Eoagri

Health status: biophysical condition

Biophysical condition provides an assessment of crop health status by quantifying a key vigor variable such plant greenness. The product generates spatial maps and time-series analytics, classifying crop vigor into five standard categories — very low, low, average (seasonal baseline), high, and very high. The five status categories are adjusted from vigor values of each targeted crop, and in each agroecological region, so the same vigor values do not apply to categorize neither to the same crop in other agroecological region neither to other crops.

Biophysical condition facilitates actionable insights for effective crop management in terms of plant greenness assessment.

• Precision agriculture

• Plant greenness

• Crop management

• Yield optimization

• Agriculture insurance and financing

Health status: leaf area & photosynthesis activity

Leaf area & photosynthesis activity provide an assessment of crop health status by quantifying key canopy variables such as Leaf Area Index (LAI) and the fraction of Absorbed Photosynthetically Active Radiation (fAPAR). The product generates spatial maps and time-series analytics, classifying LAI or fAPAR into five standard categories — very low, low, average (seasonal baseline), high, and very high. The five status categories are adjusted for each targeted crop, and in each agroecological region, so the same values do not apply to categorize neither to the same crop in other agroecological region neither to other crops.

Leaf area & photosynthesis activity facilitate actionable insights for crop management in terms of plant growth, and photosynthetic efficiency.
•    Precision agriculture
•    Plant growth
•    Crop management
•    Yield optimization
•    Agriculture insurance and financing
 

Health status: nutritional balance

Nutritional balance provides an assessment of crop health status by quantifying key nutritional variables such as light absorption by pigments (Redgeabs), chlorophyll a and b content (Chla+b), and nitrogen concentration (Ni). The product generates spatial maps and time-series analytics, classifying Redgeabs, Chla+b or Ni into five standard categories — very low, low, average (seasonal baseline), high, and very high. The five status categories are adjusted for each targeted crop, and in each agroecological region, so the same values do not apply to categorize neither to the same crop in other agroecological region neither to other crops.

Nutritional balance facilitates actionable insights for crop management in terms of photosynthetic efficiency and nutrient availability of farmlands.

• Precision agriculture

• Plant photosynthetic efficiency

• Nutrient availability

• Crop management

• Yield optimization

• Agriculture insurance and financing

Health status: humidity rates

Humidity rates provide an assessment of crop health status by quantifying a key humidity variable such as the moisture content in the interface soil/crop (Humidity). The product generates spatial maps and time-series analytics, classifying Humidity into five standard categories — very low, low, average (seasonal baseline), high, and very high. The five status categories are adjusted for each targeted crop, and in each agroecological region, so the same values do not apply to categorize neither to the same crop in other agroecological region neither to other crops.

Humidity rates facilitate actionable insights for crop management in terms of moisture content.

• Precision agriculture

• Moisture content

• Crop management

• Yield optimization

• Agriculture insurance and financing

Parcel condition

Parcel condition provides an assessment of crop heath status by comparing a key canopy variable such as Leaf Area Index (LAI) against the averaged LAI in the corresponding agroecological zone. The information is reduced to a simple notification system with three unique labels: Alarm, Warning, and No risk. The approach is dedicated to each targeted crop, and in each agroecological region, so those labels do not apply neither to the targeted crop in other agroecological region neither to other crops. 

Parcel condition facilitates actionable insights for crop management in terms of comparing green canopies between plots within an agroecological region.

• Precision agriculture

• Plant growth

• Crop management

• Yield optimization

• Agriculture insurance and financing

Dry matter productivity

Dry Matter Productivity (DMP) provides an assessment of crop heath status by estimating a key growth performance variable such as dry matter. The deployed approach integrates satellite-derived indicators with meteorological data under a Light Use Efficiency (LUE) model, which relates absorbed radiation to biomass synthesis. The methodology involves 10-day time-series preparation, calculation of net primary productivity, and subsequent derivation of dry matter using adjusted parameters in each agroecological region. 10-day temporal resolution offers a baseline to measure increment or decrement rates, and to detect key phenological times together with their corresponding biomass accumulated.

Dry matter productivity facilitates actionable insights for crop management in terms of biomass production.

