Germination: The moment when the first green leaf appears above the soil surface is marked as the germination phase of the plants. From this moment, in addition to the stored nutrients in the grain, the plant begins to feed on the substances produced by photosynthesis in the green leaves. (BBCH 05-09)

Third leaf phase: It is customary to mark the third leaf phase, when the third leaf reaches a length of 1-2 cm. This is necessary because usually at this point a so-called tillering node is established below the soil surface and a new phase in the plant’s development, called tillering, begins. (BBCH 13)

Tillering phase: It begins with the appearance of the first tiller under the coleoptile or sheath of the first green leaf. It represents the growth of the lateral buds of the central stem. When sowing is done on time and under normal weather conditions in autumn, 2-3 tillers develop in winter cereals, which are laid down during the embryonic period. During the tillering phase, the growth of most of the secondary roots also takes place. The appearance and growth of each tiller is associated with the development of roots connected to the base, and in some cases to the upper part of the corresponding internode. (BBCH 31)

Heading: Occurs when the tip of the panicle or half of the ear emerges from the axil of the uppermost leaf. (BBCH 51)

Flowering/Anthesis: Begins in the middle of the panicle. Wheat usually flowers 4-5 days after emergence. In self-pollinating plants, the stamens burst open in the flowers, their pollen falls on the stigmas, germinates, and fertilization occurs within a few hours. In cross-pollinating plants, the flower bracts open wide, the stamens are thrown out, and their pollen is carried by the wind and fertilizes other flowers. (BBCH 61)

Grain formation, filling and ripening: After fertilization, morphological, anatomical and chemical changes occur in the seed bud. The actual grain is formed in it. The growth of the grain and its composition are influenced by temperature, moisture and the nutrition of the plant. (BBCH 71-79)

The project has already seen promising results with crops like sunflower, maize, wheat, and roses. For example, in sunflower fields, the logistic regression models developed by PHENOFUSE have shown a high degree of accuracy in predicting NDVI based on radar backscatter data. This has allowed the researchers to double the number of data points available for analysis, significantly improving the temporal resolution of crop monitoring. By having more frequent data points, the project models can track subtle changes in crop development that might otherwise be missed, leading to more accurate yield predictions.

But the potential of machine learning in the PHENOFUSE project goes beyond just filling in data gaps. By applying these models, the project models can predict future crop growth based on historical data, allowing farmers and agricultural planners to make proactive decisions. For instance, the models can help predict when crops will reach critical phenological stages, such as flowering or harvest, enabling farmers to optimize irrigation, fertilization, and pest control efforts. This leads to more efficient use of resources and better crop yields, which is especially important in regions facing water scarcity or other environmental challenges.

One of the key advantages of using machine learning in this context is that the models can be continually refined and improved as more data becomes available. As the PHENOFUSE project collects more satellite observations and ground truth data, machine learning models can be trained to recognize increasingly complex patterns in crop growth. This will allow for more precise predictions and a better understanding of how different environmental factors, such as weather conditions or soil moisture, impact crop development.

Agromo is also exploring the use of machine learning to predict crop yields based on satellite data. By analyzing the relationship between radar backscatter data and vegetation indices, the Agromo team will develop models that estimate crop yields with a high degree of accuracy. This is particularly valuable for large-scale agricultural operations, where small improvements in yield prediction can have a significant impact on profitability.

In addition to its practical applications in agriculture, the PHENOFUSE project has contributed to the broader scientific understanding of how radar and optical satellite data can be used together to monitor vegetation. Stay tuned for more exciting news on Agromo’s science-based innovative modeling stack in the near future!

Sentinel-1 and Sentinel-2 are part of the European Space Agency’s Copernicus program, designed to provide high-resolution Earth Observation data. Sentinel-1’s radar sensors operate in the C-band, capturing backscatter data that measures the roughness and moisture of the Earth’s surface. This makes it particularly useful for observing agricultural fields, as it can detect changes in soil moisture and plant structure. Sentinel-2, on the other hand, is equipped with optical sensors that capture data in visible and infrared wavelengths, allowing researchers to calculate vegetation indices like NDVI and LAI, which are widely used to monitor plant health and biomass. The challenge with optical data, however, is that it is frequently disrupted by cloud cover, which limits the amount of usable data. This is where Sentinel-1’s radar capabilities come into play. Radar is unaffected by clouds, meaning it can provide continuous data, even in bad weather conditions.

