1. Introduction
Strong extratropical sea surface temperature (SST) fronts form along the boundary between warm and cold ocean currents, such as between the Kuroshio Extension and the Oyashio (e.g., Kida et al. 2015) or between the Brazil and Malvinas Currents. Many recent studies have focused on the air–sea interactions around midlatitude SST fronts, because these fronts are thought to have as great an influence on the overlying atmosphere as SST fronts in tropical oceans (e.g., Lindzen and Nigam 1987; Wallace et al. 1989; Hayes et al. 1989; see the next paragraph). Direct atmospheric soundings by a single vessel were carried out over the Kuroshio Extension by Tokinaga et al. (2006) and Tanimoto et al. (2009), and around the Brazil Current by Pezzi et al. (2005). However, it takes roughly a full day for a research vessel to carry out a set of observations across a SST front. Therefore, when single-vessel radiosonde observations document a change in the vertical profile of wind across a SST front, it may reflect a mixture of temporal and spatial changes. This study gathered data on anomalous winds ascribed to an underlying SST front during a field campaign in which radiosonde observations were made simultaneously from three ships. These observations provide direct evidence of the relationship between the SST front and the anomalous winds.
The “pressure adjustment mechanism” of Lindzen and Nigam (1987) and the “vertical mixing mechanism” of Wallace et al. (1989) and Hayes et al. (1989) both apply to SST fronts in extratropical as well as tropical environments. In the former mechanism, cold and warm SST anomalies locally generate positive and negative anomalies, respectively, in sea level pressure (SLP), and surface winds can thereby be modulated by the locally generated SLP gradient across the SST gradient. Minobe et al. (2008) reported that the pressure adjustment mechanism enhances convective rainfall through a low pressure anomaly generated on the warm side of the Gulf Stream. By analyzing in situ observations, Tanimoto et al. (2011) identified a climatological SLP minimum situated locally along the Kuroshio Extension, which is consistent with the pressure adjustment mechanism. In the vertical mixing mechanism, a warm SST anomaly enhances vertical mixing of wind momentum within the marine atmospheric boundary layer (MABL). Compared to cold SST, vertical wind shear over warm SST is thus weaker and surface winds are stronger, acting to increase the heat release from the ocean. Nonaka and Xie (2003) showed from satellite observations that the vertical mixing mechanism is operative along the Kuroshio Extension, where faster and slower wind speeds are associated with warm and cold SST anomalies, respectively. Likewise, Tokinaga et al. (2005) analyzed satellite and in situ data around the Malvinas–Brazil Current convergence and showed that vertical wind shear was weak in warm SST areas and strong in cold ones, again concordant with the vertical mixing mechanism. Xu and Xu (2015) examined the atmospheric response to the Kuroshio SST front in the East China Sea. Through their analysis of satellite data and numerical modeling, they revealed that the pressure adjustment mixing contributes more to the atmospheric response under alongfront prevailing winds, whereas the vertical mixing mechanism dominates the atmospheric adjustment under cross-front winds.
Tanimoto et al. (2009) analyzed atmospheric sounding data across the Kuroshio Extension front in the baiu season. They found that when the baiu front was displaced northward of the Kuroshio Extension front, southwesterly winds enhanced the near-surface stratification over the cool water, whereas when the baiu front was displaced southward of the Kuroshio Extension front, the stratification was reduced by northerly winds over the warm water. Furthermore, direct atmospheric soundings have shown that the vertical mixing mechanism acts to weaken vertical wind shear over the warm Kuroshio Extension (Tokinaga et al. 2006) and Brazil Current (Pezzi et al. 2005). However, it takes roughly a full day for a research vessel to conduct a set of observations across a SST front. Therefore, a change in the vertical profile of wind speed across a SST front, if documented by single-vessel radiosonde observations, may reflect a mixture of temporal and spatial changes (e.g., the mixture of a diurnal cycle and a passage of synoptic-scale disturbance). As a result of this problem, unambiguous spatial changes in the vertical structures of the MABL across the SST front produced by the pressure adjustment mechanism and the vertical mixing mechanism have not yet been captured by direct atmospheric soundings.
Sounding data obtained by a single vessel can include temporal changes in the atmosphere, which may mask more persistent signatures forced by the SST distribution. A practical attempt has been made to deal with this problem by eliminating linear observational trends (e.g., Kasamo et al. 2014) under the assumption that temporal changes due to the passage of synoptic-scale weather systems are more or less linear, but this assumption is not always valid. Kawai et al. (2014) attempted to estimate temporal changes in SLP associated with weather systems by utilizing buoy measurements. However, buoys are not always located near atmospheric sounding points, and they do not measure vertical atmospheric profiles, either.
