CLOUD-BASED MODULAR SYSTEM FOR ACQUISITION AND VISUALIZATION OF OCO-2 REMOTELY SENSED CO<sub>2</sub> DATA

In this paper is presented a cloud-based framework for automated acquisition and visualization of Orbiting Carbon Observatory-2 (OCO-2) satellite CO₂ data. The system employs an ETL pipeline with OPeNDAP protocol for selective data retrieval, reducing network overhead while processing L2 Standard and L2 Lite FP products. Built on Amazon Web Services (AWS) infrastructure, using Python (Pandas, Dash, Plotly) and Docker orchestration, the modular architecture implements dependency injection for runtime flexibility. The deployed system achieves daily automated ingestion with 2.25 km 1.29 km spatial resolution, enabling the real-time monitoring through interactive web visualization. The system is designed as a foundation for future analytical research, providing ready integration points for machine learning models to perform advanced CO₂ pattern recognition and predictive analysis.