A hypothesis on variability of surfacewater p CO 2 under the rapid sea-ice retreatduring summer in the Arctic Ocean

The Arctic Ocean has been recognized to play an active role in global carbon cycle due to its potential to disproportionally take up CO 2 from the atmosphere especially under a dwindling ice extent. It is estimated that the Arctic Ocean will be responsible for up to 5%–14% of the total global oceanic CO 2 uptake even though it comprises only 3% of the world ocean surface area. Over the past decade, the sea ice in the Arctic Ocean has significantly retreated during summer. The rapid sea ice retreat is thought to cause corresponding changes in the distribution of p CO 2 in the surface waters of Arctic Ocean. However, the response of carbon dynamics to a changing Arctic Ocean remains poorly quantified. It has been postulated that an ice-free condition in the Arctic Ocean basins would allow for uptake of a substantial amount of carbon dioxide (CO 2 ) from the atmosphere as a result of sea ice melt and increasing primary productivity. However, the other direct observation-based study of high-resolution survey of sea-surface CO 2 concentration across the Canada Basin, showing a great high p CO 2 relative to earlier observations and predicted that the Arctic Ocean basin will not become a large atmospheric CO 2 sink under ice-free conditions. Due to the existing quite different views, thus, there is an implication of observation-based learning in the variations of the surface seawater p CO 2 under a scenario of rapid sea ice melting. In the summer of 2008, during the 3rd Chinese National Arctic Research Expedition (CHINARE) cruise, significant variability of p CO 2 in surface water has been observed along a trans-Arctic section of 150°–160°W from the Chukchi Sea slope to 88°N in the western Arctic Ocean. We propose a hypothesis of “low-low-high” variation trend in surface water p CO 2 levels along a decreasing ice-cover gradient across three different zones in the Arctic Ocean under a rapid sea ice melting scenario in the summer. Specifically, there are (1) “low” p CO 2 levels of ~270–280 μ atm in the sea ice-covered zone (north of 84°N) with a high ice concentration >70%; (2) “low” p CO 2 of ~250–270 μ atm in a sea ice melting zone (78°–81°N) with a medium ice extent of around 50%; and (3) “high” p CO 2 of ~320–365 μ atm in an ice-free zone (72°–77°N) with a low ice concentration of p CO 2 variation trend could be attributed to different driving forces. The low p CO 2 in the heavily ice-covered northern basin is likely influenced by a combination of several processes, including mixing of various source waters, CO 2 fixation by ice algae, ice-water gas exchange, and temperature change, although their relative roles remain to be quantified. The low p CO 2 observed in the partially ice-covered northern Canada Basin could be primarily derived from biological CO 2 fixation and the dissolution of CaCO 3 precipitates, both of which consume CO 2 . Primary productivity enhanced by the early melting of sea ice plays an important role in the decreased sea-surface p CO 2 . In addition, sea ice melting also promotes the dissolution of authigenic CaCO 3 in the form of ikaite within the sea ice. Finally, results of tracer data and model simulations suggest that the rapid CO 2 invasion from the atmosphere and low biological export of CO 2 owing to a shallow mixed-layer depth, strong subsurface water stratification, and limited nutrient supply to the surface water are both responsible for the observed high p CO 2 values in the ice-free southern Canada Basin.