Journal of Climate Roles of intraseasonal disturbances and diabatic heating in the East Asian Jet Stream variabilities associated with the East Asian winter monsoon --Manuscript Draft-- Manuscript Number: Full Title: Article Type: Corresponding Author: Corresponding Author's Institution: First Author: Order of Authors: JCLI-D-16-0390 Roles of intraseasonal disturbances and diabatic heating in the East Asian Jet Stream variabilities associated with the East Asian winter monsoon Article Nagio Hirota National Institute for Environmental Studies Tsukuba, Ibaraki JAPAN National Institute for Environmental Studies Nagio Hirota Nagio Hirota Mai Ohta Yousuke Yamashita Masaaki Takahashi Abstract: In this study, we evaluated the relative importance of diabatic heating and intraseasonal disturbances on the variabilities of the East Asian Jet Stream (EAJS) associated with the East Asian winter monsoon (EAWM). First, we selected the strong monsoon years and the weak monsoon years based on the EAWM index of Jhun and Lee [2004], which is highly correlated with the monsoon northerlies between the Eurasian continent and the Pacific. The EAJS is stronger and narrower in strong monsoon years and weaker and wider in weak monsoon years. Model experiments were performed to investigate the atmospheric response to the diabatic heating and the nonlinear feedback from the intraseasonal disturbances. The diabatic heating is closely related to the convective activities. The intraseasonal disturbances include high-frequency components with periods of 3-10 days and lowfrequency components with periods of 10-90 days. The model results indicated that the diabatic heating and the high- and low-frequency components contribute to the stronger and narrower EAJS in the strong monsoon years. Consistently, the anomalous wave activities for both the high- and low-frequency components in the strong monsoon years are divergent in the midlatiudes over the Pacific. Conversely, in the weak monsoon years, the weaker and wider EAJS is primarily driven by the diabatic heating with some contributions from the low-frequency components over northern China. Powered by Editorial Manager and ProduXion Manager from Aries Systems Corporation
Cost Estimation and Agreement Worksheet Click here to access/download Cost Estimation and Agreement Worksheet Cost Estimation and Agreement Worksheet_hirota.pdf
Response to Reviewers Click here to download Response to Reviewers 170227_reply.docx Dear Reviewer 1, Thank you very much for your very constructive comments and suggestions. We deeply appreciate your spending your valuable time for the review. We revised the manuscript based on your comments and suggestions. Please find following replies to your comments. Your comments and suggestions are indicated in italic font, and modifications in the revised manuscript are indicated with red color. 1) Motivation and the methods used in the study In the response to my last review, the authors clarify that the motivation of the study is to investigate "to what extent the forcings by diabatic heating and intra-seasonal disturbances driving the variabilities of EAJS", and approach the topic by using a linear model to compare the jet response to forcings of the diabatic heating and the nonlinear terms associated with the intra-seasonal disturbances. First, I don't think diagnosing the anomalous diabatic heating/intra-seasonal disturbances, or investigating the jet response to those anomalous terms using a linear model will help identify the dominant driver of the jet variabilities. This is because, in the extra-tropics, both the intra-seasonal disturbance and the diabatic heating are strongly driven by the westerly jet. For example, the diabatic heating in the midlatitude, which the authors argued as the dominant driver, is mostly driven 1
by the atmospheric circulations. Thus, the diagnosed anomalies of those fields in the extra-tropics are more likely a consequence of the anomalous jet, instead of a driver. Thank you for your comments. We agree that the diabatic heating/intra-seasonal disturbances are affected by the westerly jet as described: L. 407-409: However, in the real atmosphere, the accelerated (or decelerated) zonal wind of the EAJS influences storm activities, thus the diabatic heating and the nonlinear terms. And we agree that our method of a linear model cannot discuss such influence from the jet to the diabatic heating/intra-seasonal disturbances: L. 405-406: As previously mentioned, in the model used in this study, the diabatic heating and the eddy forcing are prescribed, and both are not influenced by simulated linear dynamics. However, we think the diabatic heating/intra-seasonal disturbances are also affecting the jet variabilities. We are trying to quantify the effects from the diabatic heating/intraseasonal diturbances to the jet variabilities: L. 97-102: As will be described in Section 4, a linear model used in this study isolates and quantifies the impacts of the forcings on the EAJS variabilities. Note that quantifying impacts of diabatic heating and eddy forcing by using a linear model is discussed in many previous studies [Hoskins and Karoly 1981; Branstator 1990; Valdes and Hoskins 1989; Held et al. 1989; Hoskins and Valdes 1990; 2
Watanabe and Kimoto 2000; Watanabe and Jin 2003; Hirota et al. 2005; Mori and Watanabe 2008; Hirota et al. 2012]. Second, using a linear model will not help quantify the contributions of the diabatic heating and the intra-seasonal disturbances either. This is not only because that the diagnosed "forcing" actually in a large portion represents a response, but also that a linear model can not well simulate the nonlinear eddy feedback of high and low frequency eddies. We are sorry for the confusion. As indicated in the equation L. 274: we are not trying to simulate the nonlinear processes. All effects of the nonlinear eddy feedback processes are diagnosed from the reanalysis data and prescribed as the forcings (F) in the model: L. 258-260: This model does not include the eddy processes and the moist processes. Instead, we prescribed (not neglected) the corresponding external forcings diagnosed from the reanalysis data. 2) Robustness of the results The authors added significance test in the revised manuscript. From the new figures, I found that, for the anomalous intra-seasonal disturbances, almost no values in the extra-tropics passed the 3
significance test. Even for the midlatitude diabatic heating, which the authors argued as the dominant driver, there are very few values in the extra-tropics passed the significance test. This also questions the numerical results, in which those anomalous fields are input as anomalous forcings. The results in the manuscript can not well support the authors' main conclusions. Thank you for your comments. The relatively small significance may be because of the small number of the sample. We added a discussion about this issue as follows: L. 225-231: The significance of the anomalous heating over the East China Sea, which will be shown important for the EAJS variabilities in the next section, is relatively small. This is possibly because of the small number (7 years) of the sample for the strong and weak monsoon composites. Although we mainly investigated the 7-year composites to discuss the strong and weak monsoon separately, when we examined a regression map of the diabatic heating with respect to the EAWMI using the 32-year data, the significance of anomalous heating over East China Sea is largely enhanced (not shown). The omitted Figure in the manuscript is shown below: 4
