EGU logo

European Geosciences Union

www.egu.eu

PhD position in Digital Soil Mapping and Remote Sensing of cropland soils

Position
PhD position in Digital Soil Mapping and Remote Sensing of cropland soils

Employer

INRA ECOSYS

The INRA-AgroParisTech ECOSYS Joint Research Unit was created on 1 January 2015 through merging of the Environment and Field Crops (EGC) unit (head of unit, Enrique Barriuso between 2010 and 2015), the INRA Physicochemistry and Ecotoxicology of Soils of Contaminated Agrosystems (PESSAC) unit (head of unit, Christian Mougin, 2010-15), and the INRA and AgroParisTech staff of the Soil Organic Matter team of the BIOEMCO unit. This last group of staff had been previously integrated to the EGC unit since 1 January 2014. The EGC unit was created at the Grignon site in 2000 by the grouping of two units: the Bioclimatology and Soil Science unit and the Plant Ecophysiology team of the Agronomy unit. The PESSAC unit was created at the Versailles site in 2006 as the INRA pole of terrestrial ecotoxicology. This unit was the result of the merging of the Soil Science unit from Versailles (INRA Environment and Agronomy division – EA) and the Xenobiotics and Environment team of the Phytopharmacy unit (INRA Plant Health and Environment division – SPE).
The ECOSYS supervisors are INRA (EA and SPE divisions, with the EA division as pilot division) and AgroParisTech (Agronomic, Forestry, Water and Environment Sciences and Engineering division – SIAFEE). Currently, the unit is located in three buildings, two at the Grignon site and one at the Versailles site. The whole unit will be regrouped in Saclay in 2021.

Homepage: https://www6.versailles-grignon.inra.fr/ecosys_eng/Research


Location
THIVERVAL-GRIGNON or ORLEANS or SACLAY, France

Sector
Academic

Relevant divisions
Earth and Space Science Informatics (ESSI)
Geosciences Instrumentation and Data Systems (GI)
Soil System Sciences (SSS)

Type
Full time

Level
Entry level

Salary
min. 31435 € / Year, ~2620 €/month including payroll taxes, employer costs and premiums

Required education
Master

Application deadline
Open until the position is filled

Posted
15 July 2019

Job description

The 4 per 1000 initiative launched at the Paris climate summit (COP21) in December 2015 aspires to increase global SOM stocks by 0.4 percent per year as a compensation for the global emissions of greenhouse gases by anthropogenic sources (Minasny et al., 2017) and targets the agricultural soils in priority. In keeping with the 4 per 1000 initiative, there is a crucial need for assessing and monitoring the SOM stocks or organic C stocks of soils for agricultural areas. So far, at regional scales, there is no efficient and straightforward way of monitoring topsoil organic C (SOC). Both scarcity, spatial and spectral resolution of the satellite data that were acquired in the past did not enable to monitor SOC from space. This PhD project is backed by the several projects aspiring to take advantage of the recently available satellite Sentinel time series for this purpose.
Now that one can collect a large number of Sentinel images over the time of crop cycles, there are several challenges to face: i) target the acquisition time(s) that are best adapted for predicting SOC contents; ii) construct methods for mosaïcking bare soil pixels in order to increase the predicted area; iii) carry out mixed methods combining spatial statistics and multidate imagery, for producing spatially exhaustive maps and their related uncertainties.
The objective of this PhD thesis is to evaluate the capacity of Sentinel 2 and/or 1 satellite time series to monitor SOC for agroecosystems. It questions the quantitative performance that is reachable through plurimensual and multiyear optical Sentinel-2 combined or not to Sentinel-1 radar time series, for diversified croplands with contrasted soil types, climatic conditions and agricultural practices, located within the Parisian Basin (Versailles Plain, Beauce), region of Toulouse or in French Brittany.
Varied models will be constructed and compared, being either (i) spatial-geostatistical SOC prediction models based on spatial data only in the absence of information derived from image reflectance spectra; (ii) reflectance-derived SOC prediction models, particularly based on spectral S2 data (Vaudour et al., 2019); (iii) mixed spatial-spectral models based on both (Loiseau et al., 2019).


How to apply

please contact Emmanuelle Vaudour (emmanuelle.vaudour@agroparistech.fr) and Dominique Arrouays (dominique.arrouays@inra.fr)
and provide curriculum vitae and letter of motivation