Traditional yield forecasting relies on detailed biophysical crop simulations that demand extensive input data – data that is often scarce, outdated, or unreliable in exactly the regions where food security monitoring matters most. Simpler Earth Observation-based indices like vegetation greenness can track crop health but struggle to capture the underlying drivers of productivity – climate, water, and soil – together, at a resolution fine enough to guide field-level decisions. Existing global productivity frameworks, meanwhile, tend to rely on long-term averages at coarse spatial resolution, which can mask exactly the localized and seasonal stress episodes that make or break a harvest.

NECTAR addresses this gap by integrating three biophysical suitability factors – climate, water availability, and soil conditions – into a single, dynamic productivity score computed at 250-metre resolution for maize, rice, and wheat, from the year 2000 through to the present:

• Climate suitability – derived from monthly mean temperature, benchmarked against each crop's ecological growing requirements.
• Water availability – calculated from the ratio of actual to potential evapotranspiration, translated into crop-specific water sufficiency across each growth stage.
• Soil suitability – combining soil pH, water-holding capacity, and organic carbon content, each weighted for its role in supporting crop growth.

These three factors are combined into a weighted composite score (climate 30%, water availability 50%, soil quality 20%), reflecting water's outsized role as the most common limiting factor in crop growth. Critically, NECTAR does not simply average conditions across the season – it identifies the single worst-performing month and lets that month weigh heavily on the final score. This means a two-week drought during a critical growth stage shows up clearly in the index, rather than being smoothed away by an otherwise favourable season – giving NECTAR genuine sensitivity to the kind of short, sharp stress episodes that conventional long-term-average assessments miss entirely.

Validated against independent evidence

NECTAR has been tested and validated across 11 African countries – Burkina Faso, Côte d'Ivoire, Ethiopia, Malawi, Mali, Niger, Nigeria, Senegal, Togo, Uganda, and the United Republic of Tanzania – using two independent lines of evidence:

• Net Primary Productivity (NPP), an established, independent measure of land productivity: NECTAR showed strong seasonal agreement with NPP across districts in Ethiopia and Malawi, confirming that it captures genuine spatial productivity patterns within a growing season.
• Simulated crop yield, generated using a physically based crop-growth model in Malawi: NECTAR explained approximately 75% of the variability in simulated maize yield – a strong correlation that confirms its value as a productivity indicator across diverse agricultural zones.
  

Granular, actionable insight

Because NECTAR operates at 250-metre resolution, it resolves productivity variation within a single district that coarser, long-term global assessments can overlook entirely – differences driven by localized soil fertility, seasonal water stress, or microclimate that would otherwise be averaged away. Applied across the 11 study countries, NECTAR consistently identified where maize, rice, and wheat production is most viable: maize showing strong potential across central and southern regions of countries like Ethiopia, Malawi, Nigeria and Uganda; rice thriving where irrigation infrastructure is established; and wheat remaining constrained to cooler, higher-altitude zones and limited by drought and disease pressure elsewhere.

Supporting food security and planning

NECTAR's combination of high resolution and seasonal sensitivity makes it a practical tool for:

• Targeted agricultural planning – identifying high-potential zones for crop-specific investment.
• Sustainable land management – flagging where water access or soil health improvements would have the greatest impact.
• Early warning for food security risk – surfacing declining productivity signals in near real time, well before a harvest fails, giving governments and partners a window to act.