Fig. R1: A regression map of the diabatic heating (K day -1 ) with respect to the EAWMI at 500 hpa. The hatchings indicate the areas exceeding the 90% significance level. As for the intraseasonal disturbances, we think the important anomalies supporting our main conclusion is significant: L. 196-198: In strong monsoon years, a significant divergence anomaly of W appears around northern Japan and over the eastern North Pacific, which is generally consistent with the westerly wind intensification of the EAJS (Fig. 2a). L. 211-213: In the strong monsoon years, a weak divergence anomaly of W appears near Japan (30 N, 140 E), and a significant divergence anomaly is identified over the Central Pacific (20 N, 170 W) (Fig. 6a). L.213-214: Similarly, in weak monsoon years, a significant convergence anomaly of W exists near Japan (Fig. 6b). 5
3) On the numerical results I'm not surprised that in the authors' numerical results, the diabatic heating plays a dominant role. Because such linear model is capable to capture the stationary waves induced by diabatic heating, but most time it cannot capture the eddy feedback, as a linear model cannot well simulate such nonlinear process. Thus, I don't think the numerical results can help quantify the relative contribution of diabatic heating and intra-seasonal disturbance to the EAJS variabilities. Please see the latter part of the response to your comment 1). Second, I don't quite trust the numerical results in the manuscript either. The authors only integrate the numerical model for 30 days and the results shown in the manuscript are merely the time average of the last 10 days. I understand that, for such a simplified model, not many processes can be simulated. However, given that the time scale of the diabatic heating in free troposphere is set to be 30 days in the model, a common sense would be to run the model much longer than this time scale to reach an equilibrium state and obtain stable statistics. Thus, I doubt that a 10 day averaged model result is not reliable to represent the seasonal mean atmospheric circulation as well as the influence of the intra-seasonal disturbance. 6
Thank you for your comment. We examined the time evolutions of the model output. Figure R2 shows the linear responses to the total forcings (diabatic heating + nonlinear terms associated with the high and low frequency disturbances) at day 10, day 20, and day 30 for the strong monsoon years. We describe the results as follows: L. 270-272: The model response reached a near-steady state on the 20th day (not shown). This fast response of the atmosphere is consistent with previous studies [Rodwell and Hoskins 1996; Enomoto et al. 2003; Watanabe and Jin 2003]. Please note the damping time scale and integration time are based on previous studies examining the seasonal mean atmospheric circulations [Rodwell and Hoskins 1996; Enomoto et al. 2003; Watanabe and Jin 2003]. 7
Fig. R2. The steady-state response of the zonal wind (m s -1 ) at 300 hpa to the total forcings for strong monsoon winters at (a) day 10, (b) day 20, and (c) day 30. (d) The time series of the zonal wind averaged over (25--45 N,80--180 E) for strong and weak monsoon winters. Thank you very much again. Sincerely yours, Nagio Hirota 8
Manuscript (non-latex) Click here to download Manuscript (non-latex) 170227_hirota.docx 1 2 Roles of intraseasonal disturbances and diabatic heating in the East Asian Jet Stream variabilities associated with the East Asian winter monsoon 3 4 Nagio Hirota 1, Mai Ohta 2, Yousuke Yamashita 3, and Masaaki Takahashi 12 5 1 National Institute for Environmental Studies. 6 2 Atmosphere and Ocean Research Institute. 7 3 Japan Agency for Marine-Earth Science and Technology. 8 9 10 Corresponding author: Nagio Hirota (nagioh@gmail.com) 16-2 Onogawa, Tsukuba, Ibaraki 305-8506, Japan. 11 12 13 14 1
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 Abstract In this study, we evaluated the relative importance of diabatic heating and intraseasonal disturbances on the variabilities of the East Asian Jet Stream (EAJS) associated with the East Asian winter monsoon (EAWM). First, we selected the strong monsoon years and the weak monsoon years based on the EAWM index of Jhun and Lee [2004], which is highly correlated with the monsoon northerlies between the Eurasian continent and the Pacific. The EAJS is stronger and narrower in strong monsoon years and weaker and wider in weak monsoon years. Model experiments were performed to investigate the atmospheric response to the diabatic heating and the nonlinear feedback from the intraseasonal disturbances. The diabatic heating is closely related to the convective activities. The intraseasonal disturbances include highfrequency components with periods of 3 10 days and low-frequency components with periods of 10 90 days. The model results indicated that the diabatic heating and the high- and lowfrequency components contribute to the stronger and narrower EAJS in the strong monsoon years. Consistently, the anomalous wave activities for both the high- and low-frequency components in the strong monsoon years are divergent in the midlatiudes over the Pacific. Conversely, in the weak monsoon years, the weaker and wider EAJS is primarily driven by the diabatic heating with some contributions from the low-frequency components over northern China. 33 34 35 36 37 2
38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 1. Introduction The East Asian winter monsoon (EAWM) refers to large-scale atmospheric circulations primarily driven by differential heating between the land and sea [e.g., Zhang et al., 1997]. The Siberian high and the Aleutian low are located over the colder Eurasian continent and the warmer Pacific Ocean, respectively. The zonal pressure gradients between these pressure systems balance via geostrophic northwesterlies from Siberia to East Asia around 30 N. In addition, a low pressure center is located around the Maritime Continent, which is associated with the surface northeasterlies to the south of 30 N [Wang and Chen, 2014]. The variabilities in the monsoon northerlies and the associated cold surges exert large social and economic impacts on East Asian countries [e.g., Wang et al., 2009]. The cold air advection of the monsoon northerlies enhances meridional temperature gradients around Japan, causing the East Asian Jet Stream (EAJS) to be stronger and narrower there [Jhun and Lee, 2004]. The jet s strength and structure influence synoptic wave activities in the storm track. In general, synoptic wave activities increase with the wind speed and the baroclinicity of the jet. However, when the jet speeds are greater than 45 m s 1, as they are over East Asia, the synoptic wave activities are observed to be negatively correlated with the wind speeds. Nakamura [1992] and Nakamura and Sampe [2002] explained that this reduction in storm activity with a stronger EAJS occurs because synoptic eddies move more quickly out of the storm track region, allowing them less time to amplify. In addition to the jet strength, Harnik and Chang [2004] suggested that the narrowness of the EAJS is also an important factor reducing synoptic wave activity. Meanwhile, the synoptic activity influences the EAJS via the momentum transport by the associated nonlinear eddy flux [Wettstein and Wallace 2010; Athanasiadis et al. 2011; Li and Wettsein 2012]. 3
