In this study, we examined the natural recovery of mangroves in abandoned shrimp ponds located in the Wunbaik Mangrove Forest (WMF) in Myanmar using artificial neural network (ANN) classification and a change detection approach with Sentinel-2 satellite images. In 2020, we conducted various experiments related to mangrove classification by tuning input features and hyper-parameters. The selected ANN model was used with a transfer learning approach to predict the mangrove distribution in 2015. Changes were detected using classification results from 2015 and 2020. Naturally recovering mangroves were identified by extracting the change detection results of three abandoned shrimp ponds selected during field investigation. The proposed method yielded an overall accuracy of 95.98%, a kappa coefficient of 0.92, mangrove and non-mangrove precisions of 0.95 and 0.98, respectively, recalls of 0.96, and F1 scores of 0.96 for the 2020 classification. For the 2015 prediction, transfer learning improved model performance, resulting in an overall accuracy of 97.20%, a kappa coefficient of 0.94, mangrove and non-mangrove precisions of 0.98 and 0.96, respectively, recalls of 0.98 and 0.97, and F1 scores of 0.96. The change detection results showed that mangrove forests in the WMF slightly decreased between 2015 and 2020. Naturally recovering mangroves were detected at approximately 50% of each abandoned site within a short abandonment period. This study demonstrates that the ANN method using Sentinel-2 imagery and topographic and canopy height data can produce reliable results for mangrove classification. The natural recovery of mangroves presents a valuable opportunity for mangrove rehabilitation at human-disturbed sites in the WMF.

Information regarding land use and land cover (LULC) is essential for regional land and forest management. The contribution of reliable LULC information remains a challenge depending on the use of remote sensing data and classification methods. This study conducted a multiclass LULC classification of an intricate mangrove ecosystem using the U-Net model with PlanetScope and Sentinel-2 imagery and compared it with an artificial neural network model. We mainly used the blue, green, red, and near-infrared bands, normalized difference vegetation index (NDVI), and normalized difference water index (NDWI) of each satellite image. The Digital Elevation Model (DEM) and Canopy Height Model (CHM) were also integrated to leverage the model performance in mixed ecosystems of mangrove and non-mangrove forest areas. Through a labeled image created from field ground truth points, the models were trained and evaluated using the metrics of overall accuracy, Intersection over Union, F1 score, precision, and recall of each class. The results demonstrated that the combination of PlanetScope bands, spectral indices, DEM, and CHM yielded superior performance for both the U-Net and ANN models, achieving a higher overall accuracy (94.05% and 92.82%), mean IoU (0.82 and 0.79), mean F1 scores (0.94 and 0.93), recall (0.94 and 0.93), and precision (0.94). In contrast, models utilizing the Sentinel-2 dataset showed lower overall accuracy (86.94% and 82.08%), mean IoU (0.71 and 0.63), mean F1 scores (0.87 and 0.81), recall (0.87 and 0.82), and precision (0.87 and 0.81). The best-classified image, which was produced by U-Net using the PlanetScope dataset, was exported to create an LULC map of the Wunbaik Mangrove Area in Myanmar.

Estimating Above Ground Biomass (AGB) of mangroves provides crucial information for regional and global blue carbon strategies. Despite growing interest in mangrove AGB estimation using a remote sensing approach, the effectiveness of this advanced method has not been evaluated for mangrove forests in Myanmar. This study estimated the AGB of Wunbaik Mangrove Forest (WMF) in Myanmar using machine learning models with data from Sentinel-2, Sentinel-1, and Advanced Land Observing Satellite Phased Arrayed L-band Synthetic Aperture Radar (ALOS-PALSAR) HH and HV polarizations and Canopy Height Model. Reference AGB information was derived from field inventory plots via an allometric equation. Initially, we tested machine learning models such as Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boost (XGBoost). The results revealed that the RF model outperformed the other models, yielding a higher coefficient of determination (R²) of 0.33 and a Root Mean Square Error (RMSE) of 32.05 Mg ha⁻¹. We then set up different scenarios to improve the performance of the RF model for AGB estimation. Through different feature selection approaches, we identified features highly correlated with the field AGB, enhancing the RF model’s performance and resulting in the highest improvement in R²: 0.48 and RMSE: 28.12 Mg ha⁻¹. In this study, an AGB map of the WMF was generated using the best RF model. The predicted AGB distribution ranged from 22.266 Mg ha⁻¹ to 181.948 Mg ha⁻¹ in 2019. Compared with global datasets such as GEDI L4B and ESA CCI predictions, the proposed method provides a more accurate prediction of AGB for Wunbaik Mangrove, enhancing blue carbon information in Myanmar.

Accurate soil information is crucial for sustainable agricultural planning and land management, particularly in data-scarce regions, such as the Sudan Savanna, the largest sorghum-producing area in Africa. A recent study reported that soils in this region corresponded well with the topography, having formed primarily through erosion–deposition processes, resulting in systematic variation in soil types along the landscape. Therefore, this study compared the performances of three machine learning models, i.e., Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM), for soil classification based on multisource remote sensing and topographic data. Ground-truth data with four different soil types, Lixisols, Petric Plinthosols, Pisoplinthic Petric Plinthosols, and Gleysols, were used to train and validate the models using 19 remote sensing-derived covariates including Sentinel-1 SAR, Sentinel-2 bands, spectral indices, and Topographic Wetness Index. Machine learning classification was analyzed under different scenarios of remote sensing feature combination. Results showed that the XGBoost with the selected feature combination achieved the highest performance with an overall accuracy of 78.9%, followed by RF (72.3%) and SVM (65.2%). Among the selected features, topographic parameters appeared the most important and provided complementary information for accurate soil classification. This study demonstrates the effectiveness of integrating optical, radar, and topographic information for soil mapping and provides a valuable management tool to support agricultural and environmental strategies in the Sudan Savanna.