The effects of chlorophyll on upper ocean temperature and circulation in the two upwelling regions of the South China Sea are investigated by comparing results of two experiments using the solar radiation penetration scheme with and without the effects of chlorophyll. In boreal winter, the sea surface temperature anomalies were negligible with the existence of chlorophyll throughout most of the South China Sea because of the deep mixed layer. The strong mixing stirred by surface winds brings the cold water into the mixed layer and cancels out heating due to the effect of chlorophyll. In boreal summer, the high chlorophyll concentration in the upwelling region southeast of Vietnam cools the upwelling water below the mixed layer and leads to lower sea surface temperatures. The present study confirms the processes proposed by previous studies in the equatorial Pacific that indicate that the mixed layer depth is important to the response of the surface temperature and current to biological heating.
The existence of phytoplankton pigments can substantially affect upper ocean temperature and circulation by altering the vertical distribution of solar radiation. This effect and its associated processes have been investigated using observational data (Lewis et al., 1983, 1990; Sathyendranath et al., 1991; Siegel et al., 1995; Strutton and Chavez, 2004) and numerical models (Nakamoto et al., 2000, 2001, 2002; Murtugudde et al., 2002; Sweeney et al., 2005; Marzeion et al., 2005; Lin et al., 2007, 2008; Wu et al., 2007). These previous studies indicate that a high chlorophyll concentration is able to trap more solar radiation in the ocean surface mixed layer, and hence, less energy is available to heat the water below the mixed layer. The redistribution of solar radiation in the water column therefore results in warming of the upper layer and cooling of the lower layer, which finally leads to a shallow mixed layer and strong stratification in the upper layer.
In the upwelling regions, the responses of the temperature and current in the upper layer are affected by not only the one-dimensional thermodynamic process but also the ocean dynamics processes. Recent studies of numerical model sensitivity experiments indicate that the high chlorophyll concentration leads to an enhanced upwelling and a lower sea surface temperature (SST) in the upwelling regions (Manizza et al., 2005; Anderson et al., 2007, 2009; Lin et al., 2007, 2008). Enhanced upwelling is mainly caused by the horizontal gradient of chlorophyll concentration, which enhances the horizontal gradient of the upper ocean temperature and then strengthens the circulation. A lower SST combines the effects of the enhanced upwelling and the cooled water below the mixed layer. However, the previous studies focussed almost entirely on the equatorial Pacific. The responses of the temperature and current in other upwelling regions therefore need to be investigated.
The South China Sea (SCS) is the largest marginal sea in the western Pacific (Figure 1). Under the influence of the East Asian monsoon, the SCS has a distinct seasonal variability: a cyclone circulation in boreal winter and an anticyclone circulation in boreal summer (Shaw et al., 1994, 1999; Qu, 2000; Yang et al., 2002). The upwellings and associated cold SSTs also present seasonal variation. The upwellings occur northwest of Luzon in the boreal winter (Shaw et al., 1996) and southeast of Vietnam in the boreal summer (Xie et al., 2003). Boxes A and B in Figure 1 represent the two upwelling regions, respectively. The mechanisms that drive the upwelling (Shaw et al., 1996; Xie et al., 2003) and the variations of SSTs in these regions (Qu, 2001; Xie et al., 2003; Liu et al., 2004) have been systematically investigated. Recently, the effect of upwelling on the bloom of phytoplankton has been examined using satellite products (Liu et al., 2002). However, the effects of variations of water optical properties due to the change of chlorophyll on the surface temperature and current in the SCS have not been reported in the literature.
The purpose of the present study is to investigate the effect of chlorophyll on the upper ocean in the two upwelling regions of the SCS, and to verify the processes proposed by previous studies in the equatorial Pacific. The results of the present study can also be used to understand the interaction between the physical and biological processes in the ocean and the changes of the relationship in the context of global warming.
Model and Experiments
In the present study, the LASG/IAP (State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics/Institute of Atmospheric Physics) Climate System Ocean Model (LICOM, Liu et al., 2004a, b) is employed. LICOM is the oceanic component of the Flexible Global Ocean-Atmosphere-Land System model. LICOM is one of the models that is used in the Coupled Model Intercomparison Project Phase 3 (Yu et al., 2004), which has been cited by the Fourth Assessment Report of the Intergovernmental Panel on Climate Change.
LICOM has 30 vertical layers with 15 uniform layers in the top 150 m. The horizontal resolution of the model is 1° in the zonal direction, while the meridional resolution gradually varies from (1/2)° at the equator to 1° outside of the tropics (10°S–10°N). Therefore, the horizontal resolution is rather coarse and cannot resolve mesoscale eddies and finer circulations in the SCS. However, the evaluations of the model in the results section show that LICOM can reproduce the basin scale circulation and SST in the SCS well. Moreover, the present study is primarily based on a comparison between two experiments, which can effectively remove the systematic biases of the model.