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 Previous studies have also discussed the relationship between the EAWM and quasistationary planetary waves (with a period of over ~10 days), which vary more slowly than synoptic waves. Wang et al. [2009] examined interdecadal variations of the EAWM during the period 1957 2002. Both the Siberian high and the Aleutian low were significantly weakened around 1988, possibly due to decreased snow cover over the Eurasian continent. The weakening of pressure systems corresponds to the weakening of planetary waves. They also showed that the anomalous planetary waves propagate southward, resulting in the convergence of wave activities at midlatitudes and divergence at higher latitudes. The nonlinear eddy feedback from the divergent and convergent wave activities generally accelerates and decelerates the westerlies, respectively [Hoskins et al., 1983; Plumb, 1986; Takaya and Nakamura, 2001]. Therefore, Wang et al. [2009] explained that the anomalous planetary waves associated with the weakening EAWM decelerate the EAJS and accelerate the polar jet. The weaker EAJS and the stronger polar jet are typical features of a weak EAWM [Jhun and Lee, 2004]. The variabilities of the Siberian high and the EAWM are also related with the Arctic Oscillation (AO), which is the dominant annular mode of the atmospheric variabilities in the Northern Hemisphere [Gong et al. 2001]. Furthermore, diabatic heating associated with convective activities may play a role in EAWM variabilities. Valdes and Hoskins [1989] investigated atmospheric responses to observed diabatic heating using a multilevel primitive equation model linearized about the observed zonal mean flow. Their results suggested that the formation and maintenance of the subtropical jet is associated primarily with diabatic forcings. Even though they emphasized the importance of diabatic heating at midlatitudes, diabatic heating in the tropics may also be important for EAWM variabilities. Zhang et al. [1997] showed that strong cold-surge events are found more frequently 4
84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 in La Niña years than in El Niño years. They speculated that convective activities over maritime continents are responsible for EAWM variabilities associated with the El Niño Southern Oscillation (ENSO). Meanwhile, Li and Wettsein [2012] showed that tropical diabatic heating is positively correlated with the strength of the EAJS. This study aims to advance our understanding of to what extent the EAJS is driven by the forcings of the diabatic heating and the eddy feedback associated with the intraseasonal disturbances. As for the intraseasonal disturbances, we consider the high-frequency components (e.g., the synoptic waves) and the low-frequency components (e.g., a blocking high or quasistationary planetary waves) separately. As described above, previous studies illustrated how these forcings are different under the strong and weak EAJS conditions. However, it is still not clear to what extent the forcings by diabatic heating and intraseasonal disturbances are driving the EAJS. This is a challenging issue because the observational data include results of the forcings as well as the EAJS influence on the forcings. As will be described in Section 4, a linear model used in this study isolates and quantifies the impacts of the forcings on the EAJS variabilities. Note that quantifying impacts of diabatic heating and eddy forcing by using a linear model is discussed in many previous studies [Hoskins and Karoly 1981; Branstator 1990; Valdes and Hoskins 1989; Held et al. 1989; Hoskins and Valdes 1990; Watanabe and Kimoto 2000; Watanabe and Jin 2003; Hirota et al. 2005; Mori and Watanabe 2008; Hirota et al. 2012]. We expect our results will help understand the EAWM, of which simulations are difficult even in the current state-of-the-art climate models [Wei et al., 2013; Gong et al. 2014, 2015]. 5
105 106 107 The data used in this study are described in Section 2, the analysis results are provided in Section 3, the model experiments are described in Section 4, and the summary and discussion are given in Section 5. 108 109 110 111 112 113 114 115 116 117 2. Data The data used in this study are 6 hourly, daily, and monthly data from the Japanese 25- year Reanalysis (JRA-25; Onogi et al., 2007) for 32 winters from December 1979 to February 2011. Our winter comprises three months: December, January, and February (e.g., for the 1979/80 winter, the average is for December 1979 February 1980). The datasets were provided by the cooperative research project of the JRA-25 long-term reanalysis by the Japan Meteorological Agency (JMA) and the Central Research Institute of the Electric Power Industry (CRIEPI). The climatological mean was calculated as an average from the 1979/80 to 2010/11 winters, and the anomalies were the deviations from the climatological mean. 118 119 120 121 122 123 124 125 126 3. Data Analyses 3.1 EAWM variations The climatological mean fields for the winters of 1979/80 2010/11 are characterized by the Siberian high and the Aleutian low in the sea-level pressure field and the predominant northerly winds over northeastern Asia in the lower troposphere. The EAJS flow at 300 hpa is illustrated in Fig. 1. The maximum wind speed is approximately 65 m s -1 and the core is evident just south of Japan. The position of the jet core is closely related to the enhanced meridional temperature gradient in the lower troposphere via the thermal wind relationship. As the northerly 6
