corporate_fare
INRAE
verifiedVerified ListingIntegration of Administrative, Agricultural, and Remote Sensing Data for Irrigation Mapping in Occitanie
schedulePosted 21 days ago
workInternship
bar_chartEntry
AIData AnalysisData quality validationFrenchGeomaticsGeospatialGeospatial Data ScienceMachine LearningMappingPythonRemote SensingSpatial Data
About the Role
About INRAE
The French National Research Institute for Agriculture, Food, and Environment (INRAE) is a public research establishment bringing together around 12,000 staff across 272 research, service, and experimental units in 18 locations nationwide. INRAE is a global leader in agricultural, food, plant, and animal sciences, aiming to develop solutions for multi-performing agriculture, high-quality food, and sustainable resource and ecosystem management.
Work Environment, Missions & Activities
Following the severe drought in France this summer, irrigation remains a critical issue: in 2023, it accounted for 10% of the 29 billion m³ of water abstracted and 61% of water consumption. In Occitanie, 1 in 4 farms uses irrigation, covering about 10% of agricultural land (2020 Agricultural Census). However, fine-scale spatial data on irrigation practices remain scarce—limiting evidence-based policy and research.
Existing studies have demonstrated the potential of Sentinel-1 and Sentinel-2 time series for field-level irrigation mapping using machine learning models trained on ground-truth data (e.g., Amin et al., 2024; Bazzi et al., 2021; Demarez et al., 2019; Pageot et al., 2020). Yet, such approaches often depend on expensive field validation data, hindering robust and transferable model development.
Building on the 2026 IRRIBIO project—which compared irrigation practices between organic and conventional farms in Tarn and Essonne—this internship continues promising work within the emerging CARTIRRI project. The goal is to develop a novel methodology integrating administrative, agricultural, and remote sensing data to map irrigation without requiring field validation data.
Key Tasks:
Develop and assess a method for mapping irrigated parcels by combining data from the Agricultural Census (RA), the Graphical Parcellary Register (RPG), water abstraction records, and Sentinel-1/Sentinel-2 time series.
Use high-certainty irrigated/non-irrigated parcels identified via RA2020/RPG2020 cross-referencing as reference data to train and validate machine learning models (e.g., Random Forest, XGBoost, 1D-CNN).
Apply the method to the Tarn department as a proof of concept, extending existing departmental maps.
Explore the potential for culture-specific irrigation mapping.
Evaluate the spatial extendability of the method across Occitanie by testing model transferability to other departments.
Location:
Toulouse or Montpellier (based on candidate’s location).
Required Education & Skills
Bachelor’s/Master’s level (Bac+3 to +5)
Background in remote sensing, agro-environmental issues, and data science for heterogeneous data processing.
Strong interest in geospatial data analysis, satellite time series, and machine learning.
Recommended training: M2 or final year of engineering school in remote sensing, geomatics, data science, and/or AI, with specialization in geospatial data analysis, satellite time series processing, and machine learning.
Desired Competencies:
Rigor, organization, autonomy, active listening, and strong oral/written communication skills.
Knowledge in: remote sensing, data science, machine learning, and Python-based data processing.
Quality of Life at INRAE
As a public research establishment, INRAE offers (depending on contract type and duration):
Up to 30 days of annual leave + 15 RTT days (for full-time)
Skill development: training, career counseling
50% coverage of collective health insurance contributions
Parental support: childcare vouchers (CESU), leisure benefits
Social support: advice, listening, social aid and loans
Vacation & leisure benefits: holiday vouchers, preferential accommodation
Sports and cultural activities
Collective catering
Application Process
Submit your CV and motivation letter.
Note: As a public research establishment, INRAE staff are subject to civil service regulations, including neutrality and secularism (laïcité). Staff must refrain from expressing religious, philosophical, or political convictions through behavior or attire in the exercise of their duties—whether or not in public-facing roles.
Learn more: fonction publique.gouv.fr (https://www.fonction-publique.gouv.fr)
domain
INRAE
Leading-edge innovations and technical excellence in the geospatial domain.