The surface boundary condition of temperature takes the form of bulk formula following Large and Yeager (2004). The forcing fields are from Common Ocean-ice Reference Experiments (Griffies et al., 2009). Canuto et al.'s (2001) scheme is used for vertical mixing, and the Gent and McWilliams (1990) scheme is used for parameterizing the effects of mesoscale eddies. Simulated seasonal cycles of SST and sea surface height (SSH) in the SCS are evaluated against the observation in the following section.
In LICOM, a two-exponential formula of Paulson and Simpson (1977, hereafter PS77) is used to parameterize solar radiation penetration:
The Ohlmann (2003, hereafter O03) scheme was used to represent the effect of chlorophyll concentration on the solar radiation penetration. The Ri(i = 1, 2) and 1/Ki(i = 1, 2) in the O03 scheme are all functions of chlorophyll concentration and can be found in Table 1 of Ohlmann (2003).
|A in January||B in July|
|A in January||B in July|
Two experiments were conducted using the PS77 and O03 schemes. The climatological monthly mean of the Sea-viewing Wide Field-of-view Sensor chlorophyll concentration (http://oceancolor.gsfc.nasa.gov/) from 1998 to 2007 was used to compute the shortwave absorption profile. Both experiments were run for 18 years (1990–2007) and were forced by daily data from Common Ocean-Ice Reference Experiments.
Chlorophyll affects the vertical distribution of solar radiation by changing the radiation attenuation depth. Figure 2 shows the chlorophyll concentration (shaded) and attenuation depth (contour) computed with the method of Ohlmann (2003) in January and July. In general, the chlorophyll concentration is high in boreal winter (Figure 2a) and low in boreal summer (Figure 2b). In both seasons, the attenuation depth is shallower than 23 m, which indicates that more heat will be trapped in the upper layers. Therefore, the SST in the O03 run should be higher than that in the PS77 run in the absence of ocean dynamics. The chlorophyll concentration is relatively high in the upwelling region: northwest of Luzon in boreal winter and southeast of Vietnam in boreal summer. Although the chlorophyll concentration is high in the coastal regions, we do not focus on this region in the present study because of the coarse resolution of the model.
In the present study, the coarse resolution version of LICOM, which is primarily used in climate study, is employed to investigate the effects of chlorophyll on the upper layer temperature and current. The ability of LICOM to reproduce the upper layer circulation and stratification in the SCS is evaluated against the observation first. Figure 3 shows the SST for LICOM and for the Optimum Interpolation Sea Surface Temperature (Reynolds et al., 2002) in January and July. Because of its geographical location, the SST in the SCS presents remarkable seasonal variation related with the Asian monsoon. A cool anomaly occurs northwest of Luzon in boreal winter (Figure 3a) and southeast of Vietnam in boreal summer (Figure 3b). LICOM has reproduced a reasonable SST pattern in both seasons, including the two cool anomalies. In general, the simulated SST is approximately 1°C higher than that in the observation. The warm bias in LICOM may be partially attributed to the warm water that enters the SCS from the exaggerated large entrance between the SCS and the Sulu Sea.
Figure 4 shows the SSH anomalies for LICOM (1998–2007) and for Developing Use of Altimetry for Climate Studies (1998–2007, http://www.aviso.oceanobs.com), which merged altimetry data from all satellite missions in January and July. The primary feature of the circulation in the SCS also shows strong seasonality: there is a cyclone circulation in boreal winter and an anticyclone circulation in boreal summer. These features are associated with negative and positive SSH anomalies, respectively. LICOM generally simulates the large-scale circulation well. In boreal winter, the SSH anomaly in the region northwest of Luzon has a magnitude comparable with the observation. During this time, the SSH anomaly orientation is mainly west–east, and the observation is southwest–northeast (Figure 4a). LICOM reproduces the location of the negative SSH anomaly southeast of Vietnam in boreal summer well, but the magnitude is smaller because of the coarse resolution. This result also indicates a slow western boundary current in LICOM.
In summary, LICOM has the ability to reproduce the large-scale pattern and seasonal variation of SST and circulation in the upper layer in the SCS. Therefore, the model can be used for the present study. The warm bias in the SST and the slow western boundary are mainly due to the coarse resolution.
The response of temperature and circulation
The effects of chlorophyll are quantified by the difference between O03 and PS77. Figure 5 shows the SST and surface current differences in January and July. Because of the absorption of sunlight by phytoplankton pigments, the ocean usually warms at the surface layer and cools at the subsurface layer. However, the entire SCS cools with a small magnitude (less than 0.05°C) at the surface in boreal winter when the chlorophyll concentration is high (Figure 5a). The warm temperature differences only occur around the Luzon strait and the entrance from the Sulu Sea to the SCS. Subsurface cooling is also shown in Figure 6a. It is interesting that there is no clear anomaly center collocating within the upwelling center northwest of the Luzon (Figure 4a). In addition, there is no anomalous divergence center of the currents in this region. These results indicate that, unlike the processes proposed by previous studies, the response of the upper ocean SST in the upwelling region of Luzon is not dominated by the cold water upwelled from the subsurface.
