-
The Tasman Sea is a boundary sea off the southeastern coast of Australia, surrounded by Tasmania Island in the west, New Zealand in the east, and the Coral Sea in the north. From an oceanic perspective, the Tasman Sea is affected by a strong western boundary current, the East Australian Current (EAC), which parallels the east coast of Australia and flows southward. Its main mass meanders away from Australia around 32ºS and flows eastward into the Tasman Sea, giving rise to intensified oceanic eddy activity and air-sea exchanges (Sprintall et al., 1995). Meanwhile, the warm water brought poleward by EAC is in strong contrast with the cold water in the south, resulting in an oceanic temperature front called Tasman Front (Andrews et al., 1980; Sloyan and O’Kane, 2015). From an atmospheric perspective, the Tasman Sea is situated in the exit region of the upper atmospheric subtropical westerly jet during austral summer (DJF) (Fig. 1a) and is also closely connected with the climatological storm track of the Southern Hemisphere (SH), although the SH upper westerly jet and the storm track exhibit evident seasonal variations over a year (Fig. 1). In terms of its atmospheric and oceanic features, the Tasman Sea region bears a resemblance to the extension region of the Kuroshio in the Northwest Pacific or to the Gulf Stream in the North Atlantic, where active air-sea interaction can be found and where sea surface temperature (SST), as well as the SST front, may exert a substantial influence on the overlying atmosphere (Nakamura et al., 2004; Kwon et al., 2010). In other words, the SST in the Tasman Sea may provide some potential predictability for the climate system.
Figure 1. The 200-hPa climatological zonal wind (contour, units: m s–1) contoured every 5 m s–1 above 25 m s–1 and the upper-tropospheric storm track (shaded, units: m s–1) based on the standard deviation of the band-pass filtered (2.5–6-day periods) 300-hPa meridional wind for (a) DJF, (b) MAM, (c) JJA, and (d) SON. Shaded areas denote values greater than 7 m s–1. The box in each panel represents the Tasman Sea region (26°−46°S, 150°−174°E).
Previous studies suggest that, on interannual timescales, the Tasman Sea SST is an important factor for local and remote climates. First, it influences the weather and climate of Australia and New Zealand (Hopkins and Holland, 1997; Pook et al., 2006, 2013; Risbey et al., 2009) through impacting the formation and maintenance of the atmospheric blocking high over the Tasman Sea (Simpson and Downey, 1975; Baines, 1983) and the activity of Australian east coastal cyclones (Browning and Goodwin, 2013). Second, it affects the occurrence of extreme marine heatwaves (MHWs), the synoptic-scale anomalous warm water events with durations of five days or longer (Hobday et al., 2016), which may cause devastation of the marine ecosystems and even severe economic tensions (Garrabou et al., 2009; Mills et al., 2013; Wernberg et al., 2013; Oliver et al., 2017). During those years with higher SST, the occurrence of MHWs in the Tasman Sea generally increases (Oliver et al., 2018). Finally, the SST in the Tasman Sea may be a potential precursor for the remote Asian climate by influencing the cross-equatorial atmospheric teleconnection or the lower-tropospheric cross-equatorial flows (Liess et al., 2014; Zhao et al., 2019). A recent study suggests that warming in the Tasman Sea promotes increases in air temperature over the Antarctic Peninsula through a poleward shift of Southern Ocean storm tracks (Sato et al., 2021). Therefore, understanding the variability of SST in the Tasman Sea is of great social and climatic significance.
The processes responsible for the interannual variability of SST in the Tasman Sea are complicated. Frankignoul and Hasselmann (1977) suggested that a large fraction of the SST variability could be explained as a red-noise oceanic response to shorter time scale atmospheric random forcing such as surface heat flux. Consistent with this, Fauchereau et al. (2003) demonstrated that the SST anomalies (SSTAs) in the Tasman Sea seemed to be a response to anomalous latent heat release related to anomalous near-surface winds. More recently, by using the ECCOv4 ocean reanalysis, Bowen et al. (2017) found that the air-sea heat flux contributes to the SSTAs around New Zealand. In addition to these atmospheric flux forcings, the SST is concurrently affected by the heat transport of the ocean current like the EAC, whose strength and extension respond to changes in the South Pacific wind field (Hill et al., 2008; Holbrook et al., 2011; Wu et al., 2012; Chung et al., 2017; Li et al., 2020). Both Ridgway (2007) and Roemmich et al. (2007) suggested that SST warmth in the Tasman Sea was related to the enhanced southward transport of EAC. In agreement with this, Hill et al. (2008) found the variations in temperature at Tasmania can be explained by the heat transport of EAC, and a recent study by Li et al. (2020) showed that a total of 51% of the historical MHWs in the Tasman Sea was primarily due to the increased poleward transports of EAC. Mechanically, the southward extension of EAC, particularly the portion south of 33°S, often causes an unsteady train of mesoscale eddies, increasing eddy mixing (Stammer et al., 2006).
