Human soft skills, the key to using agricultural satellite data

Extreme weather events, population growth, and resource constraints are putting unprecedented pressure on agriculture. To respond, farmers around the world need not just tools and inputs, but timely, actionable information. Harnessing better data—from weather forecasts and soil health to market trends and water availability—can help farmers anticipate shocks, optimize resources, and make more informed decisions that reduce risk and improve the resilience of crops.

Few tools are as powerful as satellite data—enabling precision agriculture, driving early warning systems, and guiding smarter allocation of resources. The agricultural satellite imaging market topped $4 billion last year and is expected to grow to over $14 billion by 2030, with annual growth of close to 20 percent.

This data typically includes imagery and sensor readings that capture environmental conditions, land use, and atmospheric patterns. Based on the reflection of the sun’s light on crops for example, satellite data can determine the state of different types of crops and predict their yields. 

Crucially, the rise of openly accessible satellite data has been a game-changer—making real-time monitoring vastly more affordable and within reach for even the most resource-constrained countries.

For example, NASA and the European Space Agency provide free, high-quality data that can be useful for a wide range of stakeholders – from governments, regional institutions, to young farmers and agricultural analysts. WorldCover v2 (Figure 1) provides detailed land cover data to support water management, monitor crop health, and guide agroforestry efforts.

In many cases, lack of technology infrastructure and data availability are no longer a problem. The biggest barrier now is the lack of skills to process, interpret, and apply that data effectively. That includes coding, geographic information systems, and leveraging open access satellite imagery to produce information relevant for agricultural monitoring. 

Figure 1. Global Land Cover Map from WorldCover V2(2021) (Data source – ESA)

Artificial Intelligence as a solution for data processing and management

To support data processing and analysis, technicians can now tap into AI-generated code from platforms offered by companies like Google and Microsoft, reducing the time and expertise needed to generate insights. 

For instance, user-friendly monitoring platforms such as Crop Watch Cloud provide timely information on agricultural conditions. This platform is customized to agricultural monitoring and includes modules allowing users to customize the information they need, like field boundaries, then track crop conditions throughout the growing season. 

Figure 2. CropWatch Cloud Platform

The World Bank provides trainings in how to use these platforms. This includes a capacity-building program on the CropWatch Cloud platform for members of Ethiopia’s Agricultural Transformation Institute and the Centre for Geospatial Applications in Rural Development. This training, funded by the China-World Bank Group Partnership Facility, taught participants the intricacies of advanced analytics techniques like field delineation, crop-type mapping, and yield estimation. The Bank also supported the East Africa Learning Exchange, organized by the Group on Earth Observation and Global Agriculture Monitoring (GEOGLAM) and the IGAD Climate Prediction and Applications Center (ICPAC). The learning forum explored crop monitoring tools, data platforms, and assessed the importance of collaboration to improve food security and enable data-driven agricultural decisions.

Local data still crucial

Even with the abundant availability of public data, country-generated data continues to be important for agricultural monitoring. Local stakeholders have a better understanding of the nuances of their agricultural systems, such as crop types, seasonal variations, and farming practices, which are essential for calibrating and validating satellite-derived models. 

Moreover, when governments and local institutions help generate the data, they’re more likely to trust and use it. Local ownership not only builds skills and strengthens institutions, but also makes monitoring efforts more sustainable. The key is to combine global datasets with locally sourced data for the greatest impact.

The tools to transform agriculture are already within reach. With the right skills, data-driven farming could truly soar.

Source : World Bank