127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 winter monsoon flow over northeastern Asia strengthens, it brings in additional cold air and produces a stronger meridional temperature gradient. Therefore, a stronger monsoon flow leads to a stronger EAJS over the East Asian region [e.g., Jhun and Lee, 2004; Lee et al., 2010]. To measure the strength of the EAWM, the East Asian winter monsoon index (EAWMI) formulated by Jhun and Lee [2004] was adopted. The EAWMI is defined as the difference in the 300 hpa zonal wind anomaly from the climatological mean between the two regions, (110 E 170 E, 27.5 N 37.5 N) and (80 E 140 E, 50 N 60 N). This index is closely related to the monsoon northerlies and has a large negative correction of -0.8 with the surface temperature around Korea and Japan [Jhun and Lee, 2004]. In general, it is difficult to measure the entire EAWM using a single index. However, for this study, the EAWMI is suitable because our intention is to investigate the localized baroclinicity associated with the EAJS over northeast Asia. We removed the linear trend from the EAWMI time series because we are focusing on the interannual variabilities. We selected the seven strongest monsoon years (EAWMI larger than the 0.86 standard deviations) and the seven weakest monsoon years (EAWMI smaller than the - 0.96 standard deviations) from the period of 1979/80 2010/11. The seven strong winter monsoon years were 1980/81, 1983/84, 1994/95, 1999/2000, 2000/01, 2005/06, and 2007/08 whereas the seven weak winter monsoon years were 1986/87, 1988/89, 1989/90, 1991/92, 1992/93, 1997/98, and 2006/07. Moreover, we have confirmed that similar results are obtained even when using other indices for the winter monsoon as will be discussed in Section 5. A bootstrap method was used to test the statistical significance as follows. We randomly selected 7 years from the analyzed 32 years and calculated an anomaly of the 7-year-average. We repeated this calculation for 1000 times and generated a distribution of the 7-year-average anomalies. The anomalies of 7
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 the strong and weak monsoon years are considered significant at level of 90% if they are in the top or bottom 5% of the distribution (See also Fox [2008] and Yamashita et al. [2015]). The composite maps of the zonal wind anomalies at 300 hpa for the strong and weak monsoon years are shown in Fig. 2. Most of the anomalies over East Asia were significant at a 90% level (hatchings in Fig. 2). The anomaly patterns of the meteorological variables in the strong monsoon years are in clear contrast to those in the weak monsoon years. In strong monsoon years, the westerly wind anomaly around Japan corresponds to the strengthening of the EAJS, and the easterly wind anomaly to its north and south indicates that the latitudinal width of the EAJS is narrow compared with the climatological mean (Fig. 2a). In weak monsoon years, the EAJS becomes weak and an easterly wind anomaly appears around Japan (Fig. 2b). The latitudinal width of the EAJS is wider in weak monsoon years than in strong monsoon years. The zonal wind anomalies averaged over (25 45 N, 80 180 E) are 2.0 m s -1 in strong monsoon years and -2.0 m s -1 in the weak monsoon years. A high- (low-) pressure anomaly appears around the Siberian high and a low- (high-) pressure anomaly around the Aleutian low during the strong (weak) monsoon years [cf. Fig. 6 in Jhun and Lee, 2004]. At 850 hpa, a cooler (warmer) temperature anomaly spreads from the northeast of China to Japan, and a warmer (cooler) temperature anomaly covers the surrounding areas during the strong (weak) monsoon years. 166 167 168 169 170 171 3.2 Intraseasonal disturbances and diabatic heating To examine the intraseasonal disturbances, we first filtered the 6 hourly data in the time domain using the Hamming filter [Hino, 1977; Hamming, 1989]. We defined the waves with periods of 3 10 days as high-frequency intraseasonal components, whereas the waves with periods of 10 90 days were defined as low-frequency intraseasonal components. Our spectral 8
172 173 174 175 176 177 178 analysis indicated that they were the dominant periods of the waves associated with the EAWM variabilities (not shown). The interactions between the intraseasonal waves and the seasonal (winter) mean fields were investigated using the wave activity flux W formulated by Takaya and Nakamura [2001]. The divergence (convergence) of the wave activity flux indicates a source (sink) region for wave activity and is proportional to the tendency of the pseudo wave energy. Its expression in the pressure coordinate system is given as 179 180 181 182 183 184 185 186 187 188 189 190 191 + CuM where ψ is the stream function, f is the Coriolis parameter, S is the atmospheric stability parameter, Cu is the prescribed phase velocity, M is the pseudomomentum, and other notations are standard. Overbars ( ) and primes ( ) denote the basic-state quantities and the anomalies, respectively. The subscripts x, y, and p indicate the derivatives. The phase velocity (Cu) is estimated based on one-point correlation maps of 250 hpa stream function like in Takaya and Nakamura (2001). Figure 3 shows the horizontal and vertical fluxes of the 3 10-day time-filtered W at 300 hpa for strong and weak monsoon years and the difference between them. Eastward and vertical upward fluxes are seen over eastern Japan in both strong and weak monsoon years. These areas correspond to the North Pacific storm track region. Comparing the strong and weak years (Fig. 3c), vertical upward fluxes are much weaker over the Pacific during the strong monsoon years. 9
192 193 The weakened wave activities in the strong monsoon years are consistent with the findings of Nakamura [1992], Nakamura et al. [2002], and Harnik and Chang [2004]. 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 The divergence of W (i.e., W) is applied as a measure of the effects of the highfrequency disturbances on the seasonal mean fields. Figure 4 illustrates the W anomalies in the strong and weak monsoon years. In strong monsoon years, a significant divergence anomaly of W appears around northern Japan and over the eastern North Pacific, which is generally consistent with the westerly wind intensification of the EAJS (Fig. 2a). In weak monsoon years, a convergence anomaly of W appears over Japan (Fig. 4b) and over the eastern North Pacific corresponding to the easterly anomaly of EAJS, although their statistical significance is small. Quantitative impacts of the intraseasonal waves on the EAJS will further be examined using the model in the next section. Figure 5 displays the horizontal and vertical fluxes of W associated with the lowfrequency disturbances at 300 hpa for strong monsoon years, weak monsoon years, and the difference between them. Eastward fluxes are seen in both the strong and weak monsoon years over the North Pacific. Comparing the difference in the fluxes between the strong and weak monsoon years (Fig. 5c), the eastward propagation is weaker around Japan in strong monsoon years. The downward anomaly around (35 N, 170 W) and the upward anomaly around (15 N, 160 W) are significant. 210 211 212 213 Figure 6 illustrates the anomalous W associated with the low frequency components in strong and weak monsoon years. In the strong monsoon years, a weak divergence anomaly of W appears near Japan (30 N, 140 E), and a significant divergence anomaly is identified over the Central Pacific (20 N, 170 W) (Fig. 6a). Similarly, in weak monsoon years, a significant 10
