Abstract
This article presents MontadoDB, a dataset comprising samples of pasture quality parameters, weather data, and satellite images from a set of paddocks in the Alentejo region (southern Portugal), the Beira Interior region (Portugal), and Extremadura (Spain). These regions are part of the Montado system, an agro-silvo-pastoral system featuring low-density trees, pastures, and livestock grazing. The pasture parameters were collected along three vegetative cycles from 2018 to 2021 in 8 different paddocks. After the collecting procedure, pasture samples were submitted to laboratory analysis to determine the reference values of pasture moisture content (PMC, in %), crude protein (CP, in g/100 g), and neutral detergent fiber (NDF, in g/100 g), using standard analytical methods. For each pasture sample collected, GPS coordinates were recorded. Using these coordinates, the Google Earth Engine (GEE) platform was utilized to collect satellite multispectral data and weather information. The MontaDB dataset is publicly available, enabling research focused on the analysis and design of machine learning models for predicting pasture parameters. It can also work as a reference dataset for further experiments based on pasture samples, thereby enhancing research on advanced algorithms that rely on large-scale datasets.