Overview of methods and technologies used for smart management of pastures and meadows

Abstract

The present study presents a review of technologies and methods used for smart management of pastures and meadows. Initially, the sources of data used are summarized. They are divided into two major categories: remote sensing and ground sampling. Next, the study defines three major directions in the field: prediction of biomass, assessment of the degradation/erosion/soil quality and evaluation of the biodiversity. The performed review showed that studies that are aimed at evaluating the above ground biomass mostly rely on RGB and multispectral spatial data and ground sampling. The data processing includes estimation of different vegetation indices and application of classification/clustering/regression algorithms. The second direction is commonly aimed at creating erosion/degradation maps and assessing the impact of grazing. They usually rely on satellite images and field samples for training/verification. The data processing methods include creation of NDVI maps and application of classification algorithms. The third direction is aimed at creating classification models and maps representing the biodiversity of pastures and meadows. They rely on satellite and UAV obtained multispectral images and field samples. A wide range of vegetation indices and classification algorithms is used for the data processing. The obtained review analysis in this study could be useful for agronomists and engineers working in the field of precision animal husbandry and pasture/meadow management.