Aside from local and adjacent atmospheric and oceanic factors, SSTAs in the Tasman Sea are also influenced by remote forcings like El Niño–Southern Oscillation (ENSO), the most prominent mode of interannual climate variability (Philander et al., 1984; McPhaden et al., 2006; Deser et al., 2010). ENSO excites a Pacific-South America (PSA) teleconnection to influence the southern extratropics (Hoskins and Karoly, 1981). The surface atmospheric feedback processes linked to ENSO may alter the extratropical SST through a so-called atmospheric bridge (Alexander et al., 2002). Verdy et al. (2006) illustrated that ENSO drives a low-level anomalous circulation pattern over the South Pacific, which causes surface heat flux anomalies and subsequently the SSTAs in the regions of the Antarctic Circumpolar Current. Similarly, Ciasto and England (2011) demonstrated that the ENSO-related atmospheric circulation coincides well with the turbulent heat flux and contributes to the SST variability in the Southern Ocean. Guan et al. (2014) suggested that the ENSO-related PSA pattern generates persistent anomalies in sea level pressure and surface winds around New Zealand and the mid-latitudinal South Pacific. These anomalies cause surface air-sea heat flux anomalies through the evaporation-wind feedback or Ekman drift, and the arched SSTA pattern, which consists of the two major centers of the South Pacific Ocean Dipole (SPOD) pattern during austral summer. In addition, the ENSO-associated oceanic flow and oceanic waves may also affect the extratropical SST through an oceanic tunnel (oceanic bridge) (Liu and Alexander, 2007). Holbrook et al. (2005) suggested that the 3–3.5-yr oscillation of oceanic temperature variability in the upper Southwest Pacific was connected to the EAC and its extension, which might be explained as the forced result of westward propagating oceanic Rossby waves. Also, studies demonstrated the ENSO-related SSTAs in the Southern Hemisphere projects strongly onto the ENSO-related Ekman heat transport due to ocean dynamics (Ciasto and Thompson, 2008; Ciasto and England, 2011). Cetina-Heredia et al. (2014) identified a lagged increase/decrease of southward EAC heat transport approximately 6–9 months after the end of an El Niño/La Niña event. The increased/decreased southward heat transport might alter the SST nearby (Ridgway, 2007; Roemmich et al., 2007; Li et al., 2020). These studies suggest a substantial remote influence of ENSO on southern mid-latitudinal SSTAs, including those in the Tasman Sea, where the strongest SST variability occurs in the peak phase of ENSO, December-January-February (DJF) (Fig. 2).
Figure 2. Spatial distribution of seasonal mean climatological SST (contour, units: °C) and the standard deviation (shaded, units: °C) of detrended SSTAs for (a) DJF, (b) MAM, (c) JJA, and (d) SON. The box in each panel represents the Tasman Sea region as in Fig.1.
However, from several randomly selected cases as displayed in Table 1, the connection of SST in the Tasman Sea to ENSO seems asymmetric with respect to the opposite phases of ENSO (Fig. 3). In La Niña years, the SSTAs generally increase over the whole basin (Figs. 3b, d, f). In contrast, in El Niño years there is no general basin cooling, but a dipolar pattern with warmth in the northwest and cooling in the southeast (Figs. 3a, c, e). In our understanding, such an asymmetric relationship has not been sufficiently investigated. Research on this issue will contribute to a better understanding of the interannual variability of SST in the Tasman Sea, also yielding insight into the non-linear relationship between East Australian rainfall and ENSO (Power et al., 2006; Cai et al., 2010; King et al., 2015; Chung and Power, 2017), since the moisture availability in the southwest Pacific has been proposed as a possible mechanism explaining such a relationship (King et al., 2015).
Years El Niño 1957, 1965, 1968, 1972, 1982, 1986, 1991, 1994, 1997, 2009, 2015 La Niña 1955, 1970, 1973, 1975, 1988, 1998, 1999, 2007, 2010, 2017 Table 1. Years of El Niño and La Niña during the period of 1950–2018 following the threshold of one standard deviation of DJF mean Niño-3.4 index.
Figure 3. The DJF-mean global SST anomaly patterns (units: °C) in (a) 1957, (b) 1970, (c) 1972, (d) 1973, (e) 2009, and (f) 1998. The left and right columns show SSTAs in El Niño years and La Niña years, respectively. The year in the top left of each panel is the year of December. The box in each panel represents the Tasman Sea region as in Fig. 1.
This paper is structured as follows. Section 2 outlines the data and methods adopted in this study. Section 3 examines the connection of SSTAs in the Tasman Sea with those in the tropical central-eastern Pacific and identifies the asymmetric connection with respect to the phase of ENSO. Section 4 analyzes the possible mechanisms for the asymmetry of SST in terms of local air-sea heat fluxes and heat advection by oceanic currents. The roles of the above two factors are further diagnosed by examining a heat budge in the local ocean mixed layer in section 5. Finally, a summary is presented in section 6.
Years | |
El Niño | 1957, 1965, 1968, 1972, 1982, 1986, 1991, 1994, 1997, 2009, 2015 |
La Niña | 1955, 1970, 1973, 1975, 1988, 1998, 1999, 2007, 2010, 2017 |