Funded by: H. Hovnanian Family Foundation
PI: Habet Madoyan
Researchers: Vahe Movsisyan, Gayane Shahnazaryan, Anna Zatikyan, Aleksandr Arakelyan
Â
Project description
This project represents the successful completion of a comprehensive research and capacity-building initiative aimed at advancing environmental data science and hydrometeorological analytics in Armenia. The work was carried out by a dedicated research team at the American University of Armenia (AUA), in close collaboration with the Hydrometeorology and Monitoring Center of the Republic of Armenia.
A critical foundation of this project was the earlier support from the Afeyan Foundation, which enabled the initial formation of the research team. That grant served as the inception point for building a multidisciplinary group combining expertise in data science, environmental analytics, and applied statistics. The current project builds directly on that foundation, demonstrating the continuity and scalability of the team’s research capabilities. Over time, the team has evolved into a stable and productive research unit capable of delivering both scientific outputs and practical solutions for national institutions.
The project delivered three major components. First, advanced forecasting models for spring flood characteristics were developed, replacing legacy Excel-based approaches with robust AI models. These models enable automated forecasting of peak river discharges, as well as the timing of flood onset and duration. A fully functional analytical platform was developed, allowing specialists to upload data, run forecasts, visualize results, and generate reports in a streamlined and reproducible manner. This significantly modernizes the forecasting workflow and enhances the reliability and usability of outputs for decision-making.
Second, the project strengthened institutional capacity through targeted, hands-on training. A structured training program was delivered to Hydrometeorology and Monitoring Center staff on processing climate observation data. Participants gained practical skills in data cleaning, homogenization, and analysis of precipitation, temperature, and wind data, supported by pre-developed scripts and real national datasets. This component ensured that the developed methodologies are not only implemented but also sustained within the institution.
Third, the project initiated the development of an automated air quality dashboard, designed to integrate real-time monitoring data and provide dynamic visualization and reporting capabilities. The dashboard enables continuous tracking of key air pollutants and supports multi-level analysis through interactive visualizations and automated reporting. This work contributes to the broader goal of improving environmental data accessibility, strengthening analytical workflows, and supporting evidence-based policymaking in the area of air quality management.
Beyond technical outputs, the project contributed to scientific advancement and international collaboration. The research team expanded to include internationally recognized experts from the Helmholtz-Centre for Environmental Research, strengthening the scientific rigor of the work. Project results have been prepared for academic dissemination, with conference submissions and a peer-reviewed manuscript produced as part of the final outputs.