국내기사
국제 사례 비교를 통한 한국형 위성영상 기반 작황 모니터링 시스템 설계 방안 = Designing a Korean satellite imagery-based agricultural monitoring system through comparative analysis of international approaches
Recent climate change, extreme weather events, and global supply chain disruptions have intensified price volatility in the international grain market and heightened uncertainty in food security. For South Korea, with its low grain self-sufficiency rate, such external shocks can directly lead to food security instability. This underscores the urgent need for a systematic information system capable of providing timely assessments of crop conditions in major grain-producing countries and integrating this information into policy decision-making. This study aims to develop a system design framework tailored to Korea's circumstances by analyzing and comparing existing satellite-based crop monitoring systems in operation abroad, namely: Crop Explorer, Monitoring Agricultural ResourceS (MARS), the Global Information and Early Warning System (GIEWS), and CropWatch. The analysis showed that all four systems share the common feature of integrating satellite-derived vegetation indices with meteorological data to assess crop conditions and yields. At the same time, their variations in scope, objectives, and delivery methods illustrate how system designs are adapted to specific regional and policy needs. These differences provide practical insights for developing a framework suited to Korea's agricultural context and policy priorities. The Crop Explorer provides an open platform that integrates satellite and meteorological data with periodic updates. The MARS combines vegetation indices with crop growth models to deliver precision yield forecasts designed to support EU agricultural policies. The GIEWS emphasizes early warning of food crises through the Agricultural Stress Index (ASI), whereas CropWatch employs multi-layered spatial analysis and automated data processing to monitor crop conditions both in China and globally. Based on these findings, this study proposes a system design for Korea centered on five core components: (1) constructing high-resolution cropland maps using deep learning techniques, (2) employing advanced vegetation indices that capture crop physiological processes, (3) integrating reanalyzed meteorological datasets, (4) establishing automated large-scale data processing pipelines, and (5) providing a user-friendly web-based visualization platform. The proposed design is expected to serve as an evidence-based decision-support tool, enhancing the accuracy and timeliness of crop yield information, and thereby contributing to food supply-demand policies, disaster response, and the formulation of mid- to long-term agricultural strategies.