For 5 days in early July 2012, as part of an intensive observation campaign of the Japanese “Hotspot Project” on extratropical air–sea interaction,1 simultaneous GPS radiosonde launches were carried out repeatedly from three vessels that were aligned meridionally across the SST front along the Kuroshio Extension, as described in detail by Kawai et al. (2015). Simultaneous radiosonde launches by multiple vessels aligned across a SST front would offer a unique opportunity to separate cross-frontal variations in atmospheric conditions forced by the SST gradients from temporal changes associated with atmospheric disturbances, although the arrangement of multiple vessels is likely to be difficult. This unprecedented set of observations provides detailed atmospheric fields in a meridional cross section across the SST front for 2–6 July 2012. Kawai et al. (2015) reported that the observations captured abrupt changes of the mesoscale MABL structure across the SST front. For example, the cloud-base height increased sharply across the Kuroshio Extension front from the cooler to the warmer side; this increase can be explained by enhanced turbulent mixing in the MABL over the higher SST under the influence of the northerly winds.
This study analyzed observational data obtained by the aforementioned simultaneous radiosonde measurements with a particular focus on anomalous winds. The main purpose of this paper is to present observational evidence for anomalous winds generated by the sharp SST gradient across the Kuroshio Extension. Section 3 outlines the contrasting oceanic effects on overlying air temperature and pressure distributions produced by the cold and warm SSTs across the front. We then demonstrate the following two processes: 1) anomalous winds forced by the local pressure gradient, which is consistent with the pressure adjustment mechanism; and 2) enhanced vertical mixing in cool air moving over warm waters, which is consistent with the vertical mixing mechanism. Section 4 provides a summary and the conclusions.
2. Observations
The three-ship observations were carried out from 0600 Japan standard time (JST = UTC + 9 h) 2 July to 0000 JST 7 July 2012 (Kawai et al. 2015). The three ships—R/V Tansei-maru of the Japan Agency for Marine-Earth Science and Technology (JAMSTEC), R/V Wakataka-maru of Tohoku National Fisheries Research Institute in Japan, and R/V Seisui-maru of Mie University—were kept aligned meridionally along 143°E, across the SST front along the Kuroshio Extension (Fig. 1). The vessels shifted northward or southward in a synchronous manner, each within a different half-degree section along 143°E, and switched directions every 4 h. GPS radiosondes were launched simultaneously from each ship every hour or every 2 h in such a way that observation points were either about 14 km (0.125°) or about 28 km (0.25°) apart (Fig. 2). This approach meant that up to 15 atmospheric vertical profiles could be acquired across the SST front over a period of only 4 h if radiosondes were launched every hour, and we succeeded in capturing detailed atmospheric fields on a meridional section across the SST front over a 5-day period. GPS radiosondes sample the data every second, equivalently with 3-m vertical spacing between data points for an ascending rate of 3 m s−1. We interpolated the observation data to produce a three-dimensional [latitude–height–time: 0.125° (~14 km) × 10 m × 1 h] gridded dataset.
SST distribution based on the Japan Coastal Ocean Predictability Experiment 2 (JCOPE2; Miyazawa et al. 2009) data from an ocean forecast system used to capture synoptic SST conditions on 2 Jul 2012 (colored and contoured, °C). The vertical black bar represents the observation line along which three research vessels moved back and forth.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
Latitude–time section of SST measured by in situ observation (colors and contours, °C). Time is represented by four digits, the first two being the day of July 2012 and the last two being the hour in JST. Black dots correspond to the atmospheric sounding points.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
For these observations, RS-06G GPS radiosondes manufactured by Meisei Electric Co., Ltd., were used, and each ship was equipped with three radiosonde receivers of the same type. It takes about 2 h to obtain an atmospheric profile, but with three receivers, each ship could carry out observations as frequently as every hour. The sensors have the nominal accuracy of ±0.5°C in temperature, ±7.0% in relative humidity, ±1.0 hPa in pressure, and ±2 m s−1 in wind. Temperature and relative humidity of each radiosonde sensor were calibrated about 30 min before the launch. The in situ observations are input as surface data in the GPS radiosonde observation system. The height of the in situ observations are 12, 11.7, and 9.4 m above the sea level in R/V Tansei-maru, R/V Wakataka-maru, and R/V Seisui-maru, respectively.
