By Ricardo Rojas, Agricultural Engineer - Avium Technical Advisor; Nicolás Martínez, Computer Engineer - Avium R&D Technological Support; Carlos Tapia, M.Sc., Avium Technical Director.
What seemed unimaginable a couple of decades ago is now a reality. Precision technology is here to stay and scale up the development of agriculture. An example of thisOr drones are increasingly making a greater contribution to cherry production, providing data that is both specific and accurate.
The adoption of new technological tools and/or new technologies has been a major challenge for the Chilean fruit industry. The use of these in agriculture has always generated a certain amount of disbelief, because in the short term it can be very expensive or also because the information generated is not fully utilized by those who interpret it, causing the technological tools and/or new technologies, from the receiver's point of view, not to generate value in order to improve or make management more efficient at the orchard level.
However, with each passing day the cost of using technology becomes more accessible, allowing for a favorable benefit. Therefore, the use of sensors, drones and satellites, which a few years ago were thought to be very remote in agriculture, are today very important instruments in decision-making, both in the management of tasks carried out in orchards and in everyday life.
The Avium team has conducted various investigations into the use of these technologies; the initial idea was to completely change the traditional management of agriculture, leaving behind recommendations for large, unrepresentative surfaces, based mainly on averages and, in the specific case of cherry cultivation, using new or existing technologies to characterize the spatial variability that plants present within a field and/or plot.
The work carried out focuses on the interpretation of images collected at a macro level by satellites and at a more micro level by a Drone. These are processed and converted to multispectral images to perform an interpretation, observing in the field agronomic variables of interest in the crop.
In this context, different production situations of orchards located in the O'Higgins and Maule regions were analyzed and, it is worth mentioning, that the work carried out by Avium aims to adapt these tools to the crop to generate reliable data, which when analyzed, cause a positive, real and quantifiable impact on cherry production.
Preliminary results:
Over time, Avium has characterized the different plant conditions and their variability at the orchard level, using the following plant categories according to their “vigor” condition:
–Type 0: Dead plant or space without plants.
–Type 1: Plant with little or no annual shoot growth (0-10 cm). Average number of branches well below the normal size of the quarter, showing two to three lateral shoots with uncontrolled growth (excessive vigor), its leaves are smaller and have a grayish-green hue.
–Type 2: Plant with annual growth of between 30 and 60 cm. Balanced plant.
–Type 3: Plants with annual growth of over 60 cm. They appear to be plants with excess vegetative development, more unbalanced and shadier trees.
This plant-by-plant characterization allows each production unit on the property (block) to be visually parameterized and, using an office template, to transform this data into colored cells, which are easily displayed on a “Heat Map of Plant Types,” as shown in Image 1.
Image 1. Detailed mapping of Type Plants according to Avium category (Heat Map).
Image 2. Biomass map generated by Software using orthophoto.
In practice, it is impossible for each producer to characterize each of the plants in the different production units on the farm, taking into account the amount of labor that would be needed to carry out this task, which would consist of 3-4 days/hectare. In addition, it is necessary to differentiate the vigor index and its changes over time, which makes it unfeasible to carry out the work in this way since it would significantly increase production costs per hectare. Faced with this problem, the use of new precision agriculture tools and/or new technologies allows for the simple and rapid characterization of large surfaces, where the only limitation is the autonomy of the device to be used.
Among the different softwares used by Avium, adapted to the particular conditions of cherry cultivation, the results show that they are quite close to the reality of the plot, as can be seen in Image 2 processed by the program, and in parallel it can be seen that the trend observed in both illustrations is reliable when analyzing certain sectors that are marked in red in Image 3.
Image 3. Software-generated biomass map versus Surface Mapping of Type Plants. (Source: Agrodreams-Copeval)
Furthermore, the aim of using these new tools is to gather data to be interpreted and transformed and thus provide readings of different agronomic indices of great relevance and high economic impact, such as effective canopy volume (ECV).
This is why one of the biggest problems faced today in the orchard is the excessive use of phytosanitary products, often considering the efficiency of their application. It is very common that, in the orchard, a greater volume of water is being applied, in addition to a greater use of chemical/biological products. This bad policy of applying more to make it “better” is more common than it appears to be, mainly due to misinformation and often by repeating recipes from neighbouring fields or default recommendations. To avoid this problem, an effective measurement of the effective canopy volume, which is directly related to the amount of water and chemical/biological products to be used, is extremely important information when planning production costs.
The first results of obtaining the VHA quickly and easily for each particular condition of the orchard showed very promising results, adjusting in a similar way to the calculation of the VHA, measured in the field in the traditional way. Below, an image processed through artificial intelligence is observed, where it is calculated, through a Drone flight, with how many liters of product each sector within a block should be wetted/covered.
In another aspect, the use of multispectral images such as NDVI (Normalized Digital Vegetation Index), is a multispectral map that shows the real biomass condition at the time of the flight. It can indicate the homogeneity or heterogeneity of the crop and its consequences with greater precision on an index scale established by the user.
It is very important that this indicator does not represent vigor, since the latter refers to the development of the annual growth of the crop, which is why it is better to define it as a biomass index or crown volume in the case of fruit trees.
The logical step after this study is to be able to interpret agronomically what information would be generated by these indicators and to be able to relate it to other images such as orthophotos and thermal images in order to be able to reach a technical conclusion.
After examining the orthophoto and thermal image, the NDVI indicator image must be analyzed in situ since the difference in color scale does not always have a similar response. In this context, the levels that indicate “stress” do not always respond to the same problem. These defects may correspond to water stress, thermal stress, spider mite damage, soil structure, among others.
Once the main problem has been established, recommendations and actions must be generated specifically to solve this problem.
Image 4. A. Orthophoto; b. Thermal image; c. NDVI image. (Source Skyquest-UPL).
In conclusion, multispectral images allow us to:
- Gain a broad understanding of the state of the orchard's biomass, which will allow you to differentiate tasks and better optimize resources.
-They allow us to know the variations in the type of soil, therefore, the analysis can be focused through a better distribution of test pits or other tools that allow for agronomic diagnosis.
-Decrease or increase the consumption of water resources.
Obtaining the maximum benefit from the productive potential of the garden through qualifiable and quantifiable data, providing precision and support for different instructions and management that must be carried out in the crop.
