Citation: Lin, C.; Doyog, N.D. Challenges of Retrieving LULC Information in Rural-Forest Mosaic Landscapes Using Random Forest Technique. Forests 2023, 14, 816. https://doi.org/10.3390/f14040816 Academic Editors: Jan Bocianowski and Cate Macinnis-Ng Received: 31 January 2023 Revised: 30 March 2023 Accepted: 13 April 2023 Published: 15 April 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Article Challenges of Retrieving LULC Information in Rural-Forest Mosaic Landscapes Using Random Forest Technique Chinsu Lin 1, * and Nova D. Doyog 1,2 1 Department of Forestry and Natural Resources, National Chiayi University, 300 University Road, Chiayi 600035, Taiwan 2 College of Forestry, Benguet State University, LaTrinidad 2601, Benguet, Philippines * Correspondence: chinsu@mail.ncyu.edu.tw Abstract: Land use and land cover (LULC) information plays a crucial role in determining the trend of the global carbon cycle in various fields, such as urban land planning, agriculture, rural management, and sustainable development, and serves as an up-to-date indicator of forest changes. Accurate and reliable LULC information is needed to address the detailed changes in conservation-based and development-based classes. This study integrates Sentinel-2 multispectral surface reflectance and vegetation indices, and lidar-based canopy height and slope to generate a random forest model for 3-level LULC classification. The challenges for LULC classification by RF approach are discussed by comparing it with the SVM model. To summarize, the RF model achieved an overall accuracy (OA) of 0.79 and a macro F1-score of 0.72 for the Level-III classification. In contrast, the SVM model outperformed the RF model by 0.04 and 0.09 in OA and macro F1-score, respectively. The accuracy difference increased to 0.89 vs. 0.96 for OA and 0.79 vs. 0.91 for macro F1-score for the Level-I classification. The mapping reliability of the RF model for different classes with nearly identical features was challenging with regard to precision and recall measures which are both inconsistent in the RF model. Therefore, further research is needed to close the knowledge gap associated with reliable and high thematic LULC mapping using the RF classifier. Keywords: forest classification; mapping; forest degradation; agriculture; machine learning; sustainability 1. Introduction Capturing and reporting the status of earth’s land use and land cover (LULC) is indispensable in reducing emissions from deforestation and degradation (REDD) and in mitigating global warming as it plays a crucial role in determining the trends of the global carbon cycle [1–4]. Areas of forest coverage, healthy and structural components such as forest types and species, are critical information for forest monitoring and sustainable management [5]. LULC conversion is a complex process that involves anthropogenic activ- ities and biological, environmental, and meteorological factors, and could also directly or indirectly affect its environment in terms of climate, global climate change patterns, services such as economy, biodiversity, forest growth, food, and water cycle [6,7], and aesthetic and economic value [8]. Change in LULC could hasten the risk of natural hazards due to the loss of protective services that LULC offers and could also lead to increased fragmentation and structural destruction of protected areas [9]. Diverse degrees of LULC changes can significantly affect the ecosystem as this interferes with natural ecological processes, such as nutrient cycling, water cycle, energy flow, and succession. The changes that provide critical information to address and control the impacts should be measured and assessed. From the point of view of forest resources assessment, forest inventory must be reliable in measurement, report, and validation [10]. Derivation of accurate LULC to disclose details of forest status is of particular importance in achieving sustainable management. Accurate mapping of LULC is highly related to the resolutions and features inherited from the diverse remotely sensed data delivered by spaceborne, airborne, and drone Forests 2023, 14, 816. https://doi.org/10.3390/f14040816 https://www.mdpi.com/journal/forests