If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
1. Nature of Pre-existing Cyclonic Circulation and the Scale Analysis
 We thank Wada et al.  (hereinafter referred to as WA) for the comments on the paper entitled “Importance of pre-existing oceanic conditions to upper ocean response induced by Super Typhoon Hai-Tang” by Z.-W. Zheng, C.-R. Ho, and N.-J. Kuo [Zheng et al., 2008] (hereinafter referred to as ZH). After examining the comments, however, we still insist that the basic understanding and the major conclusions given in the paper are correct. In this note, we will provide our deeper understanding and new evidence to further support our conclusions, i.e., the pre-existing of cyclonic circulation or cold core eddies is a major physical mechanism for that typhoon-induced extremely upper ocean cooling responses are defined in certain areas rather than other areas along its trail.
ZH pointed out that there were a series of negative sea surface height anomaly (SSHA) features or cyclonic circulation structures in the tail of Hai-Tang. Their locations coincide with a series of sea surface temperature (SST) cooling centers formed in Hai-Tang's tail. In fact, as shown in Figure 1, these negative SSHA features represent a train of cold core mesoscale eddies, which are generated by the nonlinearity of equatorial Rossby waves. The eddy has a cold core with water temperature 5–10°C lower than surrounding water, which is located at 100–800 m below the sea surface [Parker, 1971; Kiyomitsu, 1975; Cheney and Richardson, 1976]. The horizontal length scale (diameter) of the eddies Le is O(200–300 km). The wavelength (separate distance) is O(600–800 km). In general, the eddies propagate westward with a phase speed of O(0.1 ms−1). It is also possible, sometimes, that the eddies are stationary or even propagate eastward due to the shear effects of mean current [Zheng et al., 1994]. Their time span is O(100–180 d) [Hu et al., 2001]. The azimuthal velocity of the eddy can be calculated with a geostrophic approximation in a cylindrical coordinate system in the form of vθ = gf−1 ∂ς/∂r, where vθ is the azimuth velocity components, g is the gravitational acceleration, f (= 2Ωsinθ) is the Coriolis parameter, and ζ is the sea surface elevation. For cold eddies C1 and C2 shown by ZH (Figure 1) and WA (Figure 1a), we have θ = 23°, and ∂ς/∂r = O(40 cm/100 km). Thus, we obtain an estimate of a maximum azimuthal velocity of the eddies as O(0.7 ms−1). The maximum angular velocity is O(7 × 10−6 s−1), and the period is O(10 d). In addition, ZH and WA indicate that the SST recovery time of the cooling centers Te is O(5 d).
2. Mechanisms for Formation of Extremely Cooling Centers
2.1. Sea Surface Processes
 Before a typhoon comes, heating from the intensive subtropic solar radiation and low sea states keep the skin layer water temperature, which is sensed by satellite SST sensors (infrared or microwave), quite uniform inside eddies and outside surrounding waters. Once the typhoon comes, two sea surface processes are immediately strengthened. 1) Evaporation. High typhoon winds extremely strengthen sea surface evaporation. Latent heat release makes SST decreased along the typhoon tail. However, this process can not obviously enhance the SST difference between eddies and their surrounding waters. 2) High waves. High typhoon winds drive the sea surface wave height higher than 10 m [Yao, 2006]. Recent studies indicate that the sea surface waves play a key role in upper ocean mixing. The depth of wave-induced mixing may reach up to 80–100 m [Qiao et al., 2004]. Numerical study by Wang and Qiao  indicates that wave mixing is major contributor (36%) for SST drop (∼3°C) induced by Typhoon Matsa in the East China Sea.
 In our case, Typhoon Hai-Tang (minimum center pressure 920 mb) is much stronger than Typhoon Matsa (minimum center pressure 950 mb). The typhoon-driven waves by Hai-Tang must be higher than that by Matsa. When it passed over a cold core eddy, wave-induced mixing will bring cold water with temperature 5–10°C lower than surrounding water from up to the depth of 80–100 m to the upper mixed layer. This wave-induced vertical mixing process continuously supplies cold water to the upper mixed layer, which maintains SST inside the cold eddy lower than the surrounding water, and makes the upper mixed layer deeper (see WA, Figure 1c (middle)). The time for wave growth is only a couple of hours. Response time for wave-induced mixing is also a couple of hours, and wave-induced mixing will maintain for a couple of days after typhoon passage.
