National Crop and Statistical Modelling Expert at Food and… · Sri…

IMPORTANT NOTICE: Please note that Closure Date and Time displayed above are based on date and time settings of your personal device

FAO seeks gender, geographical and linguistic diversity in order to best serve FAO Members in all regions.

• FAO is committed to achieving workforce diversity in terms of gender, nationality, background and culture
• Qualified female applicants, qualified nationals of non-and under-represented Members and person with disabilities are encouraged to apply
• Everyone who works for FAO is required to adhere to the highest standards of integrity and professional conduct, and to uphold FAO's values
• FAO, as a Specialized Agency of the United Nations, has a zero-tolerance policy for conduct that is incompatible with its status, objectives and mandate, including sexual exploitation and abuse, sexual harassment, abuse of authority and discrimination
• All selected candidates will undergo rigorous reference and background checks
• All applications will be treated with the strictest confidentiality

Organizational Setting

The Food and Agriculture Organization (FAO) is a specialized agency of the United Nations that leads international efforts to defeat hunger. FAO's goal is to achieve food security for all and ensure that people have regular access to enough high-quality food to lead active, healthy lives. With 195 members - 194 countries and the European Union, FAO works in over 130 countries worldwide and has started working in Sri Lanka since 1979 with its Representation (Country Office) opened in Colombo.

The FAO contributes to the achievement of the 2030 Agenda through its Strategic Framework by supporting the transformation to more efficient, inclusive, resilient and sustainable agri-food systems, and through four Betters (Better production, Better nutrition, a Better environment and a Better life) - leaving no one behind.

Reporting Lines

The National Crop and Statistical Modelling Expert will work under the overall supervision of the FAO Representative for Sri Lanka and the Maldives, the direct supervision of the Assistant FAO Representative (Programmes), and technical guidance of the Lead Technical Officer of the project. The incumbent will work under the supervision and leadership of the Socio Economic and Planning Centre (SEPC), in coordination with the TCP Operations Specialist, Digital Agriculture Implementation Specialist, relevant government institutions and technical stakeholders to support the effective implementation, coordination, validation, and institutionalization of the CROPIX-AI initiative.

Technical Focus

The objective of this assignment is to provide technical expertise for the design, development, calibration, validation, and operationalization of paddy and maize crop growth, yield and production forecasting models, including agricultural data harmonization, statistical and AI-based modelling, crop damage assessment, forecast adjustment, and capacity building, supporting the integration and continuous improvement of accurate, timely and spatially disaggregated forecasts within CROPIX-AI under TCP/SRL/4107.

Tasks and responsibilities

The expert will provide technical support for the tasks below under SEPC’s supervision and leadership, following an agreed workplan and methodologies approved by SEPC in consultation with the FAO Lead Technical Officer.

I. Agricultural Data Assessment and Modelling Dataset Development

• Review and assess paddy and maize datasets available from DOA, DCS, DAD, DoM, Mahaweli Authority, NRMC and other relevant institutions for their suitability for model development and forecasting.
• Assess data completeness, consistency, temporal and spatial coverage, definitions, classifications, reporting frequencies and methodological differences across datasets.
• Identify and address missing values, outliers, inconsistencies, structural breaks and other data-quality issues that may affect model performance.
• Support the development and maintenance of modelling-ready datasets covering historical production, cultivated area, yield, weather, soil, fertilizer use, irrigation, crop varieties, crop management practices, and other relevant variables..
• Harmonize crop classifications, geographic references, agro-ecological zones, cropping seasons and key modelling variables.
• Support the development and maintenance of a data dictionary and metadata framework for datasets used in model development, calibration, and validation..
• Ensure datasets used for model training, testing and validation are appropriately documented, traceable and version-controlled.

II. Crop Model Design and Development

• Assist with the technical and operational tasks involved in designing and developing paddy and maize crop growth and yield forecasting models suited to Sri Lankan agro-ecological zones, climatic conditions, cropping seasons, and production systems, under the guidance of the supervisor. Support the development of model architectures, parameters, variables, and forecasting methodologies for estimating crop growth, yield, and total production..
• Assist with the technical and operational tasks required to generate crop forecasts at appropriate growth stages and spatial scales, including agro-ecological zones, irrigation systems, districts, and national levels, subject to data availability and under the guidance of the supervisor..
• Support the assessment and application of appropriate modelling methodologies, including statistical, econometric, process-based crop simulation, machine-learning, artificial intelligence, remote-sensing, and hybrid modelling approaches..
• Support the development of hybrid and/or ensemble modelling approaches that combine process-based crop simulation, statistical models, and AI/ML techniques to improve forecast accuracy and robustness..
• Assist with compiling and incorporating key yield-determining variables, including weather, soil, crop varieties, growth stages, planting dates, cultivated area, fertilizer availability and use, irrigation methods, crop management practices, historical yields, crop conditions, crop damage, and other relevant factors, under the guidance of the supervisor. .
• Support the development of growth-stage-specific modelling approaches to assess the effects of climatic and agronomic conditions on crop development and final yield.
• Assist with the compilation, integration, and processing of historical yield trends, current-season crop conditions, weather information, crop calendars, cultivated-area estimates, remote-sensing data, and other relevant datasets to support the preparation of operational production forecasts, under the guidance of the supervisor. .
• Support feature engineering, feature selection, parameter estimation, model optimization, and sensitivity analysis to improve model performance and robustness..
• Support the development of methodologies for incorporating biweekly updates on cultivated and harvested areas into yield and production forecasts.