• Precision agriculture

• Biomass production

• Crop management

• Yield optimization

• Agriculture insurance and financing

Yield estimation

Yield estimation provides an assessment of crop health status by estimating a key productivity variable such as grain yield estimation. The cumulative dry matter productivity along a season is converted to yield using locally calibrated harvest indices adjusted to each agroecological region. This calibration ensures that the model reflects the specific characteristics of targeted crops in each agroecological region. Yield estimation, typically expressed in tons per hectare, is conducted after the peak of the season (POS) and before the end of the season (EOS).

Yield estimation facilitates actionable insights for crop management in terms of optimizing productivity analysis.

• Precision agriculture

• Productivity analysis

• Crop management

• Yield optimization

• Agriculture insurance and financing

Soil water content

Soil Water Content (SWC) product is derived from curated vegetation indices fused with in-field hygrometer measurements. Scattered field measurements of soil water content at 10 centimetres are used to calibrate the satellite measurement each time it monitors the farm plots. This generates a SWC product at 10 centimetres depth with a spatial resolution of 10 meters, calibrated at the same time as it is monitored by satellite. 

Soil water content supports informed decision-making for crop health, water resource management, and yield optimization.

• Precision viticulture

• Water management

• Irrigation needs

• Soil monitoring

• Yield optimization

Irrigation needs

Irrigation needs product calculates real evapotranspiration (ETc) using a refined adaptation of the Penman-Monteith equation, incorporating both physical and vegetation-based parameters. Key refined inputs:

• Albedo dynamically derived from satellite data—using land surface temperature.

• fAPAR (Fraction of Absorbed Photosynthetically Active Radiation) to estimate crop coefficient (Kc).

The model synthesizes these curated vegetation indices with local weather station measurements, edaphological analysis, and advanced geospatial analytics. This integrated approach enables high-resolution modelling of vineyard microclimates and soil moisture dynamics. 

Irrigation needs product is updated continuously throughout the growing season, providing a real-time view of actual evapotranspiration rates. This supports precise, data-driven irrigation scheduling by aligning water application directly with the vines' physiological demand, optimizing both water efficiency and vineyard health.

• Precision viticulture

• Water management

• Irrigation needs

• Yield optimization

Diseases risk

Disease risk is an early warning system to provide predictive insights into the onset risk of major vineyard diseases, such as powdery mildew or botrytis. By correlating key climatic factors with disease development models, the system uses weather data and geospatial analytics to generate daily parcel-level risk assessments. Eight levels ranging from ‘0’ to ‘7’ are expected, where ‘0’ means no disease conditions in the last seven days and ‘7’ means disease conditions every day in the last seven days.

Disease risk provides continuous updates along key phenological states ensuring timely alerts, enabling proactive crop protection strategies, and supporting ecological farming.

• Precision viticulture

• Ecological farming

• Disease occurrence

• Early warning

• Proactive crop protection

• Yield optimization

Fertilization demand

Fertilization demand is activated on demand when a high risk is identified by Disease Risk product. Very High-Resolution (VHR) multispectral imagery is required here to assess very-detailed vine vigor and nutritional status. The analysis maps spatial variability in plant development, identifying zones with differing nutrient requirements. It facilitates the implementation of variable-rate fertilization strategies that target specific areas, optimizing nutrient utilization and supporting plant health.

Fertilization demand enables timely and precise nutrient management.

• Precision viticulture

• Early waring

• Disease occurrence

• Nutritional Status

• Ecological farming

• Canopy monitoring

Yield forecast

Yield forecast provides estimations of grain yield at plot level. Predictive models are trained using real yields from past agricultural years, along with satellite-based and climatic explanatory variables. The result is a geospatial layer enriched with plot-specific yield estimates, starting two and a half months before harvest and updated every 15 days.

Yield forecast supports actionable insights that facilitate proactive decision-making in terms of green pruning, organize the define the quality of the vintage, purchase of grapes from suppliers, and control of each vine in production, knowing which vines are no longer productive or should be grubbed up.

• Precision viticulture

• Yield forecast

• Green pruning management 

• Vintage quality control

• Harvest planning

Crop location & crop type

Crop location consists of a 10-meter resolution map that delineates cultivated and non-cultivated land using an automated machine learning model fed by satellite-based explanatory variables. In addition, an uncertainty map provides a measure of confidence for the binary classification results. The product is generated annually and is accompanied by a spatial statistical analysis, which can be adapted to administrative divisions or specific user needs to provide a more in-depth view of the distribution of arable land.