214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 convergence anomaly of W exists near Japan (Fig. 6b). The areas of the divergence (convergence) anomaly of W for strong (weak) years are consistent with the westerly (easterly) anomaly near Japan shown in Fig. 2, though the anomalous W around Japan in both years is very small. The distribution of diabatic heating anomalies in strong and weak monsoon years is shown in Fig. 7. The diabatic heating was estimated as a residual of the thermodynamic equation using the 6-hourly reanalysis data (Yanai et al. 1973). In strong monsoon years (Fig. 7a), diabatic cooling is observed in the equatorial Pacific Ocean (10 S 10 N, 160 E 120 W) and diabatic heating occurs to the west of that region (15 S 15 N, 100 E 150 E). Moreover, significant anomalies are also located over the East China Sea (25 N, 135 E) and the central North Pacifc (35 N, 180 E). The diabatic heating in weak monsoon years shows an approximately opposite pattern (Fig. 7b). The significance of the anomalous heating over the East China Sea, which will be shown important for the EAJS variabilities in the next section, is relatively small. This is possibly because of the small number (7 years) of the sample for the strong and weak monsoon composites. Although we mainly investigated the 7-year composites to discuss the strong and weak monsoon separately, when we examined a regression map of the diabatic heating with respect to the EAWMI using the 32-year data, the significance of anomalous heating over East China Sea is largely enhanced (not shown). The horizontal distributions of the diabatic heating are very similar to those of the precipitation and SST anomalies (not shown). Therefore, the diabatic heating is likely to be related to anomalous convective activities [e.g., Hirota and Takahashi, 2012]. The impacts of the diabatic heating on the EAJS will be described in the next section. 11
236 237 4. Experiments using a linear model 238 239 240 241 242 243 244 245 4.1 Model Numerical experiments were performed to examine the relative importance of the diabatic heating and the eddy feedback from the intraseasonal disturbances. The model used in this study is a linear primitive model developed by Hirota and Takahashi [2012]. The governing equations of the model are the primitive equations linearized about the climatological basic state in sigma coordinate system. The equations of anomalous variables ( ) linearized about a basic state ( ) are written as 246 (1) 247 where X is a vector of prognostic variables (vorticity, divergence, and temperature), is the 248 249 250 251 252 253 254 255 256 linear dynamical operator, and F is a forcing vector. The exact form of the equations on the model is given in the appendix. The horizontal resolution of the model is T42 (~2.8 ) and the model has 20 vertical levels. Rayleigh friction, Newtonian cooling, and horizontal diffusion were included. The e- folding time of the friction and cooling was set to 1 day for 1.0 < σ < 0.95, 3 days for 0.95 < σ < 0.9, 5 days for 0.9 < σ < 0.85, 30 days for 0.85 < σ < 0.035, and 1 day for 0.035 < σ < 0.0083 [Branstator, 1990; Watanabe and Jin, 2003]. The strong friction at the boundary layers mimics a turbulent mixing process. The e-folding time of horizontal diffusion was set to 0.5 day for the largest wave number. 12
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 The linear primitive model calculates a linear response (X ) to forcings (F ) of the diabatic heating and the nonlinear terms associated with the intraseasonal disturbances. This model does not include the eddy processes and the moist processes. Instead, we prescribed (not neglected) the corresponding external forcings diagnosed from the reanalysis data. The diabatic heating is primarily due to the condensation heating of the precipitation anomalies caused by convective activities. The nonlinear terms arise primarily from the convergence of the horizontal vorticity flux associated with the high- and low-frequency disturbances. Generally, they correspond to W (Hirota and Takahashi, 2012). This method of calculating the response to the prescribed forcings is advantageous in isolating and quantifying the impacts of the forcings on the EAJS variabilities, but is also a limitation as the influence of the EAJS on the forcings is not simulated. This limitation should be kept in mind especially when discussing regional importance of the forcings. This issue will be discussed in Section 5. The linear model was integrated for 30 days with the prescribed forcings imposed at each time step. The model response reached a near-steady state on the 20th day (not shown). This fast response of the atmosphere is consistent with previous studies [Rodwell and Hoskins 1996; Enomoto et al. 2003; Watanabe and Jin 2003]. Therefore, the responses to the forcings discussed are an average from days 20 to 30. Assuming the steady-state, the response can be written as 274 275. In this linear model, the climatological basic state, which determines the linear dynamical 276 operator, was prescribed using the reanalysis data. This is another advantage of the linear 277 278 model over the general circulation models (GCMs) where the realistic simulation of the climatology is needed even when our interest is only on the anomalies from the climatological 13
279 280 basic state. Although the climatology was prescribed in the linear model, we still need to validate simulated anomalies as described in the next subsection. 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 4.2 Results In order to validate the linear model, we calculated the steady-state responses of the zonal wind to the total external forcings of the diabatic heating with the high- and low-frequency nonlinear terms over the entire globe at all vertical levels. The responses shown in Fig. 8 reproduce the observed anomaly pattern shown in Fig. 2. The easterly-westerly-easterly (westerly-easterly-westerly) responses correspond to the stronger (weaker) and narrower (wider) EAJS in the strong (weak) monsoon years. The zonal wind responses averaged over (25 45 N, 80 180 E) are 2.0 m s -1 in strong monsoon years and -2.2 m s -1 in the weak monsoon years, which are very similar to the values in the reanalysis. Moreover, the northerly and the cold air advection that are described as being important for the EAJS variabilities (see Section 1) are also reproduced in the model (not shown). The similarities between the results from our model and the reanalysis data validate that this model can realistically simulate the linear dynamics of the anomalies associated with prescribed forcings on the climatological basic states. Next, the responses to the diabatic heating and to the nonlinear forcings were calculated separately, as illustrated in Fig. 9, for the strong and weak monsoon years. From Figs. 8a and 9a, we see that, even if the forcing is only provided by the diabatic heating (without the nonlinear effects), the westerly and easterly anomalies of the EAJS are reproduced in the strong and weak monsoon years, respectively. Moreover, the horizontal pattern of the responses around India (0 30 N, 70 E 140 E) and the Pacific Ocean (0 40 N, 160 W 60 W) is qualitatively similar to 14