III. Model Calibration, Validation and Performance Optimization

• Support SEPC in the calibration, testing, validation and optimization of developed paddy and maize forecasting models using historical and current-season datasets.
• Develop and apply robust validation methodologies using DCS official production statistics, DOA datasets, crop-cut and field observations, historical crop performance, weather observations, remote-sensing information and other verified ground observations.
• Establish appropriate training, testing and validation datasets covering multiple cropping seasons and major agro-ecological and production zones.
• Conduct back-testing and hindcasting to evaluate model performance using historical seasons and information available at relevant forecast stages.
• Assess model performance at national, district and agro-ecological-zone levels, where data permit.
• Compare alternative model specifications and identify sources of forecast error, model instability, uncertainty and data limitations.
• Continuously refine, recalibrate and retrain models as new agricultural, weather, remote-sensing and field data become available.

IV. Crop Damage, Climate Risk and Forecast Adjustment

• Support the design and development of modelling approaches to assess the effects of floods, droughts, heat stress, excessive rainfall, wind, pests, diseases, lodging, and nutrient deficiencies on paddy and maize growth and yield..
• Assess the effects of climatic and non-climatic stresses at different crop growth stages and identify relevant variables for incorporation into forecasting models.
• Support in developing methodologies for incorporating validated crop damage observations and quantified loss estimates into production forecasting models.
• Conduct historical back-testing of damage-adjusted forecasts to assess their effectiveness in improving forecast accuracy.
• Provide technical recommendations for adjusting production forecasts based on verified crop damage and loss information.

V. Multi-Source Data Integration for Crop Modelling

• Define technical requirements for integrating statistical, field, crop-cut, weather, soil, fertilizer, crop registry, remote-sensing, irrigation and crop management datasets into the forecasting workflow.
• Assist in coordinating the collection and organization of data from relevant institutions and technical service providers, including supporting the definition of data structures, interfaces, validation rules, and update frequencies required for model execution, under the guidance of the supervisor..
• Support the establishment of reliable data pipelines and quality-control procedures for recurring ingestion and updating of forecasting datasets.
• Ensure all model inputs are traceable to their source datasets and observation periods.
• Support in documenting the timing, frequency, and information requirements for generating forecasts during each cropping season, in accordance with established procedures and under the guidance of the supervisor.
• Monitor forecast performance as new crop, weather, remote-sensing and field observations become available.
• Compare successive forecasts with observed crop conditions and official production statistics to identify forecast deviations.
• Identify circumstances requiring model recalibration, retraining, updating or forecast adjustment.
• Assist in implementing established procedures for monitoring model performance and documenting areas for potential improvement, under the guidance of the supervisor. Assist in compiling and reviewing model performance data across seasons and climatic conditions, and document observations for consideration by the supervisor regarding potential improvements to model structure, variables, parameters, and algorithms.Maintain a continuous model development cycle incorporating lessons and new evidence from each cropping season.

VI. Technical Validation, Stakeholder Engagement and Capacity Building

• Support coordination with DOA technical divisions, DCS, DAD, DoM, the Mahaweli Authority, NRMC, and other relevant institutions to obtain technical inputs and validate datasets, modelling assumptions, and forecast outputs.Provide technical inputs to SEPC-led technical discussions relating to crop model design, development, calibration, validation, model performance and forecasting methodologies.
• Participate in technical consultations, validation workshops and review sessions to provide technical inputs for institutional consideration of modelling approaches, datasets and forecast outputs.
• Support SEPC in delivering practical training and knowledge transfer to relevant technical officers on crop modelling, yield forecasting, statistical modelling, model development, calibration and validation, forecast interpretation and the effective use of CROPIX-AI forecasting outputs.
• Support the development of Standard Operating Procedures for data exchange, model execution, model updating, forecast generation and validation.
• Contribute to the CROPIX Technical Working Group and related institutional mechanisms established under the project.

