The application of machine learning (ML) and remote sensing (RS) in soil and water conservation has become a powerful tool. As analytical tools continue to advance, the variety of ML algorithms and RS sources has expanded, providing opportunities for more sophisticated analyses. At the same time, researchers are required to select appropriate technologies based on the research objectives, topic, and scope of the study area. In this paper, we present a comprehensive review of the application of ML algorithms and RS that has been implemented to advance research in soil and water conservation. The key contribution of this review paper is that it provides an overview of current research areas within soil and water conservation and their effectiveness in improving prediction accuracy and resource management in categorized subfields, including soil properties, hydrology and water resources, and wildfire management. We also highlight challenges and future directions based on limitations of ML and RS applications in soil and water conservation. This review aims to serve as a reference for researchers and decision-makers by offering insights into the effectiveness of ML and RS applications in the fields of soil and water conservation.

A smart thinning operation refers to an advanced method of selecting and cutting trees to be thinned based on digitally captured forest information. In smart thinning operations, workers use the coordinates of individual trees to navigate to the target trees for thinning. However, it is difficult to accurately locate individual trees in a forest stand covered with a canopy, necessitating a precise real-time positioning system that can be used in the forest. Therefore, this study aimed to evaluate the applicability of the global navigation satellite system real-time kinematic (GNSS-RTK) device in a forest stand through analysis of its positioning accuracy within the forest environment and evaluation of the operational range of the single-baseline RTK based on analysis of the positioning precision and radio signal strength index (RSSI) change with increasing distance from the base station. The results showed that the root mean square error (RMSE) of the horizontal positioning error was highly accurate, with an average of 0.26 m in Larix kaempferi stands and 0.48 m in Pinus koraiensis stands. The RSSI decreased to a minimum of −103.3 dBm within 1 km of distance from the base station; however, this had no significant impact on the horizontal positioning precision. The conclusion is that the GNSS-RTK is suitable for use in smart thinning operations.

The United Nations Framework Convention on Climate Change (UNFCCC) recognized the importance of forests in combating climate change and agreed upon financial support for REDD+ activities in developing countries through the Warsaw REDD+ Framework (WRF). The REDD+ activities for conserving carbon stored in forests to reduce GHG emissions and to enhance the carbon sink function of forests are expected to serve as an important means for achieving the climatic goal. In this study, a set of criteria was devised to assess the REDD+ Measuring, Reporting, and Verifying (MRV) implementation capabilities of developing countries, which was applied to analyze the REDD+ MRV levels in REDD+ countries. Based on the Forest Reference Emission Level/Forest Reference Level (FREL/FRL), National Strategy (NS), National Forest Monitoring System (NFMS) and Safeguard Information Summary (SIS) the countries submitted, 36 REDD+ countries submitted only the FREL/FRL (Group I), while 5 countries fulfilled the WRF requirements and registered REDD+ reduction results (Group II), and 6 received Results-Based Payments (RBP) (Group III). From longest to shortest, the periods for which the International Fund provided support were arranged in the order of Group III, II, and I, verifying the relative importance of international support. From highest to lowest, the overall MRV capability was also arranged in the order of Group III, II, and I, although Group I or Group II was at a higher level than the other groups in some elements. REDD+ countries in the Readiness Phase (Group I) would aim to have the MRV capabilities of Groups II and III to receive RBP, and international support for REDD+ MRV capacity building could enable them to do. However, in addition to the receipt of RBP, REDD+ should be reflected in the Nationally Determined Contribution (NDC) as consistent reduction results at the national GHG inventory level, and the advancement of REDD+ MRV is expected to be a necessary and sufficient condition for REDD+ cooperation under the Paris Agreement cooperative approach framework. For the following groups, international cooperation is essential. Countries in the Readiness Phase need to be supported with the establishment of an MRV framework, which will enable them to achieve REDD+ to receive RBP and be reflected in NDCs. For REDD+ countries that have thus far met the WRF requirements, the REDD+ scope needs to be upscaled to national levels, and the MRV system should be further advanced to establish a cooperative approach system that can achieve more ambitious reduction targets through forests.

The application of machine learning (ML) and remote sensing (RS) in soil and water conserva-tion has become a powerful tool. As analytical tools continue to advance, the variety of ML al-gorithms and RS sources has expanded, providing opportunities for more sophisticated analyses. At the same time, researchers are required to select appropriate technologies based on research objectives, topic, and scope of the study area. In this paper, we present a comprehensive review on the application of ML algorithms and RS that has been implemented to advance research in soil and water conservation. The key contribution of this review paper is that it provides an overview of current research areas within soil and water conservation and their effectiveness in improving prediction accuracy and resource management in categorized subfields, including biomass-vegetation, soil properties, hydrology and water resources, and wildfire management. We also highlight challenges and future directions based on limitations of ML and RS applica-tions in soil and water conservation. This review aims to serve as a reference for researchers and decision-makers by offering insights into the effectiveness of ML and RS applications in the field of soil and water conservation.

In artificial forests, regular thinning is required to promote diameter growth for producing high-quality, large-diameter timber. However, selecting trees for thinning often relies on qualitative and subjective assessments by field workers. The purpose of this study was to develop a quantitative method for selecting trees for thinning by combining machine learning algorithms and competition indices. Our study site included the Pinus koraiensis area within a Kangwon National University research forest in the Republic of Korea. Data from a model development site were used for the basic crown classification model for Pinus koraiensis. The model was optimized by adjusting hyperparameters. Different algorithms, including Random Forest, XGBoost, and LightGBM (LGBM), were improved using Random Search. LGBM showed the highest accuracy of 71.6%. LGBM—in combination with the competition indices—was used to classify the crown class in the application site and select trees for thinning. Compared to the combination of Braathe and Martin-EK indices, the combination of LGBM and Hegyi index enabled the even distribution of the residual stand in the entire site after thinning. It lowered the distribution of hot spots, which represent competition. Thus, the combination of LGBM and Hegyi index was the most effective option to improve the spatial distribution of trees after thinning. Our findings can improve forest management by providing a quantitative and objective method for selecting trees for thinning.

This study developed a forest management plan model using reinforcement learning (Q-learning) to optimize both the economic and ecological functions of forests. Management objectives for national forests were established, and forest conditions were analyzed using GIS spatial data and administrative records. A 60-year forest management plan was formulated to predict timber production and management performance across different regions and time periods. Our analysis revealed that Scenario 3 (Carbon Storage Priority) demonstrated the highest economic value, starting at approximately KRW 576.2 billion in the initial period and escalating to KRW 775.7 billion over six 10-year periods, totaling 60 years. In addition to its economic performance, Scenario 3 effectively improved forest age class structure and ensured a stable timber supply, making it the most balanced approach for sustainable forest management. By focusing on carbon storage as a key management goal, this approach highlights the potential for achieving both economic and environmental benefits concurrently. These results suggest that reinforcement learning is a powerful tool for developing long-term forest management strategies that address multiple objectives, including economic viability, ecological sustainability, and resource optimization.