A study of the Xisha eddy is a good supplement to comprehend the local oceanic and atmospheric currents. Our work aims at analyzing the intraseasonal Air–Sea interaction process during the occurring of the Xisha warm eddy. Using a Complex Empirical Orthogonal Function statistical technique, we found an eddy that propagated southwestward along the continental slope of the northwestern South China Sea, at a similar phase speed to that of Rossby waves, at 0.12 m s−1. Intraseasonal variability in spring is related to the ocean-to-atmosphere interaction, and involves the cloud-radiation effects on surface sea temperature, as well as its impact on lower-level convergence over the South China Sea. This process, induced by the warm eddy, results in abnormal surface sea temperature, downward shortwave/longwave radiation flux, surface latent heat flux, and wind changes. We used composite results and the Weather Research and Forecasting Model to explore how the observed surface sea temperature anomalies influence precipitation in the South China Sea. The results showed that the atmospheric transport of heat and moisture improved with respect to the surface sea temperature max and the air active events have caused lower atmosphere instability, along with the lower pressure and enhanced precipitation frequency in spring.

Introduction

The Xisha Islands are located in the northwest part of the South China Sea (SCS), and are involved in the air-sea interaction on multi-time scales with respect to large-scale circulation (Qu, 2000; Su, 2004). The northern SCS (NSCS) is characterized as a region with high mesoscale eddy activity, influenced by monsoons, Kuroshio intrusion (Wyrtki, 1961; Wang et al., 2011), and complex topography. Many anticyclonic eddies originating near the Luzon Strait greatly influence the large-scale circulation of the currents (Yuan et al., 2008, Shu et al., 2011). Using the Argo drifting buoys' tracks and TOPEX/Poseidon (T/P) altimetry data, He et al. (2002) confirmed a series of activities of mesoscale eddies. Using multi-satellite remote sensing data and in situ hydrographic data, Wang et al. (2008) presented two anticyclonic eddies that propagated southwestward along the continental slope of the NSCS at 0.0 97 m s−1, similar to the phase speed of Rossby waves, which travel at 0.0105 m s−1. Xiu et al. (2010) compiled data on eddy activity in the South China Sea and pointed out that the wind stress curl is an important, but not the only, mechanism of eddy genesis in the SCS. Zu et al. (2013) studied the evolution process of warm eddies that generated southwest of Taiwan and found that their southwestward movement was due to the background flow patterns.

The air–sea interaction is more obvious in the intraseasonal time scale of the SCS (Krishnamurti et al., 1988; Fu et al., 2006), than the seasonal time scale. Drifting buoy and satellite observations with high temporal resolution showed distinct evidence of the intraseasonal sea surface temperature (SST) oscillation in the SCS (Sengupta et al., 2001; Xie et al., 2007). When the ocean SST is above 27°C, it is conducive for enhanced convective precipitation (Lau et al., 1997). Kemball-Cook and Wang (2001) suggested that the negative latent heat flux (upward negative) anomalies enhance the northward propagating convective anomalies through an increase in the moist static energy. Mao and Chan (2005), using NCEP-NCAR reanalysis data, showed intraseasonal air–sea interaction makes the overall variations differ from year to year. Zeng and Wang (2009) attempted to use the weekly satellite to estimate the latent heat flux anomalies, SST, wind speed, and near-surface air humidity, and revealed that the intraseasonal SST variation significantly changed the intraseasonal latent-heat flux, reducing it by 20% in winter. Fu and Wang (2004) demonstrated that the coupled SST-convection in the model has a structure similar to the one observed. Understanding the relationship of SST-precipitation and the air–sea interaction involved is crucial for evaluating and rectifying model forecasts (Wu et al., 2008). Fu et al. (2008) pointed out that the positive intraseasonal SST can trigger convective disturbances by increasing the temperature and moisture in the atmospheric boundary layer.

This article focuses on the relationship between the intraseasonal variations of SSHA, SSTA, rainfall, and heat fluxes during the Xisha anticyclonic eddies come up in the spring. It is interesting that the SST changes the interaction with the atmosphere by influencing the stability of the atmosphere. Using the Weather Research and Forecasting Model (WRF), potential relationships among the surface fluxes, atmosphere circulation, and SST are explored.