By intercalibrating the radar data from Sentinel-1 with the optical data from Sentinel-2, the PHENOFUSE models create a more complete and accurate picture of crop growth over the season.

The process of intercalibration involves aligning the data from both satellites to ensure consistency in the time series. This is no easy task, as radar and optical data represent different physical phenomena and are processed in different ways. Radar data, for instance, is based on the backscatter of microwave signals, while optical data is based on the reflection of sunlight. However, through sophisticated algorithms and data processing techniques, PHENOFUSE is able to establish strong correlations between radar backscatter and optical vegetation indices.

One of the key findings of the project is that radar backscatter data from Sentinel-1, particularly in VV and VH polarizations, correlates strongly with traditional vegetation indices like NDVI and LAI from Sentinel-2. This correlation allows Agromo’s researchers to create time series of vegetation indices that are uninterrupted by cloud cover. For example, by combining data from both satellites, the PHENOFUSE models can generate a time series of crop health indicators that is twice as frequent as it would be using optical data alone. This enhanced temporal resolution is critical for tracking the phenological stages of crops, which can change rapidly during the growing season.

The intercalibration process has proven particularly effective for crops like wheat, sunflower, and maize, where the radar and optical data show strong correlations. For example, in sunflower fields, the team observed a high correlation between the radar VV backscatter data and the NDVI, allowing them to reconstruct NDVI time series even when optical data was unavailable. This is a major breakthrough for agricultural monitoring, as it enables continuous observation of crop health, even in regions with frequent cloud cover.

In addition to improving the temporal resolution of crop monitoring, intercalibration also enhances the accuracy of yield predictions. By combining radar and optical data, PHENOFUSE can generate more precise models of crop development, leading to better yield forecasts. This is particularly valuable for farmers and agricultural planners, who rely on accurate yield predictions to make informed decisions about harvesting, storage, and market prices.

The PHENOFUSE project’s work on intercalibration is setting the stage for a new era of agricultural monitoring. By integrating radar and optical data, the project is providing a more detailed and accurate view of crop development, helping to ensure food security in an increasingly unpredictable world.

The key innovation in the PHENOFUSE project is the integration of radar data from Sentinel-1 with optical data from Sentinel-2. Radar observations from Sentinel-1 are particularly valuable because they can penetrate cloud cover, allowing researchers to continue monitoring crop growth even during bad weather. On the other hand, Sentinel-2’s optical sensors capture detailed images of the Earth’s surface in multiple spectral bands, which are critical for calculating traditional vegetation indices like the Normalized Difference Vegetation Index (NDVI) and Leaf Area Index (LAI). 

Through advanced data processing techniques, PHENOFUSE will transform these satellite observations into meaningful indicators of crop health. This involves calculating various vegetation indices for fields of wheat, sunflower, maize, and roses. These indices offer a quantitative measure of how healthy a crop is, how much biomass it has accumulated, and when it will reach critical growth stages. By monitoring these indices over time, the PHENOFUSE models will be able to predict important phenological phases like flowering and harvest time with greater accuracy than ever before.

But the project doesn’t stop there. By fusing radar and optical data, PHENOFUSE addresses one of the most significant challenges in Earth Observation: the inconsistency in data due to cloud cover or limited revisit times. Traditionally, cloud cover has been a major obstacle for optical satellite observations, especially in regions like Bulgaria, where the project is focused. With the integration of Sentinel-1’s radar data, which is unaffected by cloud cover, PHENOFUSE will ensure continuous monitoring, providing a complete and uninterrupted picture of crop development.

This continuous monitoring is particularly valuable for predicting crop yields. With access to near real-time data on crop growth, farmers and researchers can make informed decisions about irrigation, fertilizer application, and pest management, leading to more efficient and sustainable farming practices. In a world facing climate change and a growing global population, the ability to maximize agricultural productivity while minimizing environmental impact is more important than ever.

The PHENOFUSE project’s integrated approach to satellite data offers a glimpse into the future of precision agriculture. By combining cutting-edge Earth Observation technology with innovative data processing techniques, PHENOFUSE is helping to pave the way for a new era in agricultural monitoring and management.