Remote sensing by nanosatellites can help release agricultural credits

Geospatial technologies, notably remote sensing based on orbital data, help public and private companies in the environmental, forestry, energy, agriculture, among others, in solving their demands in an agile and efficient way. Through images captured by nanosatellites, it is possible to process, analyze and make geospatial information available, with systems that generate accurate and updated data. Another great advantage of remote monitoring is cost reduction and increased productivity. This is because companies that hire the solution are able to have more up-to-date information, build more efficient workflows, in addition to reducing operational costs, such as transportation, logistics and inspections by optimizing data collection in the field.

The use of remote sensing can also be a great help in the process of releasing agricultural credits, as it allows frequent monitoring of areas subject to financing by financial institutions, culminating in more reliable management of the financeable area and, consequently, maximizing control of operational risk and fraud (ghost areas). The observation comes from the master and doctor in Applied Geosciences and Geodynamics from UnB, Sumaia Resegue Aboud Neta, who defended the thesis “The use of remote sensing to mitigate risks in the financial system: an application for control and monitoring of agricultural production”, guided by professor Edilson Bias.

The researcher had as a starting point for the preparation of her thesis Resolution No. 4427/2015 of the Central Bank (Bacen), which recommends that banks and financial institutions, subject to rural credit operations, use remote sensing for contracting and supervision of agricultural credit operations. According to the Resolution, the non-adoption of remote inspection makes it impossible to carry out inspections by sampling, which, according to the researcher, “makes on-site inspection necessary in all agricultural enterprises, which can generate waste of time and resources, in addition to increasing operational costs due to the displacement of personnel, especially when it comes to remote regions, making the process more expensive”.

According to Sumaia, what made the work possible was the use of images from Planet nanosatellites to carry out monitoring, which were made available free of charge under the Technical Cooperation Agreement signed between UnB - University of Brasília and SCCON Geospatial - Santiago & Cintra Consultoria in March 2018. “Part of one of the greatest advances in orbital imaging, Planet nanosatellites have an extensive capacity for collecting large areas, with a high revisit rate, high radiometric, temporal and spatial definition, with daily repeatability of each scene. Ideal for crops with an annual cycle, as they allow for more frequent monitoring of the area”, says the doctor. “The processing of these images, linked to the use of big data and machine learning, allows farmers greater predictability”, she adds.

Nanosatellite technology is employed by Planet. There are more than 130 nanosatellites, weighing up to 5 kilos each, that cover the entire surface of the Earth, monitoring more than 300 million km2/day, with high radiometric (from 12 to 16 bits) and spatial (3 meters orthorectified) resolution and repeatability daily from the same point on the planet, with applications aimed at agriculture and deforestation control. Because they are small and light, which makes them easy to launch into space, they have more affordable costs when compared to traditional satellites.

According to Sumaia, to design the study, a method was proposed and an open source tool was developed to detect changes in the patterns of agricultural areas, which included three models: Peaks and Valleys (PV), Normalized Difference Vegetation Index ( NDVI) and Band Difference (DB). The models provided information about changes in the monitored areas, such as vegetation growth or loss. With the data obtained and analyzed, it was possible to assess the quality of the models and verify the viability of the proposed objective of using remote sensing to mitigate risks in the Financial System, monitoring the production cycles of each type of agricultural product, without necessarily having in-house visits. crazy.