Data and analysis methods

This study uses 3-day measurements of means of SST and precipitation from the Tropical Rainfall Measuring Mission (TRMM), which has much better spatial coverage than the daily data. The TRMM Microwave Imager (TMI) provides TMI data (Version 4) based on a 0.25° grid collected from 1998 to 2012, and was unaffected by clouds, aerosols, or atmospheric water vapor (Wentz et al., 2000). The TMI (available at http://www.ssmi.com/) revealed subseasonal SST perturbations are considerably larger than there analysis product (Sengupta et al., 2001; Vecchi and Harrison, 2002). The daily Outgoing Longwave Radiation (OLR) data is provided by the National Oceanic and Atmospheric Administration (NOAA)'s National Center for Environmental Prediction (NCEP). The merged sea surface height anomaly (SSHA) data are from TOPEX/Poseidon, Jason 1, and ERS (European Research Satellite) provided by the French Archiving, Validation and Interpretation of Satellite Oceanographic Data (AVISO) project. The average daily data, gridded with 1/3° resolution and with orbit error and tides bias removed (Jacobs et al., 2002), are used to evaluate the evolution of the Xisha's warm eddy.

SSHA, SST, and OLR showed strong seasonal signals. In order to identify the air–sea features around the warm eddies in the spring; we used Fast Fourier Transform (FFT) to filter out the seasonal components and other low frequency features. The FFT decomposes a sequence of values into components of different frequencies. Letting x0… xN-1 be complex numbers, the FFT is defined by the formula: , k = 0…N-1. This moderate formula is used to avoid any energy loss from either side of the time series, unlike Butterworth and other filters.

The resulting empirical orthogonal function is a statistical way to decompose a space–time field into spatial patterns and their associated time indices. Complex EOF (C-EOF) was used to analyze a set of time series data that presented supplemental phase lag components rotated by 90 degrees on a complex plane (called Hilbert transform) to establish the relationship between power and phase on a specific frequency band (Hannachi et al., 2007). We then combined the intraseasonally filtered SSHA from the spring (from February to May) as a time series database. C-EOF helped us to seize the evolution feature of the Xisha's warm eddies. And according to the C-EOF result, we composite subseasonal SST, OLR, and SSH anomalies with respect to the arrival of Xisha warm eddy in the last 15 years (1998–2012).

Results

The C-EOF of the intraseasonally filtered SSHA in spring shows that there is a quasi-steady sea-surface height anomaly around the Xisha Island area (111–112°E, 16–17°N) before the onset of the summer monsoon (Figure 1).The anticyclonic eddies might be generated near the Luzon Strait (Wang et al., 2005, 2013) and propagate southwestward along the continental slope of the northern SCS. During the eddy's life cycle, the high value field shows that the SSHA contour was closed and the SSHA maximum was greater than 7.5 cm, which satisfies the two basic criteria of mesoscale eddy identification (Wang et al., 2003). The warm eddy moved at about 0.12 m s−1, which is similar to the phase speed of Rossby waves in the northern SCS, and finally disappeared along the eastern coast of Vietnam. Generally the eddy will come from the northeastern of the South China Sea in March, mature around the Xisha Island area in May, and die out before the summer monsoon season. Eddies might increase the local SST about 0.5°C, which is generally around 28°C in spring. While the larger, subseasonal SST change might be related to surface heat flux changes.

According to the mesoscale eddy identification criteria, we selected 23 warm eddies of a similar kind that occurred in the spring of the years between 1998 and 2012. Three warm eddies appeared in 2006, and two warm eddies formed in 1999, 2002, 2007, 2008 and 2010, while only one warm eddy occurred in the other years we studied. We further investigated the relationship of the intraseasonal variations of SSHA, SSTA, rainfall, and OLR of the warm, spring time eddies near the Xisha Islands (Figure 2). It is interesting that SSHA and SSTA have a reasonable relationship (correlation index 0.40), but we should bear in mind that, in addition to the effect of eddy activities, the atmosphere's circulation could also affect the local temperature changes. Thus we classified these cases into three groups according to whether the OLR was in/out of phase with the SSHA. Let the correlation coefficient between OLR and SSHA ro/h denote the explained variance by the upper ocean or the atmosphere dominant. Then the spring eddies are categorized according to the following criteria:

  • The warm eddy is dominant category ro/h < −0.4;

  • Dual category −0.4 < = ro/h < = 0.4;

  • The atmosphere is dominant category ro/h > 0.4.

