Discrimination of Mangrove Stages Using Multitemporal Sentinel-1 C-Band Backscatter and Sentinel-2 Data-A Case Study in Samut Songkhram Province, Thailand

Discrimination of mangrove stage changes is useful for the conservation of this valuable natural resource. However, present-day optical satellite imagery is not fully reliable due to its high sensitivity to weather conditions and tidal variables. Here, we used the Vertical Transmit-Vertical Receive Polarization (VV) and Vertical Transmit-Horizontal Receive Polarization (VH) backscatter from the same and multiple-incidence angles from Sentinel-1 SAR C-band along with Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Normalized Difference Moisture Index (NDMI), Normalized Difference Red Edge (NDVIRE) and Chlorophyll Index Green (CIGreen) from the optical satellite imageries from Sentinel-2 to discriminate between the changes in disturbance, recovery, and healthy mangrove stages in Samut Songkhram province, Thailand. We found the mean NDVI values to be 0.08 (+/- 0.11), 0.19 (+/- 0.09), and -0.53 (+/- 0.16) for the three stages, respectively. We further found their correlation with VH backscatter from the multiple-incidence angles at about -17.98 (+/- 2.34), -16.43 (+/- 1.59), and -13.40 (+/- 1.07), respectively. The VH backscatter from multiple-incidence angles was correlated with NDVI using Pearson's correlation (r(2) = 0.62). However, Pearson's correlation of a single plot (ID2) of mangrove stage change from disturbance to recovery, and then on to the healthy mangrove stage, displayed a r(2) of 0.93 (p value is less than 0.0001, n = 34). This indicated that the multitemporal Sentinel-1 C-band backscatter and Sentinel-2 data could be used to discriminate mangrove stages, and that a reduced correlation to significant observations was the result of variations in both optical and SAR backscatter data.

Automated summary

The study successfully differentiated changes in different stages of mangroves in Samut Songkhram province, Thailand using data from Sentinel-1 and Sentinel-2 satellites, demonstrating the utility of multi-temporal data in evaluating mangroves.