301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 that of the responses to the total forcing. Therefore, we conclude that diabatic heating plays a primary role in the formation of the anomaly pattern associated with EAWM variabilities. Figures 9c and 9d show the zonal wind anomaly patterns as responses to the nonlinear eddy effects due to the high-frequency components in the strong and weak monsoon years, respectively. In the strong monsoon years, the westerly response over China and Japan reflects the contribution of the eddy feedback from the high-frequency components to the jet acceleration (Fig. 9c). Conversely, in the weak monsoon years, the response to the high-frequency components has almost no contribution to the jet deceleration (Fig. 9d). Figures 9e and 9f show the zonal wind anomaly patterns of the response to eddy effects due to the low-frequency component in the strong and weak monsoon years, respectively. In the strong monsoon years, the results show the contributions of the low-frequency components to the jet acceleration over the central North Pacific. In the weak monsoon years, the response to the low-frequency components shows some small contribution to the jet deceleration over northern China. We further performed experiments calculating the linear responses to forcings over 0-- 360 E confined to low-latitudes (20 S 20 N), mid-latitudes (20 50 N), and high-latitudes (50 80 N) separately. Contributions from each forcing to the zonal wind response averaged over the EAJS (25 45 N, 80 180 E) were summarized in Fig. 10. Although Fig. 10 shows the responses to forcings over 0 360 E, the contributions to the responses associated with the EAJS are mostly resulted from forcings over 60 E 60 W (not shown). In the strong monsoon years, the zonal wind response to the all forcings over the entire globe are 2.0 m s -1 as previously described. The largest contributor is the diabatic heating in the mid-latitudes (20 50 N). The EAJS acceleration of the low-latitude (20 S 20 N) heating is 15
324 325 326 327 328 about 0.3 m s -1, whereas that of high-latitude (50 80 N) heating compensates the accelerating anomalies by -0.2 m s -1. The nonlinear forcings associated with the high-frequency disturbances in the high-latitude and the low-frequency disturbances in the low- and mid-latitudes also show some contributions to the jet acceleration. As for the weak monsoon years, the diabatic heating in the low- and mid-latitudes shows the dominant contributions to the deceleration of the EAJS. 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 5. Discussion and Summary This study focused on understanding the relative importance of diabatic heating and nonlinear eddy feedback effects due to intraseasonal disturbances in the EAJS variabilities associated with the EAWM. As for the intraseasonal disturbances, we considered high-frequency components with periods of 3 10 days (e.g., synoptic waves) and low-frequency components with periods of 10 90 days (e.g., a blocking high or quasi-stationary planetary waves) separately based on the spectral analysis. The EAWM index of Jhun and Lee [2004] was used to select the strong monsoon years and the weak monsoon years. The composite analysis of the JRA-25 reanalysis data indicates that in the strong monsoon years, the northerlies between the Eurasian Continent and the Pacific are stronger and the EAJS is stronger and narrower, as in previous studies. The diabatic heating anomalies associated with the stronger monsoon are positive near the Philippines and the central North Pacific, and negative anomalies are located in the equatorial Pacific and over the East China Sea. These heating anomalies are related to convective activities. The high-frequency components of the intraseasonal disturbances are less active over the Pacific. The anomalies of wave activity flux for both the high- and low-frequency components are divergent around Japan. These 16
346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 anomalies of the large-scale circulations, the intraseasonal wave activities, and the diabatic heating are essentially opposite signs in weak monsoon years. We performed linear model experiments to quantify the roles of the diabatic heating and the intraseasonal disturbances in the EAJS variabilities. The model results indicated that the diabatic heating and the high- and low-frequency components all contribute to the acceleration of the EAJS in the strong monsoon years. Conversely, the deceleration of the EAJS in the weak monsoon years is primarily related to the diabatic heating. The feedback from the low-frequency components contributes to the EAJS deceleration over northern China, whereas the feedback from the high-frequency components does not decelerate the jet. The physical reasons for these asymmetric responses between the strong and weak monsoon years are beyond the scope of this study. That might be related with a nonlinear precipitation response and/or structures of the intraseasonal disturbances, which will be examined in future work. This study used the EAWMI of Jhun and Lee [2004], which is highly correlated with the EAWM circulations and the surface temperature over East Asia. To support robustness of our results, we have done the same analyses using two similar indices: the sea level pressure difference between East Siberia (50 70 N, 100 120 E) and the western North Pacific (30 50 N, 150 170 E), which is more directly related to the monsoon northerlies; 200 hpa zonal wind averaged over (30 35 N, 130 160 E), which is a more direct measure of the EAJS strength as defined by Yang et al. (2002). The correlation coefficients of EAWMI with these indices of the pressure difference and the EAJS strength are 0.84 and 0.77, respectively. The EAJS is stronger (weaker) when these indices are larger (smaller) as consistent with the previous studies. The results of the model experiments based on the composites using these alternative indices are very similar to those using the EAWMI. In particular, the EAJS acceleration is related with the 17
369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 diabatic heating and the eddy feedback due to the high- and low-frequency disturbances, whereas the jet deceleration is primary driven by the diabatic heating (not shown). Next we discuss relationships of EAWMI with AO and ENSO because they are considered important factors for the EAWM variabilities (Zhang et al. 1997; Gong et al. 2001; Yang et al. 2002). An AO index (AOI) used here is defined as the principal components of the leading Emperical Orthogonal Function mode of 1000 hpa geopotential height poleward of 20 N (http://www.cpc.ncep.noaa.gov/products/precip/cwlink/daily_ao_index/history/method.shtml). As we have done using EAWMI, we examined composites of years with the 7 largest AOI (AO+) and the 7 smallest AOI (AO-). The composite maps of zonal wind anomalies in Fig. 11a, b show that the EAJS is weaker in the AO+ years and stronger in the AO- years. The model responses to diabatic heating (Fig. 11e, f) indicate that these zonal wind anomalies are mainly related to the diabatic heating. Note that the model simulated responses to all forcings associated with AO (Fig. 11c, d) show some disagreements with the composite anomalies (Fig. 11a, b). In particular, the EAJS deceleration in the AO+ years is not reproduced over the Pacific, and the EAJS acceleration in the AO- years is slightly shifted poleward over the eastern Pacifc. These anomalies cannot be discussed using the responses of the linear model. Further exploration of the AO impacts on the EAJS is beyond the scope of this paper. As for ENSO, we used Southern Oscillation Index (SOI, http://www.cpc.noaa.gov/products/analysis_monitoring/ensocycle/soi.shtml) defined as pressure differences between Tahiti and Darwin. The composite maps of years with the 7 largest SOI (SOI+) and the 7 smallest SOI (SOI-) are shown in Fig. 12a, b. In the SOI+ (SOI-) years, the EAJS is significantly weaken (strengthen) over the central Pacific, and smaller anomalies with the opposite signs are found over East Asia (25 40 N, 80 150 E). The diabatic heating seems to 18
