Satellite views of the Bohai Sea, Yellow Sea, and East China Sea

Abstract

A comprehensive study of water properties for the Bohai Sea (BS), Yellow Sea (YS), and East China Sea (ECS) has been carried out with 8-year observations between 2002 and 2009 from the Moderate Resolution Imaging Spectroradiometer (MODIS) onboard the Aqua platform. Normalized water-leaving radiance spectra (nL_w(λ)), chlorophyll-a concentration (Chl-a), diffuse attenuation coefficient at the wavelength of 490 nm (K_d(490)), total suspended matter (TSM), and sea surface temperature (SST) are used to quantify and characterize the physical, optical, biological, and biogeochemical properties and their seasonal and interannual variability in the BS, YS, and ECS regions.

The BS, YS, and ECS feature highly turbid waters in the coastal regions and river estuaries with high K_d(490) over ∼3 m⁻¹ and TSM concentrations reach over ∼50 g m⁻³. The optical, biological, and biogeochemical property features in these three seas show considerable seasonal variability. The dominant empirical orthogonal function (EOF) mode for K_d(490) and TSM variability in the BS, YS, and ECS regions is the seasonal mode, which accounts for about two-thirds of the total variance. Phytoplankton dynamics in open oceans of the BS, YS, and ECS is also found to play an important role in the K_d(490) variation, while its impact on the ocean turbidity (K_d(490)) is much less than that of seasonal winds and sea surface thermodynamics in coastal regions. The first EOF mode in SST for the regions is seasonal and accounts for nearly 90% of the total SST variance. The major mechanisms that drive ocean color property variations in the BS, YS, and ECS are the seasonal winds, ocean stratification, and sea surface thermodynamics due to the seasonal climate change, as well as coastal bathymetry, seasonal phytoplankton blooms, and river discharges.

Introduction

The Bohai Sea (BS), Yellow Sea (YS), and East China Sea (ECS) are the three major marginal seas in the western Pacific Ocean bounded by China, Korea, and Japan (Fig. 1). The Bohai Sea, Yellow Sea, and East China Sea cover ∼7.8 × 10⁴, ∼3.8 × 10⁵, and ∼1.25 × 10⁶ km², respectively, for a total coverage of ∼1.71 × 10⁶ km². The mean depths of the BS, YS, and ECS are about 20, 44, and 188 m, respectively. All three of these oceans are characterized with high loadings of sediment concentrations (Guo and Yanagi, 1998, Shi and Wang, 2010b, Wang et al., 2007) in the continental shelf regions. The Yellow River, which is China’s second largest river, transports ∼1.64 × 10⁹ tons of sediment into the Bohai Sea yearly (Milliman and Meade, 1983). The sediment discharge into the ECS from the Yangtze River is approximately 5.0 × 10⁸ tons annually (Milliman and Meade, 1983). Other major rivers, such as the Liaohe, Yalu, and Haihe, also transport significant amounts of sediment into these three seas.

Seasonal monsoons, the Kuroshio Current, and semidiurnal and diurnal tides (Fang, 1994) are primary processes that dominate the ocean hydrography, ocean circulations, sea surface temperature (SST), ocean stratification, and ocean fronts in these three oceans. Due to the sediment deposited in these three oceans via rivers, the BS, YS, and ECS are among the most turbid ocean regions in the world (Shi and Wang, 2010a, Shi and Wang, 2010b). In the BS, huge sediment transportations and depositions from the Liaohe River and the Yellow River form two major delta wetlands: the Liaohe River estuary and the Yellow River estuary. In the YS, sediment depositions from the ancient Yellow River help form the Subei Shoal off China’s Jiangsu province. The outer shelf mud in the south of Cheju Island has been found to contain calcites in its clay fraction derived from the erosion of the old Yellow River submarine delta (Milliman et al., 1985).

