A Machine Learning Approach for Mapping Forest Vegetation in Riparian Zones in an Atlantic Biome Environment Using Sentinel-2 Imagery

Abstract

Riparian zones consist of important environmental regions, specifically to maintain the quality of water resources. Accurately mapping forest vegetation in riparian zones is an important issue, since it may provide information about numerous surface processes that occur in these areas. Recently, machine learning algorithms have gained attention as an innovative approach to extract information from remote sensing imagery, including to support the mapping task of vegetation areas. Nonetheless, studies related to machine learning application for forest vegetation mapping in the riparian zones exclusively is still limited. Therefore, this paper presents a framework for forest vegetation mapping in riparian zones based on machine learning models using orbital multispectral images. A total of 14 Sentinel-2 images registered throughout the year, covering a large riparian zone of a portion of a wide river in the Pontal do Paranapanema region, São Paulo state, Brazil, was adopted as the dataset. This area is mainly composed of the Atlantic Biome vegetation, and it is near to the last primary fragment of its biome, being an important region from the environmental planning point of view. We compared the performance of multiple machine learning algorithms like decision tree (DT), random forest (RF), support vector machine (SVM), and normal Bayes (NB). We evaluated different dates and locations with all models. Our results demonstrated that the DT learner has, overall, the highest accuracy in this task. The DT algorithm also showed high accuracy when applied on different dates and in the riparian zone of another river. We conclude that the proposed approach is appropriated to accurately map forest vegetation in riparian zones, including temporal context.

A Comparative Estimation of Maize Leaf Water Content Using Machine Learning Techniques and Unmanned Aerial Vehicle (UAV)-Based Proximal and Remotely Sensed Data

“…Decision tree regression (DTR): In this study, the fine-tuning process of the DTR algorithm was performed until no improvements were observed and the model parameters were specified. The minimum split, which is the minimum number of values that must exist at a node before the split is attempted [48], was fixed at 20. The maximum depth to which the tree is allowed to grow was set at 30.…”

Assessment of high spatial resolution satellite imagery for monitoring riparian vegetation: riverine management in the smallholding

“…Most studies analysing forest vegetation are based on the Sentinel-2 satellite (Krtalić et al, 2021). These are mainly based on the use of logarithms (Brovelli et al, 2020;Furuya et al, 2020) or vegetation indices (Hill, 2013;Nouri et al, 2014;Pace et al, 2021). The main advantage of these procedures is the use of lowcost images, in addition to the fact that they enable environmental monitoring and allow the adequacy of forest management decision-making.…”

Sử dụng ảnh vệ tinh sentinel 2 và thuật toán học máy thành lập bản đồ hiện trạng rừng tại huyện Bù Đăng, tỉnh Bình Phước

“…Xây dựng bản đồ hiện trạng rừng là một nhiệm vụ quan trọng, cung cấp cơ sở dữ liệu cho công tác quản lý và phát triển rừng (Gebhardt et al, 2014;Trung & Huong, 2019). Đặc biệt, kiểm kê hàng năm hoặc định kỳ hiện trạng rừng còn dùng để xác định tình trạng suy thoái do thiên tai hoặc do tác động của con người, từ đó đề xuất những chiến lược quản lý phù hợp (Luong et al, 2017;Furuya et al, 2020). Kiểm kê rừng có thể thực hiện bằng các phương pháp đo trên mặt đất, sử dụng ảnh vệ tinh, hoặc sử dụng kết hợp cả 2 phương pháp.…”