During the latter part of the observation period, from 1200 JST 3 July to 0000 JST 7 July, the SST front locally moved northward rapidly. As a result of this frontal displacement, fewer radiosonde launches were conducted on the cooler side of the SST front during the latter period than during the earlier period (Fig. 2), even though the northernmost positions of the vessels were shifted slightly northward in the later period. In addition, the cross-frontal contrast in air temperature in the lower troposphere was reduced during periods of southerly winds. Thus, the observational data are averaged separately for the periods of northerly (70 samples) and southerly (50 samples) winds. The 5-day averages were taken without removing the diurnal cycle beforehand. This is because the diurnal cycle of the meridional SST contrast was found to be small. We thus focus on differences in the mean structure of MABL under the influence of the SST front between the northerly and southerly wind periods, while analysis of the diurnal cycle is left for a future study.
3. Observational results
In this section, we present the data obtained from the three-ship radiosonde campaign across the Kuroshio Extension front as latitude–time cross sections. Latitude in the sections is set relative to the SST front, which moved northward during the observational period. The SST front is defined instantaneously as the latitude of the strongest meridional SST gradient based on an in situ measurement at each observation time.
a. Oceanic effect on air temperature
The total turbulent heat flux was calculated with the Coupled Ocean–Atmosphere Response Experiment (COARE 3.0 algorithm; Fairall et al. 2003) as the sum of latent and sensible heat fluxes (Fig. 3a). During the observation period, the upward turbulent heat flux from the ocean was systematically greater to the south of the SST front than to its north, and the flux appeared to reach a maximum 40–50 km or more south of the front. The heat release from the Kuroshio Extension in the period of the northerly winds was stronger than in the period of the southerly winds (not shown). The cross-frontal contrast in the turbulent heat flux was consistent with the corresponding contrast in the air–sea temperature difference (Fig. 3b). Surface air temperature (SAT) measured by in situ observation at the height of about 10 m above the sea level for a particular vessel also displayed a cross-frontal gradient (Fig. 4a). This SAT gradient was weaker than the frontal SST gradient; therefore, a cross-frontal gradient of upward turbulent heat flux was produced (Fig. 3a). It is noteworthy that the simultaneous three-ship observations were successful in capturing the southward advection of cold air across the SST front by the strong northerlies and the northward advection of warm air across the front by the strong southerlies (Kawai et al. 2015).
Latitude–time sections of (a) upward surface turbulent heat flux (W m−2) and (b) air–sea temperature difference (SAT minus SST, °C). In (a), red represents the heating of the atmosphere by the ocean, while blue represents cooling. In (b), red represents unstable stratification (SAT < SST), while blue represents stable stratification (SAT > SST). The vertical axis shows the instantaneous distance in kilometers and degrees latitude northward (N) or southward (S) of the SST front (FRNT). N01, S01, and so on, indicates the latitude distance north and south of the front (FRNT) in increments of 0.125°.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
Latitude–time sections of air temperature (contours, °C), its instantaneous deviations from the average over the entire meridional line at each observation time (colors, °C), and meridional wind velocity (vectors, m s−1) at (a) the surface, (b) 100-m altitude, and (c) 1000-m altitude. Other information is as in Fig. 3.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
Though weaker than at the surface, corresponding meridional temperature gradients were also observed at 100-m altitude, especially during the periods of northerly winds (Fig. 4b). The thermal gradient, however, diminished at 1000-m altitude (Fig. 4c). In the meridional section of potential temperature averaged for the northerly period (Fig. 5a), the strong influence of the cross-frontal SST contrast, manifested as warm and cool anomalies on the warm and cool sides of the SST front, is evident up to the altitude of approximately 300 m. To the south of the front, the stratification in the near-surface layer below the altitude of 50 m was unstable with superadiabatic lapse rates (see contour and hatching in Fig. 5a), which suggests that the cool northerly airflow was rapidly heated by the warm Kuroshio Extension. The corresponding meridional section for the southerly period (Fig. 5b) showed that the near-surface stratification was stronger than in the northerly period, especially to the north of the Kuroshio Extension front. As in the northerly period, the anomalous thermal structure as a response to the cross-frontal SST contrast was in the form of a meridional dipole, but the response was confined to the ~200-m altitude, owing to reduced turbulent mixing under the stronger stratification due to the warm southerlies. As a result of the warmness of the Kuroshio Extension, even the warm southerlies accompanied unstable near-surface stratification with superadiabatic lapse rates just to the south of the front. Though weaker than near the surface, the thermal contrast under the influence of the cross-frontal SST contrast appeared to extend up to 1500-m altitude regardless of the wind direction (Figs. 5a,b). In summary, the particular meridional and vertical distributions of the lower-tropospheric temperature were determined mainly by the heating/cooling from the ocean across the SST front, which was modulated by the horizontal thermal advection by the cross-frontal winds.