Crop type consists of a 10-meter resolution map with the detection of each crop grown within the arable land using an automated machine learning model using satellite-based explanatory variables. In addition, an uncertainty map provides a measure of confidence for the classification results. The product is generated by each phenological cycle and is accompanied by a spatial statistical analysis, which can be adapted to administrative divisions or specific user needs to provide a more in-depth view of the distribution of crop types.

• Agricultural financing

• Financial portfolios in the agri-sector 

• Traceability for certified supply chains

• Agricultural planning and support

• Agricultural land use mapping

• Risk assessment

Acreage mapping

Acreage mapping delineates individual agricultural parcels using an automated deep learning algorithm (Convolutional Neural Network). By analyzing high-resolution satellite, the CNN identifies subtle features such as boundary contrasts, vegetation indices, and texture variations. This allows the model to distinguish adjacent fields even when they lack visible physical barriers like fences or hedgerows, achieving high precision at scale.

Acreage mapping delivers individual agricultural parcels, which will be further used as input for ‘Parcel status’.

• Agricultural financing

• Financial portfolios in the agri-sector 

• Traceability for certified supply chains

• Agricultural planning and support

• Agricultural land use mapping

• Risk assessment

Crop rotation

Crop rotation provides an assessment for agricultural financing by quantifying accurate crop rotation practices in an agricultural year. A growing season detection algorithm detects all local maxima in the time series, after which, for each of the maxima, it finds a start and end of season in a pre-defined search window around the peak. In case a second growing season is detected in the same agricultural year, the processor can also classify the crop.

Crop rotation, with information at parcel level, is generated in each agricultural year, consists of the number of growth stages detected and the secondary crop classified if no single growth stage is present. The product may be accompanied by a spatial statistical analysis for each administrative division if required by agricultural financing needs.

• Agricultural financing 

• Regional and policy planning

• Financial portfolios in the agri-sector

• Traceability for certified supply chains

• Sustainable crop rotation strategies

• Carbon farming

Cover crops

Cover crops provide an assessment for agricultural financing by quantifying accurate cover crops patterns during an agricultural year. Machine learning models are trained at parcel level for each main crop during a growth cycle, so acreage mapping and crop rotation detection are mandatory preliminary steps.

Cover crops, with information at parcel level, is generated in each agricultural year, and is accompanied by a spatial statistical analysis for each administrative division to be compliant with specific agricultural financing needs.

• Agricultural financing

• Regional and policy planning

• Financial portfolios in the agri-sector

• Traceability for certified supply chains

• Sustainable crop rotation strategies

• Carbon farming

Tillage

Tillage provides an assessment for agricultural financing by identifying tillage practices using multi-temporal satellite imagery and a machine learning approach. Models are trained at parcel level for each main crop during a growth cycle, so acreage mapping, crop rotation and cover crops detection are mandatory preliminary steps

Tillage, with information at parcel level, is generated in each agricultural year, and is accompanied by a spatial statistical analysis for each administrative division to be compliant with specific agricultural financing needs.

• Agricultural financing 

• Regional and policy planning

• Financial portfolios in the agri-sector

• Traceability for certified supply chains

• Sustainable crop rotation strategies

• Carbon farming

Orchard census

Orchard census provides a high-precision geospatial solution for counting individual trees within orchard parcels. This solution uses Very High Resolution (VHR) optical satellite imagery or UAV-based data, combined with a semi-automated approach and machine learning algorithms, this solution enables accurate tree counts and leverages basic vegetation indices.

Orchard census provides actionable insights for resource planning, and operational optimization, while supporting compliance with sustainability and certification standards.

• Inventory and mapping of tree orchards for agroforestry planning.

• Monitoring orchard expansion or reduction over time.

• Supporting land-use planning and agricultural zoning.

• Estimating tree-based crop production potential.

• Guiding interventions for orchard management and sustainability.

• Assessing carbon sequestration potential in tree orchards.