To understand the slight cooling at the surface, we further investigate the upper layer stratification and the current. Figure 7a shows the differences of temperature and current as well as the Mixed Layer Depth (MLD) from the two experiments along 18°N in January. The vertical current differences in the mixed layer show small anomalous downwelling in the region northwest of Luzon (Figure 7a); this result confirms that the SST in this region is not dominated by upwelling. The MLD from the O03 experiment is 3–5 m shallower than the PS77 experiment (Figure 7a); this result means the stratification is slightly intensified because of the biological heating. The MLDs for both experiments are deep along 18°N (more than 60 m) because of the strong mixing from the surface winds. The deep mixed layer leads to large heat content and, therefore, a small heating rate of the mixed layer (Wu et al., 2007). The deep mixed layer also leads to a slight effect of the chlorophyll on the mixed layer because the MLD is much deeper than the e-folding depth of the solar radiation. These two effects both weaken the response of the temperature on chlorophyll. Figure 7a also shows that the SST from the O03 experiment is approximately 0.2°C colder than the PS77 experiment beneath the mixed layer. The strong mixing in the upper layer brings the colder water below. Because the small heating rate cannot balance the mixing terms, the SST decreases. This process will be further investigated by the heat budget analysis in the following paragraph.
In boreal summer, warm SST anomalies (Figure 5b) and cold subsurface temperature anomalies (Figure 6b) are induced by the chlorophyll concentration in most parts of the SCS except for the region along the coast of Vietnam. The surface cold anomaly extends from 105°E eastward to approximately 112°E. However, as shown in Figure 2b, a high chlorophyll concentration exists in the region in boreal summer. The cold anomalies also accompany the anomalous offshore current in the surface (Figure 5b) and toward-shore current in the subsurface (Figure 6b). These results suggest that the upwelling may dominate the SST in this region, similar to the processes in the equatorial Pacific.
Lin et al. (2008) investigated the effect of biological heating in the equatorial upwelling region. The chlorophyll concentration gradient enlarges the subsurface temperature gradient, which in turn strengthens the zonal current near the upwelling region, and finally induces an enhanced upwelling. The enhanced upwelling can lead to a cold SST anomaly. In addition, the large chlorophyll concentration cools the upwelling water below the mixed layer, which also leads to a cold SST anomaly. The subsurface cooling and the upward vertical current anomaly can be found around 109°E–111°E in Figure 7b.
To further investigate the differences between the two cases, the heat budget in the two regions was computed. Boxes A (117°E–119°E, 16°N–20°N) and B (109°E–111°E, 10°N–12°N) in Figure 1 denote the two regions where the heat budget was computed. The equation governing the ocean mixed layer temperature can be written as:Table 1.
The shortwave penetration is the main heating term. In boreal winter, the heat budget northwest of Luzon mainly balances the shortwave penetration term (0.19°C/month) and the residual term (−0.29°C/month), which is primarily caused by the mixing forced by the strong northeast monsoon. In contrast, in the region southeast of Vietnam, the residual term (−0.06°C/month) is relatively small because of weak mixing in July. The heat budget is balanced between the shortwave penetration term and the vertical advection, with a value of 0.28°C/month and −0.30°C/month, respectively. Therefore, vertical advection plays an important role in the southeast of Vietnam in boreal summer.
This article investigated the effects of chlorophyll concentration on upper ocean temperature and circulation in the two upwelling regions of the SCS by comparing the results of two experiments using the solar radiation penetration scheme with and without the effects of chlorophyll. In boreal winter, the SST anomalies induced by the chlorophyll concentration are negligible. The strong mixing stirred by strong surface wind brings the cold water beneath into the mixed layer and cancels out the heating due to the effect of chlorophyll. In boreal summer, the high chlorophyll concentration in the upwelling region southeast of Vietnam leads to enhanced upwelling and lower SST. In the non-upwelling regions, surface warming and subsurface cooling are found.
This study suggests that the depth of the mixed layer is important in the response of the surface temperature and current to chlorophyll. The deep mixed layer weakens the responses of the upper ocean on the change of the solar radiation penetration.
The present study is mainly based on numerical experiments that use a stand-alone ocean model. Thus, because of the absence of atmosphere-ocean interaction and biological-physical interaction, the effects may be largely altered, the results of which are to be studied using the atmosphere-ocean coupled model or an Earth system model.
We wish to thank two anonymous reviewers for their valuable comments that have helped to improve our paper. Jinfeng Ma, Haigang Zhan and Yan Du are supported by the CAS projects (KZCX2-YW-Q11–02, KZCX2-YW-BR-04 and LYQY200807) and Hailong Liu and Pengfei Lin are supported by the National Natural Science Foundation of China (40775054, 41023002) and the National Key Program for Developing Basic Sciences (2007CB411806, 2010CB951904).