Antarctic Bottom Water Warming in the Brazil Basin: 1990s Through 2020, From WOCE to Deep Argo

Abstract

A warming trend of 2.1 (±0.4) m°C yr −1 in bottom waters (4,500 to 5,900 dbar) spreading north from Antarctica through the Brazil Basin is quantified by comparing 2019-2020 data from a new Deep Argo regional pilot array to 1989-1995 data from full-depth zonal and meridional hydrographic sections occupied across the basin before and during the World Ocean Circulation Experiment (WOCE). Additionally, float temperatures are about 0.046°C warmer than those from a long-term climatology in those same bottom waters. Lower North Atlantic Deep Water in the basin shows no detectable warming, but Upper North Atlantic Deep Water exhibits evidence of warming in both analyses. The bottom-intensified warming results in a reduction in vertical density stratification between the bottom and deep waters of about 1% per decade. This change is in contrast with the effects of surface-intensified warming, which tends to increase vertical density stratification. Plain Language Summary Surface waters around Antarctica become very cold, salty, and dense through atmosphere and ice exchanges. After cascading down the continental shelf to the abyssal ocean, they spread northward as "Antarctic Bottom Waters." Widespread warming of these waters over the past few decades has been detected by analyses of data from decadal occupations of a set of worldwide, transoceanic research cruises. The better to monitor and understand these and other deep ocean changes, international oceanographers are working together to deploy arrays of oceanic robots (Deep Argo floats) that collect oceanographic data from the sea surface to its floor and report them in real time. We analyze data from one of these arrays, recently deployed in the Brazil Basin. By comparing the Deep Argo data collected from 2019 to 2020 with data collected from 1989-1995 during research cruises in the region, we quantify the warming trend of Antarctic Bottom Water as 0.002°C yr −1 over that time period, which is about one seventh the global sea surface temperature warming trend from 1989-2019. Furthermore, since the bottom waters are warming, but the deep waters above them are not, the density difference between them is lessening over the decades, potentially influencing ocean mixing and circulation.

Warming Trend in Antarctic Bottom Water in the Vema Channel in the South Atlantic

“…The mean potential temperature sampled during 2019–2020 SAMBAR mooring was −0.086°C, an increase of 0.059°C as compared with the −0.145°C mean of the 1991–1992 WOCE data. The warming trend of ∼0.0021°C yr −1 agrees with the most recent estimates in the Argentine Basin (Meinen et al., 2020) and the Brazil Basin (Johnson et al., 2020). Perhaps more crucially, our results suggest that this warming trend is increasing with time, with a possible acceleration beginning around 2005–2007.…”

“…The rate of warming is mostly observed in the upper 2,000 m (Johnson & Lyman, 2020; Wu et al., 2020). However, a significant increase in temperature has also been observed in the deepest layers (>4,000m) (Coles et al., 1996; Johnson et al., 2014, 2019, 2020; Purkey & Johnson, 2010; Purkey et al., 2019). In the Argentine basin, the warming seems to be related to changes in the Weddell Sea (Coles et al., 1996).…”

“…More recently, analysis of the 10‐year time series of temperature from near‐bottom moored instruments in the Argentine Basin indicate intense high‐frequency variability and linear warming trends of ∼0.002–0.004°C yr −1 (Meinen et al., 2020). In the Brazil Basin, a study of the basin‐wide temperature, based on WOCE and Deep Argo observations, estimated a warming of 0.0021 ± 0.0004°C yr −1 in Antarctic Bottom Water (AABW) (Johnson et al., 2020).…”

Persistence of the Large‐Scale Interior Deep Ocean Circulation in Global Repeat Hydrographic Sections

“…For the longitude range 15°–30°W, error in the density gradient is particularly low, notable because it is related to the meridional geostrophic shear just north of the Hunter Channel (Finucane & Hautala, 2022). As in the western South Pacific, persistence does not contradict significant warming observed below 4,000 m for the Brazil Basin (Johnson et al., 2020). The density gradient along A16–A23 in the southern hemisphere is surprisingly stable, despite the reduction in overall density that can also be clearly seen in Figure 3d.…”

Internal Wave Imprints on Temperature Fluctuations as Revealed by Rapid‐Sampling Deep Profiling Floats

“…The original purpose of maintaining over 1,000 Deep Argo floats is to close the heat, freshwater, and sea-level budgets, characterize decadal variability in deep ocean water masses and estimate the mean and decadal variability in deep ocean circulation including the meridional overturning circulations (Johnson et al, 2015). Because of the nearly 250 Deep Argo floats deployed in deep areas of the South Pacific, South Indian, and Atlantic Ocean, we are currently able to estimate the warming rate of deep water in some basins (Johnson et al, 2019(Johnson et al, , 2020 and reexamine the deep currents in some basins (Racapé et al, 2019;Tamsitt et al, 2019;Zilberman et al, 2020). Likewise, it is possible to have assessments on internal (Johnson et al, 2019).…”