Based on this classification, these eddies can be grouped as follows:

  • The warm eddy is dominant category: 1998, 1999 (first one), 2000, 2007 (second one), 2008 (second one), 2009 (first one), 2009 (second one);

  • Dual category: 1999 (second one), 2001, 2002 (first one), 2002 (second one), 2005, 2006 (first one), 2006 (second one), 2007 (first one), 2008 (second one), 2010 (first one), 2012;

The atmosphere is dominant category: 2003, 2004, 2006 (third one), 2010 (second one), 2011(Figure 2).

Figure 2y shows the composite result of the atmosphere's affect on the upper ocean for the first group. It is possible that the atmosphere affects the upper ocean at certain times (Figure 2y), while at other times, the OLR is mainly influenced by eddies, there are 5 over the past 24 years during which it has occurred in this case (about 22%), like Figure 2c and n and so on. And it also means that the eddy has an extraordinary influence in enhancing spring precipitation (Figure 2y). The second gourp shows that the increased convection and rainfall caused by the atmospheric circulation in the Xisha area offset the impact from eddies that cause a decrease in SST in Figure 2x. It is noteworthy that Figure 2y shows the evolution of composite subseasonal SST, OLR, and SSH anomalies with respect to the subseasonal rainfall peak in the Xisha Island region during the arrival of warm eddies, which suggests the influence of warm eddies on the local atmosphere. Composite analysis of warm eddy events with increased OLR anomalies in the Xisha region are significantly influenced by the SST anomalies, and then leaded to the rain anomalies. And the last group stands among upper groups.

The large subseasonal SST changes during the arrival of warm eddies might be related to surface heat flux changes. This is further investigated by the simulated surface latent heat flux and wind anomalies from a SCS atmospheric model, rather than in situ and reanalysis data, because of the low spatial resolution and coverage of the latter. A WRF model that covers a domain of −2°S–31°N, 93°–137°E with a horizontal resolution of 24 km (Figure 3) was utilized. In order to obtain precise output, the model only forecasted three days at a time. We then took those forecasts and combined them for the date range of 15 May–4 June 2011. The warm eddy reached Xisha on 22 May 2011 (Figure 2v). Figure 4 shows five days mean of SSTA between 20 and 24 May 2011. We drew a straight line along the zonal direction through the Xisha warm pool to get knowledge of its role. This model dependence shows a positive relationship between variables such as latent heat and shortwave fluxes, rainfall, and surface winds in this study from 15 May to 4 June 2011.

Latitude–time plots of daily composite variables averaged over the region of 16.5°N to 18°N are shown in Figure 5. Coherent westward migrations of the anomalies over the northwestern SCS are observed from 114°E to 105°E. When the warm eddy propagated westward, the OLR anomalies were out-phased, with SST increasing from 235w m−2 to 275 w m−2 (Figure 5a). At the same time, surface latent heat anomalies decreased by 60 w m−2 (Figure 5b). When SST increased, the longwave radiation flux reached 380 w m−2, after starting at 415 w m−2, and the shortwave radiation flux exceed to 315 w m−2. Figures 5e and f show that local winds shift from southwestward to northeastward. Obviously, when a warm eddy travels through and settles in the Xisha Island area, southwestward wind will change into northeastward ones. The large subseasonal SST changes are related to surface heat flux changes. This process leads to a change of longwave/shortwave fluxes and the enhancement of latent heat when an eddy reaches Xisha. But longwave/shortwave fluxes increase and decrease when an eddy impairs them. In this period, the convection would be more active, and the atmospheric transport of heat and moisture would increases, which results in the increase of the frequency of precipitation in the spring.

