Unprecedented IAC research uses aerial images to estimate soil erosion in crops

An unprecedented survey in Brazil is being developed at the Agronomic Institute (IAC-APTA), linked to the São Paulo Department of Agriculture and Supply, in Campinas, with the objective of estimating soil erosion in areas covered by plantations. The study is carried out using aerial images, generated by Unmanned Aerial Vehicles (UAVs), also known as drones. This technology uses basic principles of aerial photogrammetry and is already used to generate diverse information, such as calculating the volume of reservoirs, determining the number of plant species in a certain region, identifying flaws in crops and also to estimate agricultural production, through biomass calculation. However, the application of technology to estimate the soil erosion process where crops are already installed requires new knowledge and has never been conducted in the country. This research by IAC, from the Department of Agriculture and Supply of the State of São Paulo, is at the frontier of knowledge, according to the researcher responsible, Bernardo Cândido, who has already validated this technique on bare soil during his doctoral research.

“The challenge is to be able to apply aerial mapping technology to estimate erosion in an already cultivated area; on a larger scale it is a challenge, no one has advanced in this direction yet”, says Cândido, who joined the IAC in March 2019, as a post-doctor, with resources from the São Paulo State Research Support Foundation (FAPESP).

The technology enables image processing in specific software, generates relief maps of the area and land use, in addition to qualifying the most suitable region for each activity, for example, building a water reservoir or installing crops.

According to the researcher, with aerial photogrammetry it is possible to reconstruct the ground surface from moving images, a technique internationally known as structure-from-motion. “We were able to do the mapping in 3D”, says Cândido. In the research at IAC, a low-cost quadcopter drone is being used, and a state-of-the-art fixed wing imported from Switzerland, for around R$270. The fixed wing is accompanied by a thermal camera, which makes it possible to estimate the water content in the plant, and aims to monitor the water stress of crops at different times.

At IAC, this study will be specific to citrus. According to the researcher, in South America there are few units of this camera being used. The researchers intend to evaluate how the experiments are conducted and the accuracy of the measurements in two conditions: when using a more accessible vehicle and another with a high cost.

The team is defining the areas where experiments will be carried out with these resources to evaluate agricultural productivity and monitor the erosion process in the cultivation areas. In Catanduva, in the interior of São Paulo, experiments were set up to estimate water erosion in sugarcane transporters, using UAVs and aerial images.

The IAC group maintains exchanges with scientists from Germany, the United States, England and Spain, who are also interested in information about erosion in planted areas. “How to quantify erosion in real time, in a cultivated area, with low-cost equipment?” These and other questions will have to be answered over four years of studies, one of which will be conducted in Germany, at the Dresden Institute of Technology.

Research at IAC is aimed at promoting the advancement of this technology not only in soil erosion, but also in other aspects of agriculture. “A project is already underway to validate the production estimate for citrus and sugar cane, but it is still in its infancy,” says Cândido.

Uses of photogrammetry

The study involving water erosion in covered soil is unprecedented. However, mapping using aerial images is already adopted in agriculture and other segments. For example, in Brazil, there are producers who already adopt fixed wing to identify flaws in sugarcane planting and to assess plant vigor. “They use it, but there is no scientific data that validates the information generated by the vehicles”, says the researcher, highlighting that, in Brazil, the IAC research will bring the first scientifically proven data.

“We are going to accurately validate and establish the criteria to make reliable estimates, bring a scientific look to this technology already used in the market”, highlights Cândido, who developed the first thesis in Brazil to monitor and quantify water erosion from aerial images.

According to the researcher, it is possible to quantify the volume of biomass in sugarcane and estimate the relationship between productivity and biomass. With this technology, it is possible to quickly estimate production and the harvest point. “It will be very useful for staggering the harvesting period at plants,” he says.

The generation of aerial images to produce mapping can also be obtained via satellite. However, there are more benefits to using drones and other UAVs. One of the gains is in the creation of 3D structures, a resource that makes it possible to calculate volumes, while the satellite only produces in 2D. “This resource allows you to estimate what volume of water is needed to fill a dam, for example, or the amount of sugar cane to be harvested or even what the total number of trees or the percentage of a certain species in an area, in the case of a forest inventory”, he explains.

Other benefits are the speed provided by UAVs and the possibility of mapping at all times, with much greater frequency. “The satellite records the image from time to time and if there is a cloud at that moment, the record is lost”, he explains. The image resolution quality is also superior when using a drone, because it flies closer to the ground. In addition to these gains, drones have a low cost, there is equipment for R$2.

Technical cooperation with Mali

Researcher Bernardo Cândido is also participating in a project in Mali, the African country chosen to host the pilot technical cooperation project between Brazil and Mali, which involves the Federal University of Lavras (UFLA) and the IAC, with funding from the Brazilian Agency for Cooperation (ABC), with federal resources. The project aims to increase cotton production in that country, which is the largest cotton producer in Africa and one of the largest in the world, with 728 thousand tons produced in 2018.

In Mali, the soils are acidic, similar to those of some Brazilian regions, such as Tocantins and Goiás. However, the country has soil analysis methods based on French protocols, a country that has different biomes than the African savannah. There, there is still the problem of water infiltration into the soil, due to the presence of subsurface impediment layers, causing flooding and erosion in times of rain.

To develop the project, two regions were selected: Bandiagara II and Siani, the latter being less needy than the first. “We mapped the two regions to guide the best use of the soil, which is stony, and we are going to guide them to build a dam to store rainwater”, says the researcher.

The water reserve will be used to manage crops and prevent the appearance of puddles, outbreaks of disease among children. “The intention, with the technology, is to reduce soil erosion, which results in the accumulation of water during the rainy season and consequent transmission of diseases to children, who circulate in standing water”, explains the researcher from the IAC, from Agência Paulista of Agribusiness Technology (APTA).

The soil of Bandiagara II resembles that of the Brazilian Cerrado. “Here we know precise soil analysis methodologies for tropical soils, this knowledge we will take to Mali”, says Cândido. In Mali, it is hot and humid, temperatures reach 45ºC in the sun and 35ºC at night, the drought is severe, except for the rainy season, from June to August.
In Bandiagara II, where the technical level is very low and lower than that of Siani, around 400 people live, half of which are children. According to the project's medical team, practically all of them have anemia and the mortality rate is 50% up to two years of age.

“Our goal is to increase cotton productivity, which is the main local crop, and also make the soil more productive, benefiting other crops; this could impact the life of the community, which suffers from malnutrition”, says Cândido.
In October 2019, the IAC researcher was in Mali, where he carried out aerial mapping for 24 photos. These three-dimensional, high-resolution images will provide information to generate useful design tools. The drone mapped around 711 hectares in Bandiagara II and 1600 hectares in Siani. “We mapped the entire area where the population lives to analyze how the land is being used and point out the suitability of the soil for each crop”, he says.

Based on these data, protocols will be developed to analyze the soils, similar to the Brazilian ones. “We intend to enrich the soil so that this reflects on the nutrition of the population”, he emphasizes. The local population also produces maize, millet, cassava, cashew, potatoes and vegetables.

In March 2020, the first diagnosis should be ready, which will define the interventions to be carried out, such as drawing up maps of the soil level curve, possible change of planting areas and readjustment according to land use procedures. “We intend to establish water catchment areas and seedling nurseries, which can be planted by the population”, he says. This step should be implemented in July 2020.

Before that, next February, Cândido will return to Mali to give a course on aerial mapping to professionals from the Companhia Malinense de Desenvolvimento Têxtil (CMDT), which buys all cotton production in Mali. “They will have access to information about soils, from pedology to fertility”, he summarizes.