Coastal Objects: Mangrove Area Extraction Using Remote Sensing and Aerial LiDAR Data in Roxas, Oriental Mindoro

Abstract

The Phil-LiDAR 2 program aims to extract the natural resources of the Philippines from the available two points per square meter LiDAR data. Mangroves, being coastal resources, were one of the foci of this program under the Aquatic Resources Extraction from LiDAR Surveys (CoastMap). The object-based image analysis (OBIA) approach, and support vector machine (SVM) algorithm were utilized to classify three major classes from the LiDAR data, namely: mangrove, other vegetation, and non-vegetation. Object feature values used in the classification include the mean, standard deviation, mode, and texture values from the generated LiDAR derivatives. These derivatives include the Digital Surface Model (DSM), Digital Terrain Model (DTM), Canopy Height Model (CHM), Intensity, Number of Returns, Normalized DSM (NDSM), Slope, and Slope of Slope. Moreover, field data collection and validation provided key references in the supervised SVM classification and contextual editing of the extracted mangrove areas. From the implemented classification, an overall accuracy of above 90% was achieved. Focusing with the final classified mangrove coverage, management of the mangrove resources can be made proper and efficient. Furthermore, high resolution or detailed spatial information can support programs like Reducing Emissions from Deforestation and forest Degradation Plus (REDD+) and biodiversity studies.

Mangrove Resource Mapping Using Remote Sensing in the Philippines: A Systematic Review and Meta-Analysis

“…Long et al [44] have achieved an overall accuracy of 93% in using Landsat images for mangrove forests using decision tree classification only. Researchers in mangrove RS tend to use SVM [80,88], MLC [75], RF [81], and ANN [53], among others. Since these classifiers are non-parametric, they do not need the requirements for normality.…”

“…The study clarified that SVM has shown to be more effective (99%) in discriminating mangroves from other land cover. This algorithm is particularly promising when used in mangrove classification using high-resolution datasets such as LiDAR derivatives [79,86,87] compared to the results of Songcuan et al [88] and Gevaña et al [77] (92-95%) using low-resolution datasets. This study supports the findings of Pham et al [30], which showed that among the machine learning algorithms, SVM produced higher overall accuracy in both cases.…”

Multidecadal mangrove forest change in Macajalar Bay, Northern Mindanao, Philippines (1950–2020) using remote sensing and geographic information systems

“…Recent studies that document the spatial extent of mangrove forest in the country integrates very highresolution aerial data to provide ne-resolution mangrove distribution maps (Luna et al, 2017;Cabili et al, 2018). While it yields higher accuracy and better image resolution results compared to previous satellite-derived mangrove forest maps, it is often limited to smaller spatial scales due to the limited availability of high-resolution images.…”

The Determination of Priority Areas for the Restoration of Degraded Tropical Peatland Using Hydrological, Topographical, and Remote Sensing Approaches

“…These restorative activities need to be implemented according to a high-resolution map for the entire landscape of the PHU [16]. As the map will cover an extensive area, the restoration of the whole PHU area be costly as well as require high levels of energy, and time [22]. Therefore, it may be necessary to divide the tropical PHU into individual sub-PHUs (sub-domes or mini-domes), independently identifying a smaller space but still part of the whole hydrological system [15].…”