Consultant - Data scientist

Result of Service 1. The existing method for opium poppy yield estimates based on visual ranking of satellite images is improved with 2021 data and applied to estimate 2022 opium poppy yield in Afghanistan. 2. A new, effective and usable model for opium poppy yield prediction in Afghanistan, based on remote sensing and other spatial explicit information.

Work Location Home-based

Expected duration 30 Working days

Duties and Responsibilities UNODC’s Research and Trend Analysis Branch provides technical support on drugs and crime surveys to member states and counterparts and implements several surveys each year. Experts working in the branch are regularly involved in sample design and survey methodology development, data analysis as well as assessing the quality of surveys for which data is reported to UNODC. The Illicit Crop Monitoring Programme (ICMP) provides Member States with the necessary assistance to compile reliable and international comparable data on illicit crop cultivation and related drug production. In Afghanistan, studies on poppy yield and potential opium gum production are realized on annual basis, through the implementation of campaign for field data collection. Field data collection is extremely challenging in Afghanistan, because of the security situation of the country. Therefore, ICMP aims at predicting the poppy yield based on remote sensing and other available spatially explicit data.

The purpose of the consultancy is twofold: i) review and improve the currently available model applied to yield prediction, which is based on visual ranking of fields in satellite images, and ii) develop a new model for opium poppy yield prediction in Afghanistan, which integrates data obtained from multi-spectral satellite imagery and available ancillary data.

Under the overall guidance and supervision of the Geospatial Information Officer of the Programme Development and Management Unit (PDMU), The consultant is expected to carry out the following activities: 1. Review and organize historical yield data collected in the field. 2. Review and improve the current yield prediction methods (based on visual ranking of fields in satellite imagery). 3. Calculate the 2022 opium poppy yield based on the improved existing method. 4. Take stock of the available multispectral images and ancillary spatially explicit information. 5. Considering the available data, test different approach to identify the most accurate model for opium poppy yield prediction. 6. Train and tune the model using historical and ancillary data. 7. Produce a comprehensive report, describing in detail the applied methods and tools, with a thorough assessment of performances and prediction accuracy of the model. 8. Assure that the model can be replicated by ICMP technical staff.

The consultant will work in close cooperation with technical staff in PDMU.

Qualifications/special skills Academic Qualifications: An advanced university (Master's degree or equivalent) in the fields of data science, statistics, or related studies is required. A first level university degree in similar fields in combination with two additional years of qualifying experience may be accepted in lieu of the advanced university degree. PhD in a relevant area is desirable.

Experience: - A minimum of 5 years of experience in statistical modelling using multivariate analysis is required. - Research experience with yield prediction models is desirable. - Relevant scientific publications (books or articles in peer-reviewed journals) and/or conference papers are required. - Proven experience in AI, data mining, Deep Learning techniques and Big Data is desirable.

English and French are the working languages of the United Nations Secretariat. For this position, fluency in oral and written English is required. Knowledge of another United Nations Secretariat language is and advantage.

No Fee THE UNITED NATIONS DOES NOT CHARGE A FEE AT ANY STAGE OF THE RECRUITMENT PROCESS (APPLICATION, INTERVIEW MEETING, PROCESSING, OR TRAINING). THE UNITED NATIONS DOES NOT CONCERN ITSELF WITH INFORMATION ON APPLICANTS’ BANK ACCOUNTS.