Health status

Health status provides a high-precision geospatial solution for health measurement at the individual tree level in woody crop systems. Utilizing Very High Resolution (VHR) optical satellite imagery or UAV-based data, in conjunction with a semi-automated approach and machine learning algorithms, the tool can identify trees that deviate from expected conditions using five qualitative performance categories, ranging from Very poor to Very good performance. In addition, the crown area is measured and then compared with the average of each individual parcel.

Health status enables growers to achieve precision management with targeted interventions, providing actionable insights for woody crop systems health optimization, resource allocation improvement, and overall productivity enhancement.

• Inventory and mapping of tree orchards for agroforestry planning

• Monitoring orchard expansion or reduction over time

• Supporting land-use planning and agricultural zoning

• Estimating tree-based crop production potential

• Guiding interventions for orchard management and sustainability

• Assessing carbon sequestration potential in tree orchards

Commodity mapping

Commodity mapping consists of a 10-meter resolution map that delineates key commodities (e.g. oil palm, eucalyptus, poplars, coffee, cocoa, or rubber) within heterogeneous mixed-farming landscapes and complex canopy conditions using an automated machine learning model fed by satellite-based explanatory variables. Crop maps serve as a foundational layer for computing production metrics, including acreage estimations, and support downstream applications, like biomass accounting and yield estimation.

Commodity mapping empowers stakeholders to optimize resource allocation, driving data-driven decision-making for sustainable land management and improved agricultural productivity.

• Agroforestry resource management

• Agroforestry system optimization

• Biomass optimization strategies

• Sustainable land management

• Risk assessment and resilience planning

• Policy and investment decision support

Above ground biomass

Above-ground biomass quantifies Above-Ground Biomass (AGB) in commodity systems. The deployed approach integrates satellite-derived indicators with meteorological data under a Light Use Efficiency (LUE) model, which relates absorbed radiation to biomass synthesis. The methodology involves 10-day time-series preparation, calculation of net primary productivity, and subsequent derivation of biomass per hectare using adjusted parameters to each commodity system. From monthly to seasonal monitoring of AGB dynamics, the product enables precise tracking of growth patterns at both stand and sub-stand levels. 

Above-ground biomass is designed for agroforestry operators, providing them with information on the growth and development of plantations, which can be very useful for decision-making, carbon accounting initiatives, and sustainability programs.

• Agroforestry resource management

• Biomass optimization strategies

• Yield forecasting and harvest planning

• Carbon accounting and climate reporting

• Risk assessment and resilience planning

• Policy and investment decision support

Yield estimation

Yield estimation combines commodity system mapping with above-ground biomass assessment to enable yield modelling. The cumulative dry matter productivity over a season is converted into yield using locally calibrated harvest indices to reflect the specific characteristics of the commodity in question. Yield estimation, typically expressed in tons per hectare, is usually conducted after the peak of the season (POS) and before the end of the season (EOS).

Yield estimation reduces operational uncertainty, strengthens the resilience of agroforestry systems, and increases profitability.

• Agroforestry resource management

• Yield forecasting and harvest planning

• Risk assessment and resilience planning

• Sustainable land management

• Policy and investment decision support

Evapotranspiration rates

Evapotranspiration rates quantify potential evapotranspiration (ETo) using a refined adaptation of the Penman-Monteith equation, incorporating both physical and vegetation-based parameters. Key refined inputs:

• Albedo dynamically derived from satellite data—using land surface temperature.

• fAPAR (Fraction of Absorbed Photosynthetically Active Radiation) to estimate crop coefficient (Kc).

The model synthesizes curated vegetation indices with global climate data. Evapotranspiration rates are tracked on a daily, dekadal, monthly, or annual basis, benchmarked against historical averages to identify anomalies that indicate stress or favorable conditions.

Evapotranspiration rates are a key tool for producers to help them anticipate forage shortages, adjust grazing schedules, and plan supplementary feeding, thereby reducing operational risks and enhancing resource efficiency. At a broader scale, aggregated analytics at the departmental level equip cooperatives and policymakers with actionable data to design early interventions, mitigate drought impacts, and protect livestock supply chains.