Meridional sections of time-mean potential temperature (contours, K) and its instantaneous deviation from its horizontal average at each level (colors, K), both averaged separately for the period of the surface (a) northerlies and (b) southerlies. The hatching near the surface is where a superadiabatic lapse rate was observed.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
The time–latitude sections (Fig. 4) show that the cross-frontal winds changed direction several times during the observational period, owing to the passage of synoptic-scale atmospheric disturbances. Although the cross-frontal wind might contribute to the displacement of the SST front (Figs. 2 and 4a), the displacement was slight and the north–south frontal SST contrast was retained. The clear north–south contrast in near-surface air temperature thus persisted throughout the observation period. In other words, the oceanic influence on air temperature was a feature observed throughout this time period, not only under calm conditions but also under disturbed conditions due to passing weather systems. Although previous studies based on single-vessel observations have documented persistent influence of SST fronts on the atmospheric thermal structure (e.g., Tokinaga et al. 2006; Tanimoto et al. 2009), their data are likely to represent a mixture of temporal and spatial changes. Our simultaneous three-vessel observations, by contrast, provide unambiguous evidence that the SST front exerts a persistent influence on meridional contrasts in air–sea heat exchanges and the lower-tropospheric temperature.
b. Oceanic effect on air pressure
To extract the persistent atmospheric response to the underlying SST gradient, fluctuations associated with synoptic-scale atmospheric disturbances were removed from the observed data. The strong thermal impacts of the SST front were confined to the lowest 300 m in the atmosphere, and fluctuations in the free atmosphere were mostly due to synoptic-scale disturbances (Figs. 4 and 5). Following Nishikawa et al. (2014), we subtracted air pressure measured at 1500-m altitude during a given sounding from air pressures measured below 1500 m during the same sounding.
The pressure anomalies plotted in Fig. 6 are the deviations at each individual level from the mean pressure along the entire meridional line across the SST front and from the pressure anomalies at the 1500-m level during each sounding. Throughout the observational period (Fig. 6a), SLP anomalies were positive and negative on the cooler and warmer sides, respectively, of the SST front. Similar pressure anomalies were also apparent, though somewhat weaker, at 100-m altitude (Fig. 6b). In time-averaged meridional sections of the pressure anomalies (Figs. 6c,d), a northward pressure gradient corresponding to the meridional contrast in near-surface temperature (Fig. 5) was evident only below the altitude of approximately 800 m during both the northerly and southerly periods, but it is a topic for a future study why an abrupt vertical drop-off in the anomalies was identified. These results nevertheless demonstrate that a clear meridional pressure gradient associated with the SST-driven air temperature contrast extended to higher altitudes than the temperature contrast itself. A comparison between Figs. 6c and 6d reveals that the negative pressure anomalies spread farther south in the northerly period than in the southerly period, which is consistent with the advective shift of the temperature anomalies (Fig. 5), where warm anomalies shifted farther south in the northerly period. Specifically, the negative pressure anomalies in the southerly period were centered at around S01 and narrower than in the northerly period. The narrow negative pressure anomalies in the southerly period almost coincided with the near-surface warm anomalies around S01–S02 (Fig. 5b). The near-surface pressure contrast was stronger during the period of northerly winds (Fig. 6), which is again consistent with the greater north–south temperature contrast under the northerlies (Fig. 5). This SST-driven north–south pressure contrast persisted even during the passage of synoptic-scale disturbances, as was the case with the temperature contrast (Fig. 4).
Latitude–time sections of pressure anomaly (colors and contours, hPa) at (a) the surface and (b) 100-m altitude, defined as local deviations at each level from the pressure at 1500-m altitude in each observation after removing the mean pressure over the whole profile at each time. Other information is as in Fig. 3. Meridional sections of the mean pressure anomaly (colored and contoured, hPa) averaged separately for the (c) northerly and (d) southerly periods. Other information is as in Fig. 5.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
c. Oceanic effect on wind
1) Zonal wind
The anomalous meridional pressure gradient (Fig. 6) observed as a response to the SST gradient should accompany wind anomalies, including the zonal geostrophic component. In this section, we compare the observed zonal wind with the zonal component of the geostrophic wind. As was the case with the pressure data, we subtracted the 1500-m zonal wind from the corresponding observations at individual levels to isolate lower-tropospheric phenomena from fluctuations due to synoptic-scale disturbances. As evident in Fig. 7a, the deviations of the surface zonal wind from the 1500-m wind tended to be easterly almost throughout the observation period. On the meridional section of the time-mean zonal wind for the northerly period (Fig. 7b), the zonal wind actually observed was westerly above the 1000-m level, whereas it was basically easterly below with the largest intensity near the surface (about 3 m s−1). This downward strengthening of the easterly winds cannot be explained either by surface friction or by a vertical mixing effect, which should translate westerly momentum downward. Although surface friction leads to turning of the wind toward the low pressure side, the observations show that the northwesterlies at the 1500-m level turned into northeasterlies near the surface, in the direction opposite to the Ekman spiral [see section 3c(3)]. The strongest easterly winds were observed near the surface slightly to the north of the SST front, collocated with the node of the dipolar pressure anomalies forced by the frontal SST gradient (Figs. 6c and 7b). Consistently, the vertical shear of the zonal wind was weaker south of the SST front than to its north. The layer of the easterlies was shallower in the southerly period (Fig. 7c) than in the northerly period (Fig. 7b).