VII. Technical Documentation and Reporting

• Prepare comprehensive technical documentation covering model architecture, modelling methodologies, algorithms, datasets, variables, assumptions, parameters, calibration procedures, validation methods, results and limitations.
• Prepare detailed paddy and maize model development, calibration and validation reports.
• Maintain records of model versions, training and validation datasets, parameters, model configurations and performance results.
• Document model changes, recalibration, retraining and methodological improvements to ensure reproducibility and transparency.
• Provide technical recommendations for operational deployment, continuous improvement and future scaling of the crop forecasting system.
• Provide periodic technical progress updates and inputs required by the FAO Project Task Force and the Department of Agriculture.
• Contribute technical content to project reports, manuals, SOPs, training materials and knowledge products related to crop modelling, yield forecasting and agricultural data integration.

CANDIDATES WILL BE ASSESSED AGAINST THE FOLLOWING

Minimum Requirements

• Advanced university degree (Master’s degree or higher) in Agricultural Sciences, Agronomy, Crop Science, Agricultural/Biosystems Engineering, Statistics, Agricultural Economics, Data Science, Applied Mathematics, Computer Science, Remote Sensing, Environmental Science or another closely related field.
• Minimum of 8 years of relevant professional experience gained in crop modelling, agricultural forecasting, agricultural statistics, or a closely related discipline.
• Working knowledge of English (C1). Working knowledge of Sinhala and/or Tamil is required.
• National of Sri Lanka or resident of the country with a valid work permit.

FAO Core Competencies

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

Selection criteria

• Demonstrated experience in designing, calibrating and validating crop growth and yield forecasting models (statistical, process-based, machine-learning/AI, remote-sensing-based or hybrid), preferably for paddy, maize or other major staple crops.
• Proven experience in the integration and harmonization of multi-source agricultural, climatic, remote-sensing and statistical datasets, including data-quality assessment and management of large, heterogeneous datasets.
• Working experience with statistical and machine-learning software/programming languages such as R, Python, and relevant crop-simulation or modelling platforms (e.g. DSSAT, APSIM, or equivalent), including model calibration, back-testing and performance evaluation (RMSE, MAE, MAPE, bias, R²).

Please note that all candidates should adhere to FAO Values of Commitment to FAO, Respect for All and Integrity and Transparency.

FAO’s commitment to environmental sustainability is integral to our strategic objectives and operations.

ADDITIONAL INFORMATION

• FAO does not charge a fee at any stage of the recruitment process (application, interview meeting, processing).
• Applications received after the closing date will not be accepted.
• Please note that FAO only considers higher educational qualifications obtained from an institution accredited/recognized in the World Higher Education Database (WHED), a list updated by the International Association of Universities (IAU) / United Nations Educational, Scientific and Cultural Organization (UNESCO). The list can be accessed at http://www.whed.net/
• For other issues, visit the FAO employment website: http://www.fao.org/employment/home/en/
• Appointment will be subject to certification that the candidate is medically fit for appointment, accreditation, any residency or visa requirements, and security clearances. Subject to certain exemptions, vaccination against COVID-19 will in principle be required. Vaccination status will be verified as part of the medical clearance process.

HOW TO APPLY

• To apply, visit the recruitment website at Jobs at FAO and complete your online profile. We strongly recommend that your profile is accurate, complete and includes your employment records, academic qualifications, and language skills
• Candidates are requested to attach a letter of motivation to the online profile
• Once your profile is completed, please apply, and submit your application

• Please note that FAO only considers higher educational qualifications obtained from an institution accredited/recognized in the World Higher Education Database (WHED), a list updated by the International Association of Universities (IAU) / United Nations Educational, Scientific and Cultural Organization (UNESCO). The list can be accessed at http://www.whed.net/. These qualifications should be in alignment with the International Standard Classification of Education (ISCED) mappings.
• Candidates may be requested to provide performance assessments and authorization to conduct verification checks of past and present work, character, education, military and police records to ascertain any and all information which may be pertinent to the employment qualifications
• Incomplete applications will not be considered
• Personal information provided on your application may be shared within FAO and with other companies acting on FAO’s behalf to provide employment support services such as pre-screening of applications, assessment tests, background checks and other related services. You will be asked to provide your consent before submitting your application. You may withdraw consent at any time, by withdrawing your application, in such case FAO will no longer be able to consider your application
• Only applications received through the FAO recruitment portal will be considered
• Your application will be screened based on the information provided in your online profile
• We encourage applicants to submit the application well before the deadline date.

If you need help or have queries, please create a one-time registration with FAO’s client support team for further assistance: https://fao.service-now.com/csp

FAO IS A NON-SMOKING ENVIRONMENT