Satellite remote sensing has long been used to study physical, optical, biological, and biogeochemical processes in these ocean regions. SST images from the Advanced Very High Resolution Radiometer (AVHRR) were used to characterize sea surface features in the ECS (Tseng et al., 2000). Major ocean fronts in the three oceans have also been distinguished with satellite SST observations (Hickox et al., 2000). With satellite remote sensing, clear ocean-to-atmospheric feedback in the Yellow and East China Seas, triggered by the submerged ocean bottom topography (Xie et al., 2002), is also revealed. In recent years, satellite ocean color remote sensing has been increasingly used to study ocean optical, biological, and biogeochemical processes as well as to monitor the natural hazards in these regions. Monthly variation of pigment concentrations has been studied with satellite Coastal Zone Color Scanner (CZCS) observations (Tang et al., 1998). Images of chlorophyll-a concentration (Chl-a) of the Moderate Resolution Imaging Spectroradiometer (MODIS) show significant cross-shelf penetrating fronts off the southeast coast of China (Yuan et al., 2005). The occurrences of green algae blooms in the YS are also investigated with satellite ocean color observations (Hu et al., 2010, Shi and Wang, 2009b). Most recently, ocean sand ridge signatures in the BS (Shi et al., 2011a), sea ice effects in the BS (Shi and Wang, 2012a, Shi and Wang, 2012b), ocean optical and biological properties in the Korean dump site of the YS (Son et al., 2011), and the seasonal sediment plume in central ECS (Shi and Wang, 2010b), as well as the spring-neap tidal effects on satellite ocean color remote sensing in the BS, YS, and ECS (Shi et al., 2011b) have been studied using MODIS-Aqua measurements.

For the BS, YS, and ECS regions, significant ocean radiance contributions at the near-infrared (NIR) wavelengths can be found along the coast of the YS and ECS (Shi and Wang, 2009a, Wang et al., 2007). For example, the normalized water-leaving radiance (nL_w(λ)) at the NIR band can reach ∼2–3 mW cm⁻² μm⁻¹ sr⁻¹ in the Hangzhou Bay (Wang et al., 2007). The bio-optical complexities in the BS, YS, and ECS regions suggest that the standard (NIR) atmospheric correction algorithm (Gordon and Wang, 1994, IOCCG, 2010) for MODIS ocean color data processing is unable to produce valid ocean color products in order to quantify ocean optical, biological, and biogeochemical properties in these three seas. Indeed, limited ocean color products from the NASA standard (NIR) data processing, such as nL_w(λ) spectra (Gordon, 2005, Gordon and Wang, 1994, Morel and Gentili, 1991, Wang, 2006b), chlorophyll-a concentration (O’Reilly et al., 1998), diffuse attention coefficient at the wavelength of 490 nm (K_d(490)) (Lee et al., 2005, Morel et al., 2007, Mueller, 2000, Wang et al., 2009a), and total suspended matter (TSM) concentration (Miller and McKee, 2004, Tassan, 1993, Zhang et al., 2010), can be used to quantitatively evaluate and characterize the short-term and long-term variability in these three seas, and consequently study the ocean and atmospheric processes contributing to these changes. Recently, a shortwave infrared (SWIR) atmospheric correction algorithm (Wang, 2007, Wang and Shi, 2005) for satellite ocean color data processing has been proposed and demonstrated to significantly improve the ocean color products in the coastal and inland turbid waters (Wang et al., 2007, Wang et al., 2009b, Wang et al., 2011, Zhang et al., 2010). In addition, a tuned (NIR-corrected) atmospheric correction algorithm using NIR bands for MODIS-Aqua ocean color data processing is also proposed for these regions (Wang et al., 2012). Indeed, Shi and Wang (2009a) showed that the black pixel assumption, on which the atmospheric correction for satellite ocean color data processing is based, is generally valid for the SWIR bands for the turbid coastal regions such as in the BS, YS, and ECS regions.

In this study, the combined NIR-SWIR atmospheric correction algorithm (Wang, 2007, Wang and Shi, 2005, Wang and Shi, 2007, Wang et al., 2009b) is used to derive ocean optical property data for both the open ocean and coastal region waters. Seasonal and interannual changes of physical, optical, biological, and biogeochemical properties in these three seas are characterized and quantified with SST (Minnett et al., 2004), nL_w(λ) (Gordon and Wang, 1994, IOCCG, 2010), Chl-a (O’Reilly et al., 1998), K_d(490) (Wang et al., 2009a), and TSM concentration (Zhang et al., 2010) from the measurements of MODIS-Aqua. Annual and seasonal climatologies of nL_w(λ), Chl-a, K_d(490), TSM, and SST in the BS, YS, and ECS regions are derived and quantified. The evaluations of seasonal and interannual variations of all these physical, optical, biological, and biogeochemical parameters are also provided and correlated to the ocean and atmospheric processes.