(a) Latitude–time section of observed surface zonal wind velocity (vectors, m s−1) and its local departures from the corresponding value at 1500-m altitude (colors, m s−1). Leftward and rightward arrows represent the easterlies and westerlies, respectively, and the anomalous easterlies and westerlies are colored in green and yellow, respectively. Other information is as in Figs. 3 and 5. (b),(c) As in (a), but for meridional sections based on zonal wind velocities (m s−1) actually observed at individual altitudes (vectors; zero lines in red) and their local deviations as local departures from the corresponding value at 1500-m altitude (colors), both averaged separately for the (b) northerly and (c) southerly periods.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
The zonal component of the geostrophic wind anomaly was estimated from the deviation of the observed pressure at each level from that at the altitude of 1500 m (Fig. 8), so as to extract the geostrophic component associated with the low-level north–south pressure contrast formed locally in the presence of the SST front. The geostrophic zonal wind anomalies (Fig. 8a) estimated from the observed SLP anomalies (Fig. 6a) were predominantly easterly near the SST front, which is similar to the observed zonal wind anomalies (Fig. 7a), but with a slight overestimation. The easterly geostrophic wind anomaly averaged for the northerly period reached a maximum near the surface north of the SST front (Fig. 8b), which is in agreement with the observed zonal wind deviation (Fig. 7b). In the southerly period, the easterly geostrophic wind anomalies were weaker on average (Fig. 8c). Unlike in the observations, however, the estimated geostrophic wind anomaly was westerly far north of the SST front. This discrepancy between the observed and geostrophic wind anomalies may be due, at least partly, to the limited number of observations available, which is, for example, only eight north of N04 (as indicated by the dots in Fig. 2). The meridional distribution of the geostrophic easterly anomalies differs from that of the observed easterly anomalies, despite the fact that the southward pressure gradient force is correlated with the deviation of the observed zonal wind anomalies at the altitude of 100 m (correlation coefficient = 0.26; statistically significant at the 99% confidence level with 197 degrees of freedom assuming that each observational profile is independent). The correlation was confirmed to be significant up to nearly 800 m above the surface (not shown). It is, thus, possible to explain the observed easterly wind partly with the near-surface meridional pressure gradient formed by the SST front, especially under the northerly winds. This is in agreement with Xu and Xu (2015), in which the pressure adjustment mechanism contributes more to the atmospheric response under the prevailing alongfront winds. This result suggests the importance of the pressure adjustment mechanism (Lindzen and Nigam 1987) near the Kuroshio Extension front.
(a) As in Fig. 7a, but for the zonal component of surface geostrophic wind anomalies, as calculated from the observed sea level pressure anomalies plotted in Fig. 6. (b),(c) As in Figs. 7b and 7c, respectively, but for the zonal geostrophic wind anomalies.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
2) Meridional circulation
To examine the meridional circulation, meridional sections were constructed of the observed meridional wind averaged over the entire observational period and its local deviations relative to the 1500-m altitude at individual latitudes (Fig. 9a). The meridional wind anomalies thus defined were southerly below the 500-m altitude (colored in Fig. 9a), and actual meridional winds were also southerly between the 100-m and 300-m levels on the cooler side of the SST front (contoured in Fig. 9a). To investigate the influence of the SST front on the southerly from 100 to 300 m, we focused on the lowest 300 m, by setting the 300-m level as a new reference level for the calculation of local deviations of the meridional wind velocity (Fig. 9b). Below that altitude, meridional contrasts across the SST front were evident in both potential temperature (Fig. 5) and pressure (Figs. 6c,d) anomalies. As shown in Fig. 9b, the anomalous northerlies prevailed near the surface, especially to the north of the SST front, whereas the anomalous southerlies were prevalent above the 100-m level, especially to the south of the SST front. This near-surface northerly reinforcement is consistent with a frictional effect in the presence of the anomalous northward pressure gradient as a thermal response to the frontal SST gradient (Figs. 6c,d), and it might be possible to interpret the southerly reinforcement aloft as the return flow. This pattern of the anomalous meridional wind, which seems somewhat analogous to sea breeze between a cold ocean and warmer land, can be regarded as additional observational evidence that the SST front can force local circulation from the surface to about 300-m altitude. This pattern of the anomalous meridional wind was observed both in the northerly (Fig. 9c) and southerly (Fig. 9d) periods, except in the lowest 30 m above the surface, although the meridional wind anomalies in the southerly wind period were much stronger than in the northerly period.
