Global agriculture is in a difficult place. Production continues to climb, but the foundation that supports it is showing clear signs of stress. An FAO scientific assessment underscores this shift. Inputs are being pushed harder, with fertilizer use up about 40 percent and pesticide use doubled, and over 40 percent of rural populations now living in river basins where water scarcity is already a defining constraint.
Agriculture trade has expanded just as quickly. In the past five decades, the value of food and agricultural commodities has grown eightfold, binding national food systems more closely together. Even so, food insecurity persists. Hunger is increasing again, and more than 3 billion people cannot afford a healthy diet.
One of the central themes in the FAO white paper, Update on scientific findings on the interactions between agriculture, food systems and climate change, is the pace and character of environmental change.
Much of this change builds slowly and is easy to miss in a single season. Ocean warming has already lowered the maximum sustainable yield of key fish populations by a little over 4 percent in the last 80 years. Soil carbon also doesn’t follow one pattern; it changes with climate, soils and how land is managed. Coastal wetlands face rising seas, heavier storms, changing rainfall and steady development.
Agriculture fits into this picture in a complicated way. It relies heavily on freshwater more than any other sector and it also accounts for over a quarter of the energy used in food production. Growing populations and changing diets place additional pressure on the resources agriculture depends on, and that pressure builds gradually rather than all at once.
The FAO does not state this outright, yet the implication is clear: if agriculture is becoming more exposed, then visibility into that exposure must be continual.
The FAO assessment highlights a quiet but significant shift. The use of satellite data in national monitoring systems has expanded steadily. For many countries it is now the most consistent way to see how landscapes are evolving. These datasets support a wide range of national monitoring tasks, be it mapping land cover, tracking forest loss, estimating soil and biomass carbon, or watching changes in mangroves, croplands and coastal wetlands.
Slow shifts rarely show up from the ground. Regular satellite records give the seasonal and multi-year view needed to follow how coastlines, vegetation and restoration areas evolve.
Where remote sensing offers consistency, artificial intelligence offers scale. The FAO assessment points to rapid growth in machine-learning applications that classify crops, estimate yields, identify degraded land, infer farming practices and detect stress signals before they appear visibly on the ground.
These models allow researchers and governments to analyse decades of imagery, integrate multiple sensors and pick up emerging patterns that would be impossible to detect manually.
But the assessment also makes an important distinction. AI is not a shortcut; it is an amplifier, entirely dependent on the quality and stability of the underlying data.
FAO points out that many countries still don’t have the kind of consistent, well-calibrated agricultural data, including a mix of climate, soil conditions, biological cycles and management choices, needed for reliable analysis. Without that foundation, AI tools don’t perform as reliably as people hope.
Limited connectivity and computing capacity also create further obstacles, making advanced analytics difficult to scale.
Where stronger datasets do exist, AI can do considerably more. It can draw on long records and in-season observations to spot emerging stress, understand crop behaviour and support decision-support tools that link environmental conditions with management options.
Across chapters, the report makes one point repeatedly: environmental change doesn’t pause, and food systems are shaped by it constantly. Planning cannot rely on occasional snapshots.
FAO estimates that roughly 13 percent of food is lost between harvest and retail, much of it for reasons that only become clear when conditions are followed across seasons. That kind of visibility requires consistent records.
Remote sensing provides the long view. AI helps make sense of it. Field experience keeps it grounded. All three are part of the measurement base agriculture now depends on.
Agriculture has been central to EarthDaily’s work from the beginning, and the direction outlined in the FAO assessment closely matches the measurement principles we build around. The report highlights the need for consistent, calibrated, long-horizon visibility into land and crop conditions.
EarthDaily’s systems are designed for that kind of continuity and for making environmental change easier to interpret at field, regional and global scales.
Across these applications, the idea is the same: consistent satellite data, strengthened by analytics, creates the level of visibility the FAO identifies as essential for resilience.
FAO doesn’t present technology as the fix, but as the means to see what needs fixing. With environmental pressures rising, that steady view is becoming as essential as production itself.
The assessment points to a simple conclusion: agriculture resilience will depend on intelligence systems that can show how land, water and ecosystems are changing.