392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 be responsible for the anomalies over East Asia (Fig. 12e, f). The anomalies over the central Pacific is not captured in the linear model (Fig. 12c, d). The correlation coefficients of EAWMI with AOI and SOI are -0.34 (explains 12% of the variance) and 0.38 (14%), respectively. These values are significant, but each explains only a small part of the EAWM variabilities, indicating that other factors (e.g. SSTs in the extratropics, snow cover over the Eurasian continent, and so on) are also important. Comparing the horizontal pattern of the zonal wind anomalies, the anomalies of EAWMI (Fig. 2) and SOI (Fig. 12a, b) are consistent over East Asia, but their signs are opposite over the central Pacific. On the other hand, the westerly anomalies in the AO- composites (Fig. 11b) consistent with those in strong monsoon composites (Fig. 2a) are significant over the central Pacific but not over East Asia. These results suggest AO and ENSO are affecting different part of the EAJS variabilities. Note that the correlation between AOI and SOI is not significant, suggesting that AO and ENSO are affecting EAWM independently. As previously mentioned, in the model used in this study, the diabatic heating and the eddy forcing are prescribed, and both are not influenced by simulated linear dynamics. This helped to isolate and quantify the impacts of the forcings on the EAJS variability. However, in the real atmosphere, the accelerated (or decelerated) zonal wind of the EAJS influences storm activities, thus the diabatic heating and the nonlinear terms. The midlatutde diabatic heating having the largest contribution to the EAJS acceleration (Fig. 10) is partly resulted from precipitation associated with the modulated EAJS forced from the low- or high-latitudes. Such indirect remote interactions are not simulated in the linear model. Instead, the results of those interactions diagnosed using the reanalysis data are prescribed (not neglected) in the linear 19
414 415 model. To investigate these two-way interactions, GCMs with the nonlinear dynamics and the moist processes should be used in future work. 416 417 418 419 420 421 APPENDIX The governing equations of the model used in this study are the primitive equations linearized about the climatological basic state in sigma coordinate system (longitude λ, latitude φ, and sigma σ (= pressure/surface pressure)). The primitive equations of anomalous variables ( ) linearized about a basic state ( ) are 20
422 423 424 425 where ξ is vorticity, D is divergence, T is temperature, V = u cos φ, u is horizonatal wind, σ is vertical velocity, Tp is temperature deviation from reference temperature (= 300K), Tv is virtual 426 temperature, μ = sin φ, π = log(surface pressure), Q is heating rate, are nonlinear 427 428 429 430 forcings, R is radius of the earth, Ω is angular speed of the earth's rotation, Rd is gas constant of dry air, Cp is the specific heat of dry air, and κ=rd/cp. The linear damping and h 4 horizontal diffusion were included with the time scale of τ and the coefficients of ν, respectively. This model calculates a linear response for the prescribed forcings of Q' and. 21
431 432 433 434 435 436 437 438 Acknowledgments The authors appreciate three anonymous reviewers for their helpful comments and suggestions to improve the manuscript. This study was supported by the Program for Risk Information on Climate Change and KAKENHI (15H02132, JP16K16186) of the Ministry of Education, Culture, Sports, Science, and Technology, Japan, and by the Environment Research and Technology Development Fund (2-1503) of the Ministry of the Environment, Japan. The Grid Analysis and Display System was used to plot the figures. 439 22
440 References 441 442 Athanasiadis, P. J., J. M. Wallace, and J. J. Wettstein (2010), Patterns of Wintertime Jet Stream Variability and Their Relation to the Storm Tracks. J. Atmos. Sci, 67 (5), 1361 1381. 443 444 Barriopedro, D., Garcia--Herrera R., Lupo A., Hernandez E. (2006), A Climatology of Northern Hemisphere Blocking, J. Climate., 19, 1042--1063. 445 446 Branstator, G. (1990), Low-frequency patterns induced by stationary waves, J. Atmos. Sci., 47, 629--648. 447 448 449 Chang, C.-P, and K.-M. Lau (1982), Short-term planetary-scale interactions over the Tropics and midlatitudes during northern winter. Part I: Contrasts between active and inactive periods, Mon. Wea. Rev., 110, 933--946. 450 451 452 Chang, E. K. M. (2001), GCM and observational diagnoses of the seasonal and interannual variations of the Pacific storm track during the cool season, J. Atmos. Sci., 58, 1784-- 1800. 453 454 Enomoto, T., and B. J. Hoskins and Y. Matsuda (2003), The formation mechanism of the Bonin high in August, Quart J. Roy. Meteor. Soc., 587, 157--178. 455 456 Fox, J. (2008), Applied Regression Analysis and Generalized Linear Models, 2nd edition. Sage Publications, 688 pp. 457 458 Gong, D.-Y., S.-W.Wang, and J.-H. Zhu (2001), East Asian winter monsoon and Arctic oscillation. Geophys. Res. Lett., 28 (10), 2073 2076. 23
459 460 Gong H., et al. (2014), The climatology and interannual variability of the East Asian winter monsoon in CMIP5 models, Journal of Climate, 27, 1659-1678 461 462 463 Gong H., et al. (2015), Diverse influences of ENSO on the East Asian-western Pacific winter climate tied to different ENSO properties in CMIP5 models. Journal of Climate, 28, 2187-2202 464 Hamming, R. W. (1989), Digital Filters, 3rd ed., p.p. 284, Dover, Mineola, N. Y. 465 466 467 Harnik, N., and E. K. M. Chang (2004), The effects of variations in jet width on the growth of baroclinic waves: Implications for midwinter Pacific storm track variability, J. Atmos. Sci, 61, 23--40. 468 469 Held, I. M., S.W. Lyons, and S. Nigam (1989), Transients and the extratropical response to El Nino. J. Atmos. Sci., 46 (1), 163 174. 470 471 Hino, M. (1977), Spectral Analysis (in Japanese), Asakura Publishing Company, 300pp, 162 Tokyo, Japan. 472 473 Hirota, N., M. Takahashi, N. Sato, and M. Kimoto (2005) Recent climate trends in the East Asia during the Baiu season of 1979-2003. SOLA, 1, 137 140. 474 475 Hirota, N., and M. Takahashi (2012), A tripolar pattern as an internal mode of the East Asian summer monsoon. Climate Dynamics, 39, 2219--2238. 476 477 Hoskins, B. J., and P. J. Valdes (1990) On the existence of storm-tracks. J. Atmos. Sci., 47 (15), 1854 1864. 24
