The “startup” born in Córdoba that works with NASA

Those who transit the routes that cross the city of Marcos Juárez, in the southeast of Córdoba, find signs that describe this place as the “productive heart of the country”, because it is one of the best areas in Argentina for agricultural production.

It is no surprise, then, that he was born there from one of the agtech (technological solutions for the agricultural sector) with potential in the Argentine market: Sima (Integrated System for Agricultural Monitoring).

Founded by Cordovan Andrés Yerkovich (today CEO of the company) and Gerónimo Oliva, together with Agustín Rocha, Pablo Etchanchu and Mauricio Varela, this startup It already evaluates productive conditions in more than 300 thousand fields and now it is on a new leap.

It is that it sealed an alliance with the division of the National Administration of Aeronautics and Space of the United States (NASA, for its acronym in English) dedicated to bringing solutions for agricultural production.

This unit is called Nasa Harvest (harvest, in English) and is coordinated by a team of researchers from the University of Maryland, in the United States.

Artificial intelligence

The objective of the agreement is to improve the tools used to predict the yields (number of kilos harvested per hectare) of crops, by combining the information that can be obtained from satellite images, captured by NASA, with the volume of data The field that Sima owns, through its users.

Lucas Fontana is an expert in data science and is one of the collaborators who is part of the Sima team and who is behind this ambitious project.

“In the market there are already many solutions that offer crop monitoring. With this link with Nasa Harvest, the objective is to obtain a differential with respect to the competition, an added value through information that allows us to make more accurate performance predictions, thanks to artificial intelligence, ”he explains.

Until now, Sima has been offering its users the possibility of calculating yields based on the experience and knowledge of agricultural engineers.

The service will be enhanced with the use of the images provided by the Sentinel 1 and 2 satellites, and Landsat 8.

“It is not only the image, but we are going to contextualize it with the data that the producers upload. For example, what diseases and pests did they have on the crops, what products did they apply, what varieties of seeds did they use, at what distance and at what depth did they sow them, and the climatic conditions of temperatures and rainfall that were recorded ”, Fontana describes.

And it completes: “What we are generating is a predictive model using artificial intelligence, with the complement of indicators obtained in the field. It is not a crystal ball, but it is a more effective reference value ”.

At Sima they calculate that with this new tool they will be able to achieve predictions 20 percent more accurate than those made without the support of Maryland technicians.

The model is already created and tested, and all that remains is to incorporate it into the platform used by Sima users.

Fontana estimates that it will be available in a maximum period of two months.

In principle, it will only be enabled to predict the yields of wheat, corn and soybeans in Argentina.

In the medium term, the idea is to incorporate other crops, such as potatoes or rice; and other regions, such as Paraguay and Brazil.