| Sumario: | Agricultural decision making in Ethiopia is highly influenced by climate variability, however the effective use of agricultural data remains limited across meteorological, soil, and crop management domains. Differences in formats, spatial and temporal scales, and metadata standards reduce the usability of these datasets for digital and AI-based agro-advisory systems. This study presents a practical framework for standardizing and Harmonizing multi-source agricultural data to support the Intelligent Agricultural Systems Advisory Tool (ISAT) in Ethiopia. Building on pilot implementations in India and parts of Africa, this study examines the transition of ISAT toward an AI-enabled advisory system. The analysis emphasizes data corpus standardization for AI readiness and assesses platform flexibility and suitability across diverse agro-ecological contexts. Climate, soil, and crop management datasets were standardized using common units, aligned spatial and temporal scales, and crop-stage-based structuring, following FAIR data principles. The harmonized datasets were integrated into the ISAT framework by linking climate, soil, and crop information with advisory generation processes. This integration enables location-specific advisories, including recommendations on sowing, nutrient and water management, and climate risk mitigation. Overall, the study demonstrates how systematic data harmonization and integration can enhance data readiness and provide a scalable foundation for AI-driven agro-advisory services in data-constrained contexts.
|
|---|