Abstract The intensive and dynamic land-use and land-cover changes in Mato Grosso, Brazil, and the environmental costs of adopted agricultural practices over the years have attracted the attention of researchers and institutions. In order to evaluate these aspects, moderate resolution remote sensing data and techniques have been applied. Using the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor, compatible with the farm size and seasonal dynamic of Mato Grosso agriculture, it is possible to acquire data for a large coverage, repeatedly and freely. In the context of Mato Grosso, which is a large geographical area with dynamic processes operating at multiple scales, this sensor allows vegetation monitoring and seasonal detection of changes in canopy. This review paper presents the use of MODIS sensor data to improve soybean crop detection, yield prediction and deforestation monitoring in Mato Grosso over time. Different data from this sensor were collected over several years, especially vegetation indices, and a large number of methods were developed through the MODIS-derived information. The presented results show how the MODIS sensor was successfully applied to local and regional soybean crop and deforestation monitoring, offering new opportunities for land-cover mapping in Mato Grosso.