Precision Agriculture

Implementing smart remote sensing solutions to promote the establishment of sustainable farming systems.

The aim of the project is to develop a smart system for fertilisation, based on remote sensing of plant nutritional levels during the season. Data collected in the study will yield accurate fertilisation maps. Remote sensing data will be assessed together with soil sensor data in order to develop a high-quality field-based analysis system.  The aim is to avoid excess fertilisation of crops as it can have negative environmental impact, leading to groundwater pollution and greenhouse gas emissions. Insufficient fertilisation poses risks for plant growth, reduces soil fertility and productivity.

The tests take place with the following crops:

Spring wheat

Spring barley

Spring swede rape


Experiments will be carried out on test plots of the Estonian University of Life Sciences and at the cluster members’ production fields. The vegetation index corresponding to the fertilisation norms, and crop’s need-based nitrogen (N) norms will be calculated and determined for furter studies. 

In Estonia, we have joined forces with the European Space Agency, as a result of which we have access to satellite photos. Satellite images are obtained from the Tartu Observatory, a calculation model will be developed for the fertilisation maps. During three years of testing (in 2017-2019), field experiments have been carried out to determine the nutrient supply of crops in local climate conditions through remote sensing.  Crops assessed were established in the 1990s: spring wheat, spring barley and spring swede rape

The experimental plots set up at the Estonian University of Life Sciences were subjected to basic fertilisation of crops before sowing in order to ensure sufficient phosphorus and potassium reserves in the soil for the entire growing season. During the growing season, two additional top-dressings with mineral nitrogen fertiliser were applied. To determine the vegetation index and to evaluate the effect after fertilisation, nutrient supply during the growth season was measured by hand and via remote monitoring (four times). During harvest, crop yields were recorded from each test plot to see if monitoring in also reflected in yields. In the autumn, soil samples were taken from the experimental fields to analyse the nutrient supply and fertilisation effect  during the growing season and to set new norms for the upcoming season. 

Of the experiments established on the production fields, drone monitoring was performed to compare the growth levels from test plots to the data yielded from production fields located in different regions of Estonia. Recommendations were made for more effective field management for the next season.

Next, a calculation model will be developed for need-based fertilisation along with fertilisation maps and a site-based recommendations for farmers situated all over Estonia. For this, we use satellite monitoring for NDVI measurements of arable crops (photo above). 

The calculation model can be used by farmers in various programs based on GIS software. The use of fertiliser maps help avoid excess fertilisation, provide an opportunity to make decisions based on the actual situation and soil conditions. Recommendations will be made based  on NDVI determined through remote monitoring systems. 


The aim of the innovation activity was to use remote sensing to apply mineral fertilisers to plants where they are needed and in the quantities required to produce optimal yields under the given weather conditions. Overuse and underuse of mineral fertilisers is a threat to the environment. Therefore, we anticipated to develop a model for the optimal use of mineral fertilisers. Applying mineral fertilisers to the field in the wrong place, at the wrong rate and at the wrong time means that some of the nutrients are not used by the plants. These unused nutrients are partly volatilized into the atmosphere and partly leach into groundwater. Unused nutrients are also a loss to the farmer.

With the help of satellite monitoring, we can identify areas in fields where plants have not germinated, where the vegetation is sparse, where nutrient availability to plants is low, or where nutrients are readily available to plants.

We conducted field tests to find out the effect of different rates of nitrogen fertilisation on the Normalized Difference Vegetation Index (NDVI reading). The experiments were carried out in four consecutive years, which allowed us to see the effect of different weather conditions on the nutrient uptake of the plants, which was also reflected in the NDVI measurements. It was also clear from visual observation that if rainfall is evenly distributed during the growing season and plants are not water deprived, any additional nitrogen is taken up by the plants and the plants become increasingly lush and green (photo above).

Using satellite monitoring, we determined NDVI readings of the three crops under study (spring wheat, barley and swede rape), at different fertilisation levels.

We determined if and by how much additional nitrogen fertilization increases NDVI.
– From the field trials it was possible to see how much additional nitrogen fertilisation we need to apply to the crop under study if we want to increase NDVI to a higher level.
– We determined the NDVI of the test plots and fixed the yield of these test plots after harvest.

The trials were carried out on the production fields of the members of Soil Innovation Cluster (OÜ Voore Farm, Torma POÜ and Growing Crops OÜ). A sample photo of an experimental field used for remote sensing tests shown below.

In agriculture, there is no fixed result that clearly corresponds to a single activity, as there are many factors that influence plant growth during the growing season, like precipitation and temperature. 2017 was a very rainy year in Estonia, but the dry year of 2018 did not allow the plants to take up the fertiliser applied to the fields. The lack of precipitation in 2018 prevented the mineral fertiliser from dissolving in the soil.


Satellite monitoring of vegetation data can identify many problems. The processing of satellite monitoring data produces NDVI data layers (photo on the right).

Innovation to...

We can use satellite monitoring to identify problem areas in the field and, if necessary, investigate them further on the spot. 

Identify Problems:

Agro-technical faults in the field
Agronomic problems in the field
Differences in soil water regime
Nutrient deficiencies in the soil
Problems with nutrient uptake due to low soil pH
Nutrient deficiencies for plant growth
Problems caused by plant diseases and pests

The correlation between NDVI and yield is quite strong, and so we can construct a computational model for each crop between the NDVI measured during the active growth phase and the yield expected later in the field from the same area. The graphs obtained from the measurements are valid for Estonian weather conditions and for apparently waterlogged soils.

NDVI is not always positively related to the fertilisation rate, but is strongly dependent on rainfall and air temperature during the growing season of the crop. Under optimal growing conditions, the NDVI reading on a crop can tell us how much more nitrogen fertilisation would be needed to obtain the maximum yield under these conditions.

NDVI data layers shown on field map.

The innovation activities carried out have shown that satellite monitoring is an effective method for assessing the condition of crops in the field, across the whole farm at a single point in time. The analysis of the satellite monitoring data allows us to calculate the amount of fertiliser needed for the fields before fertiliser application can start. This information enables the company to plan fertiliser use and logistics at an early stage, resulting in cost savings for the company.

Tooltip content

The field trials on the implementation of remote sensing solutions were carried out in the framework of measure no 16 “Cooperation” of the Estonian Rural Development Programme 2014-2020. The budget for the innovation activity was 143 760€  (project period: 06.03.2017 – 06.01.2022).

Estonian University of Life Sciences Toomas Tõrra, Alar Astover
Tatoli AS   Jaanus Kilgi
Observatory of Tartu Kaupo Voormansik
Estonian Agricultural Research Centre – Priit Penu