Discussion

The C-EOF shows that the Xisha warm eddies propagated southwestward along the continental slope of the northern SCS at the speed of 0.12 m s−1, these results are similar with by Wang et al. (2008). At least one anticyclonic eddy typically appears in the Xisha region every spring and dies out near the eastern coast of Vietnam.

The intraseasonal variability of wind and surface heat flux in the spring is related to the air–sea interaction, which is mainly due to the increase in SST, which enhances the convection by modifying the lower-level convergence of the atmosphere over this region. During the arrival of an eddy at Xisha, the atmosphere may play an important role in reducing the local sea level in spring.

We used data from the spring of 2011 for this study, as it reflected how the local conditions influence the atmosphere. The large subseasonal SST changes during the arrival of a warm eddy might be related to surface heat flux (the air gained and then released) and wind changes (southwestward to northeastward). Using the SST tendency equation,, where Ts is the SST, Ftot is the total heat flux, q is density of water, cp is the specific heat of water at constant pressure, and h is the mixed layer depth, we were able to find the degree to which the surface fluxes accounted for the observed intraseasonal variations in SST (Hendonand Glick, 1997). In general, MLD (Mixed Layer Depth) might reach up to 60 m in the winter and shoal down to about 30 m in spring. The observed SST tendencies varied coherently with those derived from the surface heat fluxes, revealing that the surface heat flux anomalies accounted for the observed SST variability (Roxy and Tanimoto, 2012).

Conclusions

Using a 15-year time series of SSH, warm eddies in the Xisha area was studied before the summer monsoon onset using the C-EOF statistical technique. Special attention was paid to the relationship of local oceanic and atmospheric interaction. Using the WRF model, we attempted to view the large subseasonal SST changes during the arrival of warm eddies and whether they are related to surface heat flux and wind changes.

Funding

This study was supported by the National Natural Basic Research Program of China (973 Program, No. 2011CB403504) and the National Natural Science Foundation of China (Nos. 41106028, 41206011 and 41406038).

