Spatio-Temporal Modeling of Macroalgae Biomass and Alginate Content Using High-Resolution Satellite Imagery: A Case Study of Pannikiang Island, South Sulawesi
Aswar Anas (a*), M. Akbar AS (a,d), Agus Aris (a,b,d), Nurjannah Nurdin (a,b,c)

a) Research and Development Center for Marine, Coastal and Small Island, Hasanuddin University, Jl. Perintis Kemerdekaan Km.10, Makassar, 90245, Indonesia
*aswarmarineunhas[at]gmail.com
b) Department of Remote Sensing and Geographic Information System, Vocational Faculty, Hasanuddin University, Jl. Perintis Kemerdekaan Km. 10, Makassar, 90245, Indonesia
c) Marine Science Departement, Marine Science and Fisheries Faculty, Hasanuddin University, Makassar, 90245, Indonesia
d) The Environmental Science Study Program, Doctoral Program, Graduate School, Hasanuddin University, Makassar, 90245. Indonesia

Abstract

Accurate mapping of macroalgae habitats is essential for assessing their ecological functions and bioeconomic potential, particularly for high-value compounds, such as sodium alginate. With the advancement of high-resolution satellite imagery and classification algorithms, remote sensing now offers an efficient and non-destructive approach for the spatiotemporal monitoring of macroalgae. This study aimed to model the spatiotemporal dynamics of dominant macroalgae species (Sargassum and Turbinaria) and estimate their biomass and alginate content on Pannikiang Island, South Sulawesi, using time-series data from PlanetScope imagery within the Google Earth Engine (GEE) platform. The methodology included image preprocessing, extraction of vegetation indices based on green and red-edge bands, classification using the Random Forest algorithm, and regression modeling (both linear and nonlinear) between spectral indices and in situ measurements of biomass and alginate content. The results revealed that seasonal growth patterns of macroalgae can be consistently detected through variations in vegetation indices, with biomass peaks occurring during transitional monsoon periods. The predictive models demonstrated strong correlations between the vegetation indices and estimates of biomass. These findings highlight the potential of high-resolution time-series satellite imagery integrated into GEE for the sustainable monitoring of coastal ecosystems and bioeconomic assessment of macroalgae.

Keywords: Macroalgae- Spatio-Temporal- Biomass- Alginate- High-Resolution

Topic: Topic B: Applications of Remote Sensing