Research

Key findings:

• The CBR-PSO model incorporating Sentinel-1, Sentinel-2, and ALOS-2 PALSAR-2 data estimated the mangrove soil organic carbon (SOC) ranging from 44.74 to 91.92 Mg ha−1 (average = 68.76 Mg ha−1).  

• The proposed model yielded a satisfactory accuracy (R2 = 0.809, RMSE = 9.30 Mg ha−1) and outperformed the RF, the SVM, and the XGBoost models 

• Multisensor optical and SAR combined with the CBR-PSO model can be accurately estimated the mangrove SOC in the Red River Delta, Vietnam.

Key findings:

• The XGBR-GA model incorporating Sentinel-2, Sentinel-1, and ALOS-2 PALSAR-2 data to estimate the mangrove above-ground biomass (AGB), including small and shrub mangrove patches was proposed

• The proposed model yielded a promising result (R2 = 0.683, RMSE = 25.08 Mg·ha−1) and outperformed the four other ML models 

• Multisource optical and SAR combined with the XGBR-GA model can be used to estimate the mangrove AGB in North Vietnam.

Key findings:

• The XGBR model produced a promising result with R2 = 0.805, RMSE = 28.13 Mg ha-1, and produced the highest prediction performance among the five machine learning models.

• The XGBR model estimated mangrove AGB ranging between 11 and 293 Mg ha-1 (average = 106.93 Mg ha-1).

• A fusion of Sentinel-2 MSI and ALOS-2 PALSAR-2 sensors combined with the XGBR model can accurately estimate mangrove AGB in the tropics.

• RoF produced the highest overall accuracy than those of RF and CCF algorithms in mapping mangrove species and seagrass meadow

• The potential use of Sentinel multispectral and SAR data together with ensemble-based machine learning techniques to map mangrove species

Key findings

• Willingness to pay (WTP) is elicited by single-bound dichotomous choice model.

• Perceptional factors influencing the WTP of respondents are examined.

• Socio-demographic factors influencing the WTP for mangrove restoration are evaluated.

• The annual mean WTP per household is estimated as VND 192,780 (US$ 8.6).