References

Fu, X., and Wang, B.,
2004
.
The boreal summer intraseasonal oscillations simulated in a hybrid coupled atmosphere-ocean model, Mon
.
Weather Rev.
132
,
2628
2649
.
Fu, X., Wang, B., Tao, L.,
2006
.
Satellite data reveal the 3-D moisture structure of tropical intraseasonal oscillation and its coupling with underlying ocean
.
Geophys Res Lett
33
,
L03705
. doi:.
Fu, X., Yang, B., Bao, Q., Wang, B.,
2008
.
Sea surface temperature feedback extends the predictability of tropical intraseasonal oscillation
.
Mon Weather Rev
136
,
577
597
. doi:.
Hannachi, A, Jolliffe, I.T., and Stephenson, D B.,
2007
. Empirical orthogonal functions and related techniques in atmospheric: A review.
Int. J. Climatol.
Published online by Wiley InterScience
.
He, Z. G., Wang, D., Hu, J.Y.,
2002
.
Features of eddy kinetic energy and variations of upper circulation in the South China Sea
.
ActaOceanol.Sinica
,
21
(
2
),
305
314
.
Hendon, H. H., Glick, J.,
1997
.
Intraseasonal air–sea interaction in the tropical Indian and Pacific Oceans
.
J. Climate
10
,
647
661
.
Jacobs, G. A., Barron, C. N., Fox, D. N., Whitmer, K. R., Klingenberger, S., May, D., and Blaha, J. P.,
2002
.
Operational Altimeter Sea Level Products
,
Oceanography
,
15
(
1
),
13
21
.
Kemball-Cook, S., Wang B.,
2001
.
Equatorial waves and air-sea interaction in the boreal summer intraseasonal oscillation
.
J. Clim
.
14
(
13
),
2923
2942
.
Krishnamurti, TN, Oosterhof, D. K, Mehta A. V.,
1988
.
Air–sea interaction on the time–scale of 30–50 days
.
J AtmosSci
45
,
1304
1322
. doi:.
Lau, K. M, Wu, H.T., Bony, S.,
1997
.
The role of large-scale atmospheric circulation in the relationship between tropical convection and sea surface temperature
.
J. Clim
.
10
(
3
),
381
392
.
Mao, J.Y, Chan, J.C.L,
2005
.
Intraseasonal variability of the South China Sea summer monsoon
.
J. Clim
18
(
13
),
2388
2402
.
Qu, T. D.,
2000
.
Upper-layer circulation in the South China Sea
,
J. Phys. Oceanogr.
30
,
1450
1460
.
Roxy, M., Tanimoto, Y.,
2012
.
Influence of sea surface temperature on the intraseasonal variability of the South China Sea summer monsoon
.
ClimDyn.
39
(
5
),
1209
1218
. doi:.
Sengupta, D., Goswami, B., Senan, R.,
2001
.
Coherent intraseasonal oscillations of ocean and atmosphere during the Asian summer monsoon
.
Geophys. Res. Lett.
28
(
21
),
4127
4130
. doi:
Shu, Y.Q., Wang, D.X., Zhu, J., Peng, S.Q.,
2011
.
The 4-D structure of upwelling and Pearl River plume in the northern South China Sea during summer 2008 revealed by a data assimilation model
.
Ocean Modelling
36
,
228
241
.
Su, J. L.,
2004
.
Overview of the South China Sea circulation and its influence on the coastal physical oceanography near the Pearl River estuary
.
Cont. Shelf. Res.
,
24
,
1745
1760
.
Wang, D., Wang, Q., Zhou, W., Cai, S., Li, L., Hong, B.,
2013
.
An analysis of the current deflection around Dongsha Islands in the northern South China Sea, J
.
Geophys. Res. Oceans
,
118
,
490
501
, doi:.
Wang, D., Xu, H., Lin, J., Hu, J.,
2008
.
Anticyclonic eddies in the northeastern South China Sea during winter 2003/2004
,
J. Oceanogr.
,
64
,
925
935
. doi:.
Wang, G., Su, J. Chu, P.C.,
2003
.
Mesoscale eddies in the South China Sea detected from altimeter data
.
Geophys. Res. Lett.
,
30
,
2121
. doi:.
Wang, G., Su, J., Li, R.F.,
2005
.
Mesoscale eddies in the South China Sea and their impact on temperature profiles
.
ActaOceanol.Sin.
,
24
,
39
45
.
Wang, Q., Wang, Y.X., Hong, B., Zhou, W.D., Wang, D.X.,
2011
.
Different roles of Ekman pumping in the west and east segments of the South China Sea Warm Current
.
ActaOceanol.Sin
.,
30
(
3
),
1
13
.
Wentz, F.J., Gentemann, C., Smith, D., Chelton, D.,
2000
.
Satellite measurements of sea surface temperature through clouds
.
Science
288
,
847
850
. doi:.
Wu, R., Kirtman, B.P., Pegion K.,
2008
.
Local rainfall-SST relationship on subseasonal time scales in satellite observations and CFS
.
Geophys. Res. Lett.
35
(
22
),
L22706
. doi:.
Wyrtki, K.,
1961
.
Physical oceanography of the Southeast Asian waters
.
Naga Rep
.,
2
,
1
195
.
Xie, S.P., Chang, C.H., Xie, Q., Wang, D.X.,
2007
.
Intraseasonal variability in the summer South China Sea: wind jet, cold filament, and recirculations
.
J. Geophys. Res.
112
,
C10008
. doi:
Xiu, P., Chai, F., Shi, L., Xue, H.J., Chao, Y.,
2010
.
A census of eddy activities in the South China Sea during 1993–2007
,
J. Geophys. Res.
115
,
C03012
, doi:.
Yuan, Y.C., Liao, G.H., Yang, C.G.,
2008
.
The Kuroshio near the Luzon Strait and Circulation in the Northern South China Sea during August and September1994
,
J. Oceanogr.
64
,
777
788
.
Zeng, L., Wang, D.,
2009
.
Intraseasonal variability of latent-heat flux in the South China Sea
.
TheorApplClimatol.
97
(
1
),
53
64
. doi:
Zu, T., Wang D., Yan C., Belkin I., Zhuang W. and Chen J.,
2013
.
Evolution of an anticyclonic eddy southwest of Taiwan
.
Ocean Dynamics
63
,
519
531
.