USING SENTINEL SATELLITE IMAGE TO ESTIMATE BIOMASS OF MANGROVE FOREST IN VINH QUANG COMMUNE, TIEN LANG DISTRICT, HAI PHONG CITY


Authors

  • Tran Quang Bao Vietnam National University of Forestry
  • Le Sy Hoa Vietnam National University of Forestry

Keywords:

Biomass, mangroves, Sentinel 1, Sentinel 2, vegetation indices

Abstract

This research tested the use of geographic information systems using Sentinel 1 and Sentinel 2 satellite data to estimate biomass mangrove forest in Vinh Quang commune, Tien Lang district, Hai Phong province. 15 sample plots (10 m × 10 m) in the field were established for making models and evaluation, the satellite images for processing in 2017 were provided freely by ESA Corporation. The study created land cover and biomass maps from field allometric equations and estimated results from the model by maximum likelihood classification and the regression model, respectively. For land cover accuracy assessment, Kappa index was employed with 93% accuracy. NDVI, SAVI were representative indices of optical Sentinel 2 images, similarly, VV and VH backscatter, VV/VH and VH/VV from Sentinel 1A images. The study showed that Sentinel 1 backscatters were unable to generate model due to quite low R2. Compare to optical images, the NDVI index was used for biomass estimating, the total biomass was about 67,983.12 tons, average: 153.94 ± 27.01 ton/ha, maximum: 223.14 ton/ha. By comparing real numbers and estimated numbers, the results were acceptable, 23.8% average. We conclude that the optical Sentinel 2 has been more suitable to make estimating the model for mangrove biomass at a small-scale level, especially for commune level.

Metrics

Metrics Loading ...

Downloads

Abstract View: 11
PDF Downloaded: 7

Published

26-10-2018

How to Cite

Quang Bao, T., & Sy Hoa, L. (2018). USING SENTINEL SATELLITE IMAGE TO ESTIMATE BIOMASS OF MANGROVE FOREST IN VINH QUANG COMMUNE, TIEN LANG DISTRICT, HAI PHONG CITY. Journal of Forestry Science and Technology, (5), 071–079. Retrieved from https://jvnuf.vjst.net/en/article/view/911

Issue

No. 5 (2018)

Section

Resource management & Environment

Most read articles by the same author(s)