratory and will have to be processed.
Activities
1) Processing the semi automated BAMTS (GEE) for generating reference fine tuned fire polygons from landsat or sentinel data
2) Automatic BAMTS processing over the 1984-2024 period using reference fire as training samples,
3) Analysing omission and commission errors,
4) Assembling and formatting (cleaning, visual checking) fire polygons with dating and link to BDIFF reference data,
5) Writing document on processing steps and final results.
Skills
The candidate will be familiar with Google Earth Engine (GEE), R computing language and QGIS freeware. Activities will require meticulous assessment of semi automated fire data generated by BAMTS over reference fires, and will need to set up the successive steps of the BAMTS software. Writing skills in English will be required to provide the final deliverable describing the database and the processing chain. Fire polygons might be completed with fire severity (dNBR), and fire spread (geostatistics using freely downloadable global fire hotspots data). Competences in web interface (R package 'shiny') would be a plus.
Work Context
The position will be part of the PREDISPOSE project, aiming at understanding the fire/weather relationship in France, through an historical fire reconstruction using remote sensing over the landsat archives and process-based modeling of fire spread.
Constraints and risks
Computer work; use of GEE and R codes with updates. Calculation time and potentially repetitive tasks requiring concentration and meticulousness in referencing acquired data.