(a) Meridional section of meridional wind velocity (contoured for every 0.2 m s−1) and its deviation from its value at 1500-m altitude (colors, m s−1) both averaged over the entire observational period. The southerlies and northerlies are plotted with solid and dashed lines, respectively, and the anomalous southerlies and northerlies are colored in red and blue, respectively. (b) As in (a), but the reference level for calculating deviations is set to 300 m. (c) As in (b), but averaged only for the northerly period. (d) As in (b), but averaged only for the southerly period.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
Though not necessarily apparent in every observational time, this sea-breeze-like circulation was unambiguously observed in the time-mean fields in Fig. 9, which is likely a weak secondary circulation. Unlike typical sea-breeze circulation, however, the secondary circulation lacks its dominant diurnal cycle. Rather, the Coriolis force acting on the surface northerlies associated with this secondary circulation was necessary for maintaining the surface easterlies against the friction. Furthermore, the secondary circulation acted to maintain the upward-decaying geostrophic easterlies in thermal wind balance with persistent cross-frontal thermal contrast. The shallowness of the thermal contrast as shown in Fig. 5 may be one of the reasons why the secondary circulation was only 300 m in depth (Fig. 9) in spite of the dipolar pressure anomalies reaching 800 m above the surface (Fig. 6). Another possible reason may be the effect of a thin layer with enhanced stratification around 400 m above the surface (Fig. 5), which could prevent the vertical motion associated with the secondary circulation from extending above the 300-m level.
3) Reversed boundary layer spiral
We obtained vertical wind profiles from the surface to the 1500-m (Figs. 10a,b) level based on the atmospheric soundings averaged separately for the northerly and southerly periods. The wind profile averaged for the southerly period (Fig. 10b) resembled a classic Ekman spiral, where the wind vectors turned counterclockwise from southwesterly to southeasterly from 1500-m altitude down to the surface (i.e., backing downward). In contrast, the wind profile averaged for the northerly period (Fig. 10a) showed wind vectors turning clockwise from northwesterly to northeasterly (i.e., veering downward), in a direction opposite to that of a classical Ekman spiral. The wind shear observed within the lowest 1500-m layer during the northerly period thus seemed more consistent with thermal wind associated with a frontal SST gradient, as discussed above. These contrasting wind spirals were observed not only over the SST front but also over its northern and southern flanks (not shown).
Vertical wind profiles over the SST front averaged separately for the (a) northerly and (b) southerly periods, where wind vectors are plotted at 250-m intervals in oblique view and their lengths are proportional to wind speeds. Blue, green, and red dots denote projections of the corresponding wind profiles at 10-m intervals from the surface to 500 m, from 510 m to 1000 m, and from 1010 m to 1500 m, respectively. (c),(d) As in (a) and (b), respectively, but the time-mean wind vectors at individual altitudes are plotted on horizontal planes.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
It is noteworthy that the vertical shear of the wind vectors within 250 m above the surface averaged for the northerly period was smaller (Figs. 10a and 10c) than for the southerly period (Figs. 10b and 10d). This difference is more evident to the south of the SST front (not shown). Under cold advection by the surface northerly winds, the near-surface stratification was reduced owing to increased sensible heat release from the ocean and thus this enhanced the vertical mixing mechanism (Wallace et al. 1989; Hayes et al. 1989). Under warm advection by southerly winds, by contrast, the vertical momentum mixing was likely suppressed under the augmented near-surface stratification, leading to large vertical wind shear (Figs. 10b and 10d).
4. Summary and conclusions
Simultaneous radiosonde observations by three research vessels (Kawai et al. 2015) were successful in capturing the influences of a strong SST front along the Kuroshio Extension upon the overlying atmosphere. The observational evidence of the SST influences during both the northerly and southerly wind periods is as follows:
Latent and sensible heat release from the underlying sea differed greatly between the northern and southern sides of the SST front (Fig. 3a). The heat release from the Kuroshio Extension in the northerlies was stronger than in the southerlies (not shown). The resulting strong meridional contrast in air temperature was evident up to 300-m altitude on average, extending farther up to 1500-m altitude while decaying.
High and low pressure anomalies formed on the northern and southern sides, respectively, of the SST front through the pressure adjustment mechanism (Fig. 6).
The anomalous meridional pressure gradient that was yielded was associated with the easterly geostrophic wind component in the lower troposphere (Figs. 7 and 8).