• Resource and productivity mapping

• Mobility and distribution tracking

• Water resource management

• Climate and drought risk management

• Rangeland conflict identification

• Decision-making

Dry above ground biomass

Dry above-ground biomass quantifies biomass in pastures. The deployed approach integrates satellite-derived indicators with meteorological data under a Light Use Efficiency (LUE) model, which relates absorbed radiation to biomass synthesis. The methodology involves dekadal time-series preparation, calculation of net primary productivity, and subsequent derivation of biomass per hectare using calibrated in-field parameters.

Dry above-ground biomass is designed for producers providing actionable insights for variability analysis, supporting decision-making across extensive agricultural landscapes, enabling precise monitoring of pasture productivity at farm, or territorial scales.

• Resource and productivity mapping

• Mobility and distribution tracking

• Biomass production

• Harvest planning

• Decision-making

Seasonal dynamics

Seasonal dynamics deliver growth rates on pasture systems based on a comprehensive phenometric analysis over time-series dry above ground biomass. Key seasonal milestones are identified—Start of Season (SOS), Peak of Season (POS), and End of Season (EOS)— for track growth cycles measuring the Rate of Green-Up (ROI) and the Rate of Green-Down (ROD). Both indicators allow producers to anticipate biomass peaks, tracking productivity trends, and forage availability.

Seasonal dynamics enable precise tracking of growth patterns for grazing schedules and forage availability.

• Resource and productivity mapping

• Mobility and distribution tracking

• Biomass production

• Harvest planning

• Decision-making

Pastures condition

Pasture condition provides a detailed assessment of pasture status by quantifying variability of dry above-ground biomass in relation to the nearest corresponding paddocks. The product generates time-series spatial maps, classifying the variability into five categories by highlighting underperforming biomass areas, facilitating targeted interventions such as rotational grazing or supplemental feeding.

Pasture condition is designed to optimize decision-making by providing timely, accurate, and operational insights.

• Resource and productivity mapping

• Mobility and distribution tracking

• Biomass production

• Harvest planning

• Decision-making

Biomass for SOC modelling: agriculture

Above-Ground Biomass (AGB), in the form of crop residues (stalks, leaves, husks), is the primary source of fresh organic matter returned to the soil. The quantity of AGB directly determines the amount of carbon available for incorporation into the soil system. Models such as Century, DAYCENT, or RothC explicitly require residue input data. Underestimating AGB inputs leads to models predicting excessive SOC decline. Conversely, overestimation can mask real degradation risks. Precise AGB calculation is thus essential for establishing a realistic carbon mass balance.

Biomass for SOC modelling demands not only accurate snapshot AGB maps but also multitemporal analysis across several years. This demand brings indispensable preliminary steps on accurate agronomic contextualization such as crop type mapping, crop rotation system detection, and cover crops detection. This approach is non-negotiable for advancing precision carbon farming, validating climate-smart agricultural protocols, and generating trusted forecasts for soil sustainability.

• Carbon Credit Markets & Verification

• Assessing Nationally Determined Contributions (NDCs)

• Land Degradation Neutrality (LDN) Tracking

• Erosion Risk Assessment

• Soil Health Forecasting

Biomass for SOC modelling: agroforestry

Above-Ground Biomass (AGB), in the form of tree crop residues (stalks, leaves, husks), is the primary source of fresh organic matter returned to the soil. The quantity of AGB directly determines the amount of carbon available for incorporation into the soil system. Models such as Century, DAYCENT, or RothC explicitly require residue input data. Underestimating AGB inputs leads to models predicting excessive SOC decline. Conversely, overestimation can mask real degradation risks. Precise AGB calculation is thus essential for establishing a realistic carbon mass balance.

Accurate Biomass for SOC modelling in agroforestry requires more than snapshot biomass maps—it demands long-term, multitemporal analysis due to the woody growth where some events play a key role such as planting, pruning, thinning cycles or cutting to make way for new plantings.

• Carbon Credit Markets & Verification

• Assessing Nationally Determined Contributions (NDCs)

• Land Degradation Neutrality (LDN) Tracking

• Erosion Risk Assessment

• Soil Health Forecasting

Biomass for SOC modelling: agropastoral

Above-Ground Biomass (AGB), in the form of rangeland residues (stalks, leaves, husks), is the primary source of fresh organic matter returned to the soil. The quantity of AGB directly determines the amount of carbon available for incorporation into the soil system. Models such as Century, DAYCENT, or RothC explicitly require residue input data. Underestimating AGB inputs leads to models predicting excessive SOC decline. Conversely, overestimation can mask real degradation risks. Precise AGB calculation is thus essential for establishing a realistic carbon mass balance.

