Reply to comment by Jinchun Yuan et al. on “Reduction of primary production and changing of nutrient ratio in the East China Sea: Effect of the Three Gorges Dam?”

Abstract

[1] A brief discussion is presented here in response to Yuan et al.’s [2007] comment on our paper entitled ‘‘Reduction of primary production and changing of nutrient ratio in the East China Sea: Effect of the Three-Gorges Dam?’’ [Gong et al., 2006]. [2] First, we would like to point out that our research provides sufficient seasonal variations using accurate chlorophyll a (Chl a) data in the traditionally nutrient-rich (TNR) region of the East China Sea, as determined in several previous high-frequency sea-based investigations [Gong et al., 1996, 2003, 2006]. In contrast, Yuan et al. [2007] used questionable Chl a data (derived from the SeaWiFS OC4 algorithm: see second point) to study the inter-annual variations in Chl a in the entire East China Sea. This is clearly apparent in Figure 1 which illustrates the comparison between our average sea-based Chl a concentrations in the entire East China Sea (solid brown stars in Figure 1) and the SeaWiFS-derived Chl a data reported in Figure 2 of Yuan et al.’s comment. The results show that the SeaWiFS-derived Chl a data were substantially higher than our sea-based Chl a data. There can be no question that Yuan et al.’s [2007] data were invalid. [3] Second, we further question the accuracy of the SeaWiFS-derived Chl a data for the East China Sea which were used in Yuan et al.’s [2007] comments. Our reasons are as follows: [4] (1) It is widely believed that the standard SeaWiFSOC4 chlorophyll algorithm is only valid for ‘‘Case 1 waters’’ [O’Reilly et al., 1998; Carder et al., 1999]. But this type of global algorithm is flawed when it comes to ‘‘Case 2 waters’’ [Mueller and Austin, 1995; International Ocean-Colour Coordinating Group, 2000; Liew et al., 2001; Gong, 2004]. The key reason for the algorithmic failure in the case of ‘‘Case 2 waters’’ is that there is interference from the strong absorption of colored dissolved organic matter (CDOM) in the blue band that overlaps with phytoplankton chlorophyll absorption. When the absorption coefficient due to colored dissolved organic matter at 380 nm (acdom(l = 380 nm)) is greater than 0.1 m 1 [Mueller and Austin, 1995], the ocean waters are regarded as ‘‘Case 2 waters’’, and the use of the global satellite ocean-color OC4 algorithm is inappropriate. But unfortunately, Yuan et al. [2007] used the OC4 algorithm to process their Chl a data for the East China Sea, a region with ‘‘Case 2 waters.’’ [5] (2) O’Reilly et al. [1998] collected a large data set including coincident in-situ chlorophyll and remote sensing information to evaluate the performance of a wide variety of ocean color chlorophyll algorithms for SeaWiFS data. One of their conclusions is that, while the performance of the OC4 was superior to that of the OC2, it is not as suitable as the initial operational algorithm for SeaWiFS-derived Chl a data because its use requires accurate atmospheric correction and on-orbit calibrations in four bands (instead of two). They explain that this can only be assessed after the collection of sufficient data to validate and fine-tune sensor calibrations. It is currently widely known that atmospheric pollution in the shelf area of the East China Sea is very serious because of rapid industrial development in the coastal cities of China. Additionally, the high frequency of Asian dust storms has been observed in the East China Sea. As a result, the SeaWiFS-derived Chl a data are not likely very accurate without proper atmospheric correction. A second conclusion by O’Reilly et al. [1998] is that ‘‘Case 2 waters’’ present additional complications and challenges as their properties change depending on whether they are dominated by CDOM, nochlorophyllous particles, or a variable mix of both [Carder et al., 1989]. In short, specific algorithms or different parameterizations are required to process data from different regions. [6] (3) Gong [2004] presented CDOM data for four seasons from surface waters of the entire East China Sea. These data showed that the values of acdom(l = 325 nm) were higher than 0.2 m 1 (equivalent to acdom(l = 380 nm) = 0.1 m ) for most of the shelf waters year-round (Figure S1 of the auxiliary material). Figure S1 clearly shows that the values of acdom(l = 380 nm) in the TNR region are always higher than 0.1 m . These results support the optical classification of the TNR region as ‘‘Case 2 waters.’’ Additionally, a significant overestimation of the SeaWiFS-derived Chl a concentrations (2.3 mg m ) at a station located about 200 km east off the Changjiang River mouth (see star-symbol in Figure S1a) was found in