Decoding the drivers of variability in chlorophyll-a concentrations in the Pearl River Estuary: Intra-annual and inter-annual analyses of environmental influences

Environ Res. 2025 Jan 6:268:120783. doi: 10.1016/j.envres.2025.120783. Online ahead of print.

Abstract

Temporal variability and associated driving factors of sea surface chlorophyll-a concentration (Chl-a) in coastal waters have been extensively studied worldwide; however, the importance and spatial heterogeneity of these driving factors remain insufficiently documented. This study addressed this gap by investigating the Pearl River Estuary (PRE) from August 2002 to June 2016, using long-term remote sensing-derived data of Chl-a and potential driving factors, including total suspended solids (TSS), precipitation, photosynthetically active radiation (PAR), and sea surface temperature (SST); and in situ measurements of potential driving factors, including river discharge, wind speed, alongshore wind (u), cross-shore wind (v), and tidal range. A pixel-by-pixel correlation analysis was conducted to preliminarily examine the relationships between these potential driving factors and Chl-a. Subsequently, random forest regression was applied to determine the primary factors in driving the intra-annual and inter-annual variations of Chl-a. Results indicate that, at the intra-annual scale, TSS was the primary factor influencing Chl-a in the shallow (<10 m) nearshore regions and the deep (>30 m) offshore regions, while SST was most important in the intermediate zone. Additionally, in certain areas of the Lingding Bay, Chl-a was primarily affected by river discharge. At the inter-annual scale, tidal range was the primary factor in the western Lingding Bay and the western coastal waters, while Chl-a was primarily modulated by SST in the eastern Lingding Bay. In the offshore region, Chl-a was primarily modulated by river discharge, v, precipitation, and TSS. These findings are of fundamental importance for advancing our understanding of the characteristics and associated mechanisms of marine environmental variability in estuarine systems. They provide a scientific basis for informed management of estuarine environments and sustainable economic development.

Keywords: Chlorophyll-a; Machine learning; Ocean color; Phytoplankton biomass; Remote sensing.