Organizational Setting

The Agrifood Economics and Policy Division (ESA) conducts economic research and policy analysis to support the transformation to more efficient, inclusive, resilient and sustainable agrifood systems for better production, better nutrition, a better environment, and a better life, leaving no one behind. ESA provides evidence-based support to national, regional and global policy processes and initiatives related to monitoring and analysing food and agricultural policies, agribusiness and value chain development, rural transformation and poverty, food security and nutrition information and analysis, resilience, bioeconomy, and climate-smart agriculture. The division also leads the production of two FAO flagship publications: The state of food and agriculture (SOFA) and The state of food security and nutrition in the World (SOFI) and leads the technical work on the FAO’s global roadmap to Achieve SDG2 without breaching the 1.5C target.

The Agrifood Economics and Policy Division (ESA) is providing technical assistance to countries on the use of Earth Observations data to analyse land productivity with a focus on monitoring planted area, harvested area and crop yield. The final goal is to enhance better production and better nutrition in countries in line with FAO’s strategic objectives.

Reporting Lines

The selected candidate will work under the direct supervision of the Technical Adviser Geospatial in the Agrifood Economics and Policy Division (ESA) and in coordination with the Data Lab, CSI, and the technical divisions working with geospatial data.

Technical Focus

Provision of technical support for the development of crop type maps and crop yield maps at the national level including (i) implementation of EOSTAT project in selected countries, with a focus on i) installation and maintenance of Sen2Agri toolbox deployed on the cloud or on-premises, ii) development of algorithms for the classification of crops and land cover in general using optical and SAR data, iii) make use of both pixel and object based approaches, iv) set up data pipelines in frequently used cloud environments (Google, AWS, Sentinel Hub), v) development of web apps iv) support the development of Geospatial Data Science training programs.

Tasks and responsibilities

•    Preprocess Earth Observation (EO) time series and develop regularized EO data cubes.
•    Access and use both Optical and SAR data, at high and very high resolution from multiple sources
•    Identify phenological, spectral and texture indicators for monitoring land cover and land use, with a focus on crop types
•    Assess crop yield using EO data and in situ data, adopting both statistical and physical-based models (e.g. DSSAT, SALUS etc).
•    Develop and use ML and AI classification algorithms, using both pixel and object analysis approaches
•    Identify and analyze data requirements for national crop type mapping and extraction of agricultural statistics.
•    Perform validation of map products for location accuracy and area statistics bias.
•    API Integration: Integrate and utilize common EO data APIs such as Sentinel Hub, AWS Earth, Google Earth Engine, and OPENEO to access satellite imagery, environmental datasets, and ancillary data.

CANDIDATES WILL BE ASSESSED AGAINST THE FOLLOWING

Minimum Requirements

•    Bachelor  degree in Environmental Science, Data Science, Informatics, or related fields;
•    Minimum of 2 years for category C, 5 years for category B, 10 years for COF category A, and 10 years for PSA category A of relevant experience in the.EO Data Analysis;
•    Working knowledge of English.

FAO Core Competencies

•    Results Focus
•    Teamwork
•    Communication
•    Building Effective Relationships
•    Knowledge Sharing and Continuous Improvement

Selection Criteria

Extent and relevance of work experience in:
•    Two or more programming languages including Python, R and Javascript.
•    Coding in GEE, Jupyter Notebook, and Colab.
•    Machine Learning and Deep Learning frameworks and packages applied to satellite image classification in the domain of land cover and crop mapping (e.g Tensorflow, Keras, PyTorch, Scikit-Learn, ts-learn, etc).
•    Satellite time series pre-processing and analysis.
•    Ability to work independently, with minimum supervision.
•    Advanced knowledge of Spanish and French would be an asset.