Remote Sensing Analysis of Seagrass Beds in Bontosua Island, Spermonde Archipelago

MarSave International Symposium 2018: "Strengthening Marine Resilience for Sustainable Development Goals"

生态环境科学

Thalib, M.S.^1 ; Faizal, A.^1 ; La Nafie, Y.A.^1

Department of Marine Science, Faculty of Marine Science and Fisheries, Universitas Hasanuddin, Makassar

90245, Indonesia^1

关键词: Benthic habitats;    Direct observations;    Remote sensing analysis;    Remote sensing information;    Satellite imagery data;    Seabed classifications;    Seagrass habitats;    Spatial informations;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/253/1/012047/pdf
DOI  :  10.1088/1755-1315/253/1/012047

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

Research on mapping and monitoring of seagrass ecosystems using satellite imagery data is still lacking. The availability of spatial information on benthic habitat is becoming very important with the increasing awareness of environmentally based management. Various methods of analysing remote sensing information on benthic habitat have been developed and utilized, such as the application of the Lyzenga algorithm. This algorithm requires data on the variation of water depth in the coastal areas to be mapped. The purpose of this research was to study the effect of using Lyzenga algorithm on the mapping of coral reef / seagrass ecosystem by comparing the results of seabed classification from Sentinel-2a imagery processed and without using the Lyzenga algorithm. Image classification with Lyzenga algorithm is easily recognizable with the Lyzenga index value format that has been freed from the effect of water depth. In this study, benthic cover in the shallow waters around Bontosua Island, Spermonde Archipelago-Indonesia, was classified using six habitat types: aquatic, land, sand, rubble, coral and seagrass. Seagrass habitat was further classified based on percentage cover as class 4 (76-100%), class 3 (51-75%), class 2 (26-50%) and class 1 (0-25%). Based on the analysis of Sentinel-2a data, the area with seagrass cover was estimated at 31.27 Ha. Seagrass mapping using medium-resolution satellite imagery can be helpful in providing data over a more extensive area than is likely to be possible using direct observation in the field alone.