Biomass for SOC modelling in grasslands relies on accurate maps of Above-Ground Biomass (AGB) combined with multitemporal analysis across several years. This is necessary to account for seasonal growth cycles and year-to-year variability in forage production. Productivity dynamics is linked to climate variability and fertilization.

• Carbon Credit Markets & Verification

• Assessing Nationally Determined Contributions (NDCs)

• Land Degradation Neutrality (LDN) Tracking

• Erosion Risk Assessment

• Soil Health Forecasting

Top-soil: Soil organic matter

Soil organic matter quantifies changes within croplands as a direct result of farming practices at a depth of 10 cm. The time-series approach combines satellite observations with local soil organic matter data. This approach delivers highly detailed, spatially resolved maps of organic matter and calculates organic matter losses relative to a baseline scenario. This methodology is applicable to any crop, and models soil organic matter changes through a Tier 3 soil process-based approach once the initial organic matter map is established. This model integrates initial organic matter values, crop type maps and farming practices. The result is an accurate measurement of soil organic matter, which can be used to guide sustainable soil management and other applications.

• Guiding sustainable land management practices

• Detecting soil degradation and desertification risks

• Evaluating restoration efforts in degraded lands

• Informing national and regional soil health programs

• Designing interventions for carbon farming and regenerative agriculture

• Supporting biodiversity and ecosystem services analysis

Top-soil: SOC content and changes

SOC content and changes is an EO-based solution that delivers a method for estimating and tracking Soil Organic Carbon (SOC) levels and their temporal dynamics across agricultural landscapes. The product provides accurate SOC percentage estimates and changes detection at scale. By integrating EO-derived data with agronomic practices such as cover cropping and residue management, it enables stakeholders to design and implement effective carbon sequestration strategies. Its comprehensive monitoring capability supports results-based payment schemes and underpins digitized Measurement, Reporting, and Verification (MRV) systems essential for carbon markets.

• Climate Change Mitigation

• Supporting greenhouse gas inventories and carbon credit schemes

• Guiding sustainable land management practices

• Detecting soil degradation and desertification risks

• Evaluating restoration efforts in degraded lands

• Informing national and regional soil health programs

• Designing interventions for carbon farming and regenerative agriculture

• Measuring impacts of land-use change on soil carbon stocks

• Supporting biodiversity and ecosystem services analysis

Clearance detection

Clearing detection product monitors activities affecting carbon stocks through high-resolution satellite imagery and to cross-reference this data with field visit records, thereby providing accurate information on deforestation events.

Integrating time-series analysis into machine learning algorithms enables the product to accurately detect changes in land use. These results are then cross-referenced with field validation records to improve accuracy. The product supports compliance with sustainability standards and carbon accounting frameworks, while also serving as a mandatory step prior to quantifying carbon losses associated with clearance activities.

• Agroforestry carbon stocks

• Supporting the growth and integrity of carbon markets

• Monitoring, Reporting, and Verification (MRV) for carbon markets

• Quantifying carbon sequestration

• Mapping and assessing carbon stock potential

• Estimating carbon density

• Guiding policy and management

Clearance impact on stocks

Clearance impact on stocks provides a precise quantification of annual and total carbon stocks in forested landscapes, enabling stakeholders to assess the effects of forest clearance on carbon dynamics.

Utilizing high-resolution satellite imagery and advanced remote sensing techniques, combined with machine learning models, the product delivers accurate, scalable measurements of biomass and carbon fluxes over time. By monitoring changes in carbon stocks, organizations can evaluate deforestation risks, support carbon accounting frameworks, and comply with sustainability and climate mitigation policies. The product empowers decision-makers with actionable insights for conservation planning, carbon credit validation, and strategic land-use management, ensuring transparency and accountability in forest resource governance.

• Agroforestry carbon stocks

• Monitoring, Reporting, and Verification (MRV) for carbon markets

• Quantifying carbon sequestration

• Supporting the growth and integrity of carbon markets

• Mapping and assessing carbon stock potential

• Estimating carbon density

• Guiding policy and management