Graduate Student – EO-Based and Knowledge-Guided ML Crop Forecasting - Research Affiliate Program

Duties
The successful applicant will work on AAFC’s national, multidisciplinary project aimed at developing a next-generation forecasting system for potato yield using a knowledge-guided machine learning (KGML) approach that integrates Earth Observation (EO) data, process-based crop modeling, and field-scale observations into a unified forecasting framework. The research will examine how environmental, soil, and management factors influence potato yield across multiple production systems and varieties.The student will contribute to developing and validating predictive models, with opportunities to gain experience in EO-based analyses and interpret field data collected from potato farms in Prince Edward Island and other regions. The student will complete this research as part of their MSc or PhD thesis, or as part of a last-year undergraduate summer research project, providing a strong foundation for future graduate studies.
The student will:
• Develop and train deep learning and KGML models to predict potato growth and yield using synthetic crop model outputs and field-scale observations.
• Fine-tune models with observed yield data stratified by production system, region, and irrigation type.
• Conduct statistical analyses to evaluate and validate model predictions.
• Gain exposure to EO-derived indicators provided by the EO team as model inputs.
• Work semi-independently to troubleshoot models, interpret results, and solve research problems.
• Interact with a multidisciplinary team of scientists, students, and research technicians across Canada.
• Review relevant scientific literature and interpret results in the context of Canadian potato production systems.
• Contribute to scientific reports and publications, and present findings in written and oral formats at conferences.
• Work in Dr. Morteza Mesbah’s data science lab, using modern computational resources for model development.
• Participate in optional field visits in PEI to observe potato production and data collection practices, and training in ML and EO-based methods.
• Follow best practices for research ethics, data management, and reproducibility.

Work environment
The student will work in a collaborative, interdisciplinary team with scientists from across Canada. The position is primarily office- and computational-based, with optional exposure to field data. The student will work semi-independently in a research environment and interact with colleagues from multiple disciplines, including students, research technicians, and scientists.AAFC is committed to diversity and inclusion. We have several networks dedicated to ensuring that the department continues to grow as an inclusive, accessible, respectful and diverse workplace. All employees are encouraged and welcomed to join the networks and participate in their activities and events.
• The Gender and Sexual Diversity Inclusiveness Network
• The Indigenous Network Circle
• The Managers' Community
• The Persons with Disabilities Network
• The Student Panel of Representatives for Orientation, Unity and Training
• The Visible Minorities Network
• The Women in Science, Technology, Engineering and Mathematics Network
• The Young Professional's Network

Intent of the process
This project is part of a national, interdisciplinary initiative led by AAFC to advance potato yield forecasting using knowledge-guided machine learning (KGML), process-based crop modeling, and Earth Observation (EO) data. The research generates new scientific knowledge and predictive tools to improve the accuracy, resolution, and operational use of potato yield forecasts, supporting growers and decision-makers across Canada.The position is primarily office- and computational-based, with opportunities to gain experience using field data collected from potato farms in PEI and other regions. The successful student will work semi-independently in a research environment and will interact with individuals from a wide range of backgrounds, including students, research technicians, and scientists from multiple disciplines.

Positions to be filled
1

Important messages
Candidates will be required to pay for their own travel related to assessment and successful candidates will be responsible for obtaining their own living accommodations.Successful completion of both a RAP work assignment and your educational program may lead to a temporary or permanent federal public service position for which you meet the merit criteria and conditions of employment.

---

Reliability Status security clearance - Each student hired through the Research Affiliate Program (RAP) must meet the security requirements of the position as a condition of employment and, therefore will be asked by the hiring organization to complete security-relevant documents.

---

Education:
The candidate must be currently enrolled, or enrolled by the date of appointment, in a master’s or doctoral program in a relevant discipline (e.g., Biological Sciences, Agronomy, Environmental Science, Geography, Engineering, Bioresource Engineering, Computer Science, Remote Sensing, Data Science, or a related field) at an accredited Canadian post-secondary institution. This includes students who have completed a bachelor’s degree and have been admitted to a graduate program starting in the upcoming academic year.
The program must include a research component as part of the curriculum.Note: Candidates must be recognized as having full-time student status at the institution where they are currently enrolled. Individuals pending acceptance or in the process of submitting an application are encouraged to apply. Proof of enrollment will be required prior to the start date.
The student is expected to relocate or work in Prince Edward Island (PEI) for the duration of the appointment.

Experience*:
Experience analyzing environmental or agricultural systems data using programming tools (e.g., Python, R, or GIS software), including spatial, temporal, or remotely sensed datasets.
Experience integrating data or methods from more than one domain (e.g., environmental science, agronomy, remote sensing, or machine learning) within an academic or research project.
Experience working on research projects, including following protocols, collecting, cleaning, and documenting data.
Experience writing reports, research summaries, or essays for academic or professional purposes.
Experience collaborating in multidisciplinary teams, including students, researchers, or external stakeholders.* In the context of student recruitment in the Federal Public Service, experience can be acquired through studies, work experience, or volunteer activities.

Knowledge:
Knowledge of plant biology, crop growth, and agroecosystems.
Knowledge of remote sensing, Earth Observation, or geospatial data processing.
Knowledge of machine learning or statistical modeling applied to environmental, agricultural, or geospatial data.
Knowledge of research methods, data collection, and data management.

Competencies :
Planning and organization skills.
Interactive communication.
Initiative and self-motivation.
Attention to detail.
Teamwork and collaboration.

Abilities:
Ability to analyze and interpret complex datasets.
Ability to apply computational and statistical methods to model crop growth and yield.
Ability to synthesize results and communicate findings to both technical and non-technical audiences.

---

Applied / assessed at a later date

English essential - You are entitled to participate in the selection process in the official language of your choice.

---

---

Selection may be limited to members of the following employment equity groups: Indigenous (Aboriginal) peoples, persons with disabilities, visible minorities, and women.

---

Preference will be given to veterans first and then to Canadian citizens and permanent residents, with the exception of a job located in Nunavut, where Nunavut Inuit will be appointed first.

Learn more about preference to veterans.

---

---

---

---