Recent advancements of cloud-based global crop monitoring system （CropWatch）

Abstract

Agricultural condition monitoring is a critical link in the food supply chain, and its accuracy and reliability are vital for agricultural production management, macro-control of grain, market stability, and sustainable development. Global agricultural remote sensing monitoring plays a vital role in ensuring food security, which is especially prominent for China, the world’s largest food-importing country. Remote sensing has become an important means of agricultural condition monitoring. because it can provide extensive, high-frequency, and timely monitoring data. However, due to various factors, global agricultural condition monitoring still faces many challenges. The global agricultural monitoring system, CropWatch, is a monitoring system driven by remote sensing indicators and ground-measured data, providing independent global agricultural condition information, independent of statistical data, which has been stably operating for over 27 years, has released 135 agricultural monitoring reports to date, providing agricultural information services to 173 countries and regions worldwide. This article details the latest developments of CropWatch over the past five years, forming a monitoring system composed of 54 indicators, adding cultivation and early warning indicators, and highlighting monitoring and early warning functions. Using the advanced machine learning methods, it has constructed a foundation of basic data such as“farmland (field), irrigation, cropping intensity, terraces, shelterbelts,”achieving independence in basic data. The use of large language models has significantly improved the efficiency and objectivity of agricultural condition indicator analysis. In the future, a more open agricultural observation platform will be built, supported by large language models and open APIs, to lower the barriers to use. With its unique design philosophy and operational mechanism, the CropWatch agricultural monitoring system stands out in the international agricultural monitoring field and sets an example for the operationalization of remote sensing monitoring. After seven versions of updates, the current CropWatch has achieved cloud-based and service-oriented operations. It can serve all interested users or stakeholders, enabling them to customize the monitoring system and conduct independent monitoring based on their own agricultural characteristics and needs, thereby enhancing their autonomy in monitoring. Through capacity building and the empowerment of independent agricultural monitoring, users are enabled to independently undertake the construction of agricultural monitoring systems for specific regions, as well as the production of agricultural products, collaborative analysis of information, and dissemination of results. This customized solution not only reduces users’dependence on expensive computing infrastructure and storage devices but also, by opening up model functionalities, allows users to independently verify and calibrate models within a user-friendly interface. This reduces technical barriers and reliance on external information, enhancing the flexibility and operability of monitoring and providing a more flexible and efficient solution for agricultural monitoring. This has revolutionized the traditional model of agricultural monitoring system development and has promoted the capacity for agricultural monitoring in developing countries. Looking to the future, CropWatch will further build a more open agricultural monitoring platform, strengthen the protection of data privacy and sovereignty, and with the support of large language models and open APIs, create an efficient, reliable, and user-friendly agricultural monitoring system. This will serve global agricultural producers and decision-makers, provide strong support for global food security and sustainable development, and promote the popularization and in-depth application of agricultural monitoring technologies.