The clockwise (reverse Ekman) spiral in the wind profile was observed from about 1500-m altitude down to the surface under the surface northerlies (Fig. 10a), while the counterclockwise spiral was observed under the southerlies (Fig. 10b), in a manner consistent with thermal wind associated with the air temperature anomalies forced by the SST front.
A meridional circulation somewhat analogous to sea breeze was observed below the 300-m altitude, especially under the surface southerlies (Fig. 9).
All of these phenomena, except the sea-breeze analog, persisted even during the passage of synoptic-scale disturbances that were shown by the reversal of the cross-frontal winds several times (Figs. 4, 6a, 6b, and 7a).
The observations demonstrated that both the pressure adjustment mechanism and the vertical mixing mechanism were operative, but at different locations. The vertical mixing mechanism operated only over the warm SST south of the front under the northerlies, which led to the formation of a well-defined mixed layer in which wind speed and direction were rather homogeneous. The anomalous cold advection by the northerlies reduced near-surface static stability over the warm Kuroshio Extension to enhance vertical mixing locally, whereas anomalous warm advection by the southerlies enhanced static stability and thereby suppressed vertical mixing on the cooler side of the SST front.
Although these phenomena were observed only along a particular meridional section across the Kuroshio Extension front, we hypothesize that they can occur throughout the length of the front. Figure 11 illustrates lower-tropospheric features that could be realized along the Kuroshio Extension front east of Japan if the hypothesis is correct. The observations suggest that the pronounced north–south SST contrast on the northern flank of the Kuroshio Extension acts to strengthen lower-tropospheric easterly winds associated with the thermally driven northward pressure gradient. Specifically, the lower-tropospheric northerlies are veering downward in association with the enhanced surface northeasterlies, while vertical mixing becomes active along the southern flank of the Kuroshio Extension (Fig. 11a). The free-tropospheric southerlies are, by contrast, backing downward in association with the surface southeasterlies (Fig. 11b). One may wonder whether the anomalous near-surface easterlies forced by the SST front along the Kuroshio Extension can bring a cool maritime air mass toward the east coast of Japan, especially at latitudes slightly to the north of the front. As indicated by Fig. 7, the anomalous alongfront easterlies were stronger under the cool northerlies, and so the resultant coastal cooling could be, if it actually occurs. The significance of this cooling will be assessed in a future study.
Schematics illustrating the oceanic influence upon the atmosphere observed around the SST front along the Kuroshio Extension under the surface (a) northerlies and (b) southerlies. Numbers 1–5 correspond to the itemized summary in section 4.
Citation: Monthly Weather Review 144, 10; 10.1175/MWR-D-15-0442.1
The three-ship observations offered a unique opportunity to capture direct evidence for the anomalous winds and circulation within the near-surface layer driven by a midlatitude SST front. The reversal of the Ekman wind spiral may be a useful indicator that an observed wind profile is SST driven if it is observed in even a single radiosonde sounding over an SST front. Because data from many radiosonde soundings are available, statistical studies of wind profiles based on these data may be able to confirm our hypotheses.
Acknowledgments
We extend our thanks to all of those involved in the intensive observation campaign, especially the captains, crews, and cruise members of R/V Seisui-maru, R/V Wakataka-maru, and R/V Tansei-maru. This study was supported by the Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) through a Grant-in-Aid for Scientific Research in Innovative Areas (22106007, 22106003, 22106004, 22106005, 22106006, and 22106009). Comments from participants at the Research Meeting on Air–Sea Interaction, as part of the Collaborative Research Program of HyARC, Nagoya University, also were helpful. Dr. Tomita of Nagoya University calculated the latent and sensible heat fluxes. We also thank the editor and anonymous reviewers for their valuable comments and suggestions that improved the quality of the paper. We used the Generic Mapping Tools package (Wessel and Smith 1991) to construct three-dimensional (time–latitude–height) gridded data from the observational data, which were collected at unequal intervals.
REFERENCES
Fairall, C. W., E. F. Bradley, J. E. Hare, A. A. Grachev, and J. B. Edson, 2003: Bulk parameterization of air–sea fluxes: Updates and verification for the COARE algorithm. J. Climate, 16, 571–591, doi:10.1175/1520-0442(2003)016<0571:BPOASF>2.0.CO;2.
Hayes, S. P., M. J. McPhaden, and J. M. Wallace, 1989: The influence of sea surface temperature on surface wind in the eastern equatorial Pacific: Weekly to monthly variability. J. Climate, 2, 1500–1506, doi:10.1175/1520-0442(1989)002<1500:TIOSST>2.0.CO;2.
Kasamo, K., and Coauthors, 2014: Transient and local weakening of surface winds observed above the Kuroshio front in the winter East China Sea. J. Geophys. Res. Atmos., 119, 1277–1291, doi:10.1002/2013JD020610.