2.2. Typhoon-Induced Ekman Pumping
 The best-track dataset gives that Hai-Tang's diameter of 15 ms−1 wind speed is 480–900 km (WA). Namely, the horizontal length scale of Ekman pumping effect Lt is O(480–900 km). Compared with the horizontal length scale of these eddies Le, we find immediately that Lt ≫ Le. These unmatched length scales imply that Ekman pumping is not a direct mechanism for formation of neither a single cooling center, and nor a cooling center train. Of course, this does not mean that Ekman pumping has not contribution to the SST drop in the cooling centers. On the contrary, Ekman pumping exerts a large scale forcing on the cooling centers and its surrounding area at the same time. Obviously, temperature of water pumped up from the cold core of an eddy may be 5–10°C lower than water from its surrounding area even from the same depth. Therefore, like wave-induced mixing, Ekman pumping also plays an important role to further enhance the SST drop in the cooling centers.
2.3. Coupling Between the Atmospheric Cyclone and Oceanic Cyclone
 Typhoon is an extremely high wind event, which injects a huge amount of momentum into the upper ocean mixed layer along its passage for a very short duration, say, 1–2 days. Field observations indicated that typhoon drives a rotating ocean current field, like a huge and strong vortex in the upper ocean [Sun et al., 2009]. After typhoon passage, the near-inertial oscillations are generated as typhoon wakes, which may last for 5–7 days [Zheng et al., 2006]. These processes must interact with the cold eddies if they meet by chance. Coupling between the two may force the cold eddies into accelerating or decelerating, which should also have contribution to the change in water temperature in the cooling centers. This hypothesis needs to be confirmed by field measurements or theoretical/numerical modeling.
3. Summary and Discussion
 In this note, we provide our deeper understanding and new evidence on formation of extremely cooling centers appeared in the tail of Typhoon Hai-Tang. These new results support our original conclusions by ZH, i.e., the pre-existing of cyclonic circulation or cold core eddies is a major physical mechanism for that typhoon-induced extremely upper ocean cooling responses are defined in certain areas rather than other areas along its trail. The key points are summarized as follows.
 Low temperature water in the cold cores of cyclonic mesoscale eddies is a material source for formation of an extremely cooling center train in the tail of Typhoon Hai-Tang. The eddies are generated by equatorial Rossby waves, and were pre-existing before the typhoon passage. The scale analysis further excludes the contribution of typhoon-induced Ekman pumping to formation of a single eddy or an eddy train. 2) Typhoon-induced huge sea surface waves should be major contributors to the SST drop and mixed layer enhancement in the cooling centers. 3) Ekman pumping should also have contribution to the SST drop in the cooling centers, although it has not contribution to formation of centers. 4) Coupling between the atmospheric cyclone and oceanic cyclone may be another factor to have influence on water temperature in the cooling centers through modifying the rotation of eddies. This is a topic worthy pursuing in the future.
 We think that this note may serve as a reply to the comments by Wada et al., in particular, to their comment on the impacts of typhoon-induced Ekman pumping. We also think that typhoon-induced upper ocean response is a complex process which involves many factors and mechanisms, such as wind stress and translation speed of typhoon, heat flux, mixed layer thickness, entrainment, upwelling, and advection. To quantify and to separate contributions of different factors and mechanisms constitute challenges to all the investigators in the field. For our research team, a series of numerical experiments based on a three-dimension, high resolution, realistic bathymetry ocean numerical model, Regional Ocean Model System (ROMS), are under way.
 The altimeter products were provided by AVISO with support by CNES. This work was partly supported by the National Science Council of Taiwan through grant NSC 95-2611-M-019-008-MY3 and US NOAA NESDIS ORS Program 05-01-11-000 (for Zheng and Tai).