Work

GCOM-C/SGLI ground-area-based chlorophyll content estimation: a computationally efficient algorithm free from LAI and vegetation classification

Year: 2026

Type: article

Abstract: Satellite-derived terrestrial chlorophyll (Chl) content is generally estimated using model inversion or simple relationships with vegetation indices, which require parameterization for each species or plant functional type. Model inversion approaches require essential external inputs, such as leaf area index (LAI), carotenoid content, and land-cover classification. Although satellite-derived LAI products are generated frequently enough to capture both natural and artificial vegetation disturbances, land-cover classification products are produced less frequently, and carotenoid content products are rarely available. Thus, this study aimed to address these limitations by directly estimating Chl content from GCOM-C satellite observations, thereby eliminating the requirement of external inputs. For direct retrieval, reflectance must vary with Chl content up to a sufficiently high level; among the two green bands in GCOM-C/SGLI, Band 6 satisfies this criterion. Therefore, we developed an estimation method using this band to derive a proposed ground-area-based total Chl content (Chl ground ), the vertically integrated Chl content from the ground up through different canopy layers. This offers a computationally efficient, near-real-time detection method without requiring ancillary satellite-derived information. The results indicate that the general, broad-leaf, needle-leaf, and grass models achieved R 2 values of 0.86, 0.93, 0.91, and 0.94, respectively, even under varying soil moisture and the fraction of water-body area within a satellite-vegetated pixel. This independently derived Chl ground metric provides new information and insights into vegetation dynamics. Monthly global Chl ground distribution maps generated in this study reveal that Chl ground differs from LAI and other ecological parameters in both spatial distribution and peak timing, consistent with general ecological knowledge.

Source: Ecological Informatics

DOI: 10.1016/j.ecoinf.2026.104097