Kawai, Y., H. Tomita, M. F. Cronin, and N. A. Bond, 2014: Atmospheric pressure response to mesoscale sea surface temperature variations in the Kuroshio Extension region: In situ evidence. J. Geophys. Res. Atmos., 119, 8015–8031, doi:10.1002/2013JD021126.
Kawai, Y., T. Miyama, S. Iizuka, A. Manda, M. Yoshioka, S. Katagiri, Y. Tachibana, and H. Nakamura, 2015: Marine atmospheric boundary layer and low-level cloud responses to the Kuroshio Extension front in the early summer of 2012: Three-vessel simultaneous observations and numerical simulations. J. Oceanogr., 71, 511–526, doi:10.1007/s10872-014-0266-0.
Kida, S., and Coauthors, 2015: Oceanic fronts and jets around Japan: A review. J. Oceanogr., 71, 469–497, doi:10.1007/s10872-015-0283-7.
Lindzen, R. S., and S. Nigam, 1987: On the role of sea surface temperature gradients in forcing low-level winds and convergence in the tropics. J. Atmos. Sci., 44, 2418–2436, doi:10.1175/1520-0469(1987)044<2418:OTROSS>2.0.CO;2.
Minobe, S., A. Kuwano-Yoshida, N. Komori, S.-P. Xie, and R. J. Small, 2008: Influence of the Gulf Stream on the troposphere. Nature, 452, 206–209, doi:10.1038/nature06690.
Miyazawa, Y., and Coauthors, 2009: Water mass variability in the western North Pacific detected in a 15-year eddy resolving ocean reanalysis. J. Oceanogr., 65, 737–756, doi:10.1007/s10872-009-0063-3.
Nishikawa, H., Y. Tachibana, and Y. Udagawa, 2014: Radiosonde observational evidence of the impact of an extremely cold SST spot on a mesoscale anticyclone. J. Geophys. Res. Atmos., 119, 9183–9195, doi:10.1002/2014JD021538.
Nonaka, M., and S.-P. Xie, 2003: Covariations of sea surface temperature and wind over the Kuroshio and its extension: Evidence for ocean-to-atmosphere feedback. J. Climate, 16, 1404–1413, doi:10.1175/1520-0442(2003)16<1404:COSSTA>2.0.CO;2.
Pezzi, L. P., R. B. Souza, M. S. Dourado, C. A. E. Garcia, M. M. Mata, and M. A. F. Silva-Dias, 2005: Ocean–atmosphere in situ observations at the Brazil-Malvinas Confluence region. Geophys. Res. Lett., 32, L22603, doi:10.1029/2005GL023866.
Tanimoto, Y., S.-P. Xie, K. Kai, H. Okajima, H. Tokinaga, T. Murayama, M. Nonaka, and H. Nakamura, 2009: Observations of marine atmospheric boundary layer transitions across the summer Kuroshio Extension. J. Climate, 22, 1360–1374, doi:10.1175/2008JCLI2420.1.
Tanimoto, Y., T. Kanenari, H. Tokinaga, and S.-P. Xie, 2011: Sea level pressure minimum along the Kuroshio and its extension. J. Climate, 24, 4419–4434, doi:10.1175/2011JCLI4062.1.
Tokinaga, H., Y. Tanimoto, and S.-P. Xie, 2005: SST induced surface wind variations over the Brazil Malvinas confluence: Satellite and in situ observations. J. Climate, 18, 3470–3482, doi:10.1175/JCLI3485.1.
Tokinaga, H., and Coauthors, 2006: Atmospheric sounding over the winter Kuroshio Extension: Effect of surface stability on atmospheric boundary layer structure. Geophys. Res. Lett., 33, L04703, doi:10.1029/2005GL025102.
Wallace, J. M., T. P. Mitchell, and C. Deser, 1989: The influence of sea surface temperature on surface wind in the eastern equatorial Pacific: Seasonal and interannual variability. J. Climate, 2, 1492–1499, doi:10.1175/1520-0442(1989)002<1492:TIOSST>2.0.CO;2.
Wessel, P., and W. H. F. Smith, 1991: Free software helps map and display data. Eos, Trans. Amer. Geophys. Union, 72, 441–446, doi:10.1029/90EO00319.
Xu, M., and H. Xu, 2015: Atmospheric responses to Kuroshio SST Front in the East China Sea under different prevailing winds in winter and spring. J. Climate, 28, 3191–3211, doi:10.1175/JCLI-D-13-00675.1.
See online at http://www.atmos.rcast.u-tokyo.ac.jp/hotspot/index_eng.html.
