Research papers Air–sea CO 2 fluxes in the southern Yellow Sea: An examination of the continental shelf pump hypothesis Liang Xue a,b,1 , Longjun Zhang a,n , Wei-Jun Cai b , Li-Qing Jiang c a Key Laboratory of Marine Environmental Science and Ecology, Ministry of Education, Ocean University of China, Qingdao 266100, China b Department of Marine Sciences, The University of Georgia, Athens, GA 30602, USA c Earth Resources Technology, Inc. 6100 Frost Place, Suite A, Laurel, MD 20707, USA article info Article history: Received 17 November 2010 Received in revised form 18 August 2011 Accepted 6 September 2011 Available online 14 September 2011 Keywords: Air–sea CO 2 fluxes Continental shelf pump Carbon export Southern Yellow Sea abstract The southern Yellow Sea (SYS), located to the north of the East China Sea (ECS), was considered part of the ECS when Tsunogai et al. (1999) proposed the ‘‘continental shelf pump’’ (CSP) hypothesis. However, the original CSP carbon dioxide (CO 2 ) uptake flux (2.9 mol C m 2 yr 1 ) appears to have been overestimated, primarily due to the differences between the SYS and the ECS in terms of their CO 2 system. In this paper, we estimated air–sea CO 2 fluxes in the SYS using the surface water partial pressure of CO 2 (pCO 2 ) measured in winter, spring, and summer, as well as that estimated in fall via the relationship of pCO 2 with salinity, temperature, and chlorophyll a. The results indicate that overall, the entire investigated area was a net source of atmospheric CO 2 during summer, winter, and fall, whereas it was a net sink during spring. Spatially, the nearshore area was almost a permanent CO 2 source, while the central SYS shifted from being a CO 2 sink in spring to a source in the other seasons of the year. Overall, the SYS is a net source of atmospheric CO 2 on an annual scale, releasing 7.38 Tg C (1 Tg ¼10 12 g) to the atmosphere annually. Thus, the updated CO 2 uptake flux in the combined SYS and ECS is reduced to 0.86 mol C m 2 yr 1 . If this value is extrapolated globally following Tsunogai et al. (1999), the global continental shelf would be a sink of 0.29 Pg C yr 1 , instead of 1 Pg C yr 1 (1 Pg ¼10 15 g). The SYS as a net annual source of atmospheric CO 2 is in sharp contrast to most mid- and high-latitude continental shelves, which are CO 2 sinks. We argue that unlike the ECS and the North Sea where carbon on the shelf could be exported to the open ocean, the SYS lacks the physical conditions required by the CSP to transport carbon off the shelf effectively. The global validity of the CSP theory is thus questionable. & 2011 Elsevier Ltd. All rights reserved. 1. Introduction The rapid increase of atmospheric carbon dioxide (CO 2 ) is causing global climate change and ocean acidification. Since the Industrial Revolution, atmospheric CO 2 has increased by  38% from 280 to 390 ppm (parts per million in volume) (Le Que ´ re ´ et al., 2009). During this time, the ocean absorbed  48% of the anthropogenic CO 2 from fossil fuel burning and cement production (Sabine et al., 2004). A recent study shows that the CO 2 uptake rate for the global ocean is 1.6 70.9 Pg C yr 1 (1 Pg ¼ 10 15 g) (Takahashi et al., 2009). Thus, the ocean, as a sink of atmospheric CO 2 , plays a key role in alleviating the increase of CO 2 in the atmosphere, despite the weakening in its CO 2 uptake since 1958 suggested by ocean model results (e.g., Sarmiento et al., 2010). Continental shelves are among the most geochemically and biologically active areas of the biosphere, and thus they play a disproportionally important role in ocean carbon cycling (e.g., Gattuso et al., 1998; Muller-Karger et al., 2005; Fennel and Wilkin, 2009), despite their modest surface area (7–8% of the total ocean). It has been reported that shelf seas may be large sinks of atmospheric CO 2 (Tsunogai et al., 1999; Thomas et al., 2004; Borges et al., 2005; Cai et al., 2006; Chen and Borges, 2009; Liu et al., 2010). In one of the earliest of these studies, Tsunogai et al. (1999) coined the term ‘‘continental shelf pump’’ (CSP) as a mechanism describing the uptake of atmospheric CO 2 in the East China Sea (ECS) (note that Tsunogai et al. (1999) considered the southern Yellow Sea (SYS) as part of the ECS in their paper). They estimated that global shelves would absorb 1 Pg C yr 1 of atmo- spheric CO 2 , if the ECS CO 2 uptake rate (2.9 mol C m 2 yr 1 ) could be applied globally. However, more recent studies suggest the shelf seas are smaller CO 2 sinks of 0.22–0.45 Pg C yr 1 (Thomas et al., 2004; Borges et al., 2005; Cai et al., 2006; Chen and Borges, 2009; Liu et al., 2010; Laruelle et al., 2010; Cai, 2011). Cai et al. (2006) further pointed out that although the shelf seas in the mid- and high-latitudes were generally sinks of atmospheric CO 2 , subtropical and tropical shelf seas in the low-latitudes could be the sources. Unfortunately, most shelf seas are greatly Contents lists available at SciVerse ScienceDirect journal homepage: www.elsevier.com/locate/csr Continental Shelf Research 0278-4343/$ - see front matter & 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.csr.2011.09.002 n Corresponding author. Tel./fax: þ86 532 66782967. E-mail addresses: xueliang1001@gmail.com, longjunz@ouc.edu.cn (L. Zhang). 1 Present address: Center for Ocean and Climate Research, First Institute of Oceanography, State Oceanic Administration, Qingdao 266061, China. Continental Shelf Research 31 (2011) 1904–1914