478 479 Hoskins, B.J. and D.J. Karoly (1981) The steady response of a spherical atmosphere to thermal and orographic forcing. J. Atmos. Sci., 38, 1179 1196. 480 481 Hoskins, B. J., I. N. James, and G. H. White (1983), The shape, propagation and mean-flow interaction of large-scale weather systems, J. Atmos. Sci., 40, 1595--1612. 482 483 Jhun, J. G., and E. J. Lee (2004), A New East Asian winter monsoon index and associated characteristics of the winter monsoon, J. Climate., 15, 711--726. 484 485 Lee, Y.-Y., G.-H. Lim, and J.-S. Kug (2010), Influence of the East Asian winter monsoon on the storm track activity over the North Pacific, J. Geophys. Res., 115, D09102. 486 487 Li, C., and J. J.Wettstein (2012), Thermally Driven and Eddy-Driven Jet Variability in Reanalysis. J. Climate, 25 (5), 1587 1596. 488 489 Mori, M. and M. Watanabe (2008) The Growth and Triggering Mechanisms of the PNA: A MJO-PNA Coherence. J. Meteor. Soc. Japan, 86, 213 236. 490 491 Nakamura, H. (1992), Midwinter suppression of baroclinic wave activity during the life cycles of low-frequency circulation anomalies, J. Atmos. Sci., 47, 1100--1116. 492 493 494 Nakamura H., and T. Sampe (2002), Trapping of synoptic-scale disturbances into the north-acific subtropical jet core in midwinter. Geophysical Research Letters, 29, doi:10.1029/2002gl015535. 495 496 Nakamura H., et al. (2002), Interannual and decadal modulations recently observed in the pacific storm track activity and east Asian winter monsoon. Journal of Climate, 15, 1855-1874. 497 Onogi, K., et al. (2007), The JRA-25 Reanalysis., J. Meteor. Soc. Japan., 85, 369--432. 25
498 499 500 Plumb, R. A. (1986), Three-dimensional propagation of transient quasi-geostrophic eddies and its relationship with the eddy forcing of the time-mean flow, J. Atmos. Sci., 43, 1657-- 1678. 501 502 Rodwell, M.J. and B.J. Hoskins (1996) Monsoons and the dynamics of deserts. Quart. J. Roy. Meteor. Soc., 122, 1385 1404. 503 504 505 Takaya, K. and H. Nakamura (2005), Geographical dependence of upper-level blocking formation associated with intraseasonal amplification of the Siberian high, J. Atmos. Sci., 62, 4441--4449. 506 507 Valdes, P.J., and B. J. Hoskins (1989), Linear stationary wave simulations of the time mean climatological flow, J. Atmos. Sci., 46, 2509--2527. 508 509 Wallace, J. M., and D.S. Gutzler (1981), Teleconnections in the geopotential height field during the Northern Hemisphere winter, Mon Wea Rev., 109, 784--812. 510 511 Wang, L. and W. Chen (2014), An intensity index for the East Asian winter monsoon, Journal of Climate, 27, 2361-2374 512 513 514 Wang, L., R.-H. Huang, L. Gu, W. Chen, and L. Kang (2009), Interdecadal variations of the East Asian Winter Monsoon and their association with quasi-stationary planetary wave activity, J. Climate., 22, 4860--4872. 515 516 Wang, B., R. Wu, and X. Fu (2000), Pacific--East Asian teleconnection: How does ENSO affect East Asain climate? J. Clim., 13, 1517--1536. 26
517 518 Watanabe, M. and M. Kimoto (2000) Atmosphere-ocean thermal coupling in the North Atlantic: a positive feedback. Quart. J. Roy. Meteor. Soc., 126, 3343 3369. 519 520 Watanabe, M and F. Jin (2003), Mosit Linear Baroclinic Model: Coupled Dynamical Convective Response to El Nino, J. Clim., 16, 1121--1139. 521 522 523 Wei, K., Xu, T., Du, Z., Gong, H., and Xie, B. (2014), How well do the current state-of-the-art CMIP5 models characterise the climatology of the East Asian winter monsoon?. Climate dynamics, 43(5-6), 1241--1255. 524 525 526 Wettstein, J. J., and J. M. Wallace (2010), Observed patterns of month to month storm-track variability and their relationship to the background flow. J. Atmos. Sci., 67 (5), 1420 1437. 527 528 529 530 Yamashita, Y., H. Akiyoshi, T. G. Shepherd, and M. Takahashi (2015), The combined influences of westerly phase of the Quasi-Biennial Oscillation and 11-year solar maximum conditions on the Northern Hemisphere extratropical winter circulation, J. Meteor. Soc. Japan, 93(6), 629 644. 531 532 Yanai, M., S. Esbensen, and J.-H. Chu (1973), Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. J. Atmos. Sci., 30, 611 627. 533 534 Yang S., et al. (2002), Variations of the East Asian jet stream and Asian-Pacific-American winter climate anomalies. Journal of Climate, 15, 306-325. 27
535 536 537 Zhang, Y., Sperber, K. R., and Boyle, J. S. (1997), Climatology and interannual variation of the East Asian winter monsoon: Results from the 1979-95 NCEP/NCAR reanalysis. Monthly Weather Review, 125(10), 2605--2619. 538 28
539 List of Figures 540 Figure 1. Winter climatological zonal wind speed (m s -1 ) at 300 hpa. 541 542 543 Figure 2. Composite maps of the zonal wind anomaly (m s -1 ) at 300 hpa for (a) strong monsoon years and (b) weak monsoon years. The hatchings indicate the areas exceeding the 90% significance level. 544 545 546 547 548 Figure 3. Three-dimensional wave activity fluxes, W, associated with the high-frequency disturbances for (a) strong monsoon years, (b) weak monsoon years, and (c) their difference (strong minus weak). The vectors indicate the horizontal components of W and the shading indicates the vertical component of W at 300 hpa with units of m 2 s -2. The hatchings in (c) indicate the areas exceeding the 90% significance level. 549 550 551 Figure 4. W anomalies associated with the high-frequency disturbances for (a) strong and (b) weak monsoon years at 300 hpa in units of 10-6 m s -2. The hatchings indicate the areas exceeding the 90% significance level. 552 Figure 5. Same as Fig. 3, but for the low-frequency disturbances. 553 Figure 6. Same as Fig. 4, but for the low-frequency disturbances. 554 555 Figure 7. The diabatic heating anomaly for (a) strong and (b) weak monsoon winters at 500 hpa with units of 10-5 K s -1. The hatchings indicate the areas exceeding the 90% significance level. 556 557 Figure 8. The steady-state response of the zonal wind (m s -1 ) at 300 hpa to the total forcings for (a) strong and (b) weak monsoon winters. 29
558 559 560 561 Figure 9. The steady-state response of the zonal wind (m s -1 ) at 300 hpa to the forcing separated by (a, b) diabatic heating, (c, d) nonlinear effects due to the high-frequency disturbances, and (e, f) nonlinear effects due to the low-frequency disturbances for strong monsoon winters (left) and weak monsoon winters (right). 562 563 564 565 566 567 568 Figure 10. The response of 300 hpa zonal wind (m s -1 ) over (25--45 N, 80--180 E) to the total forcings (All), diabatic heating (Q ), nonlinear effects due to the high-frequency disturbances (Nh ), and nonlinear effects due to the low-frequency disturbances (Nl ) for the strong (left) and weak (right) monsoon years. Grey bars, red bars, orange bars, and blue bars show the response to the forcings over the entire globe (90 S--90 N), low-latitudes (20 S--20 N), mid-latitudes (20 -- 50 N), high-latitudes (50--80 N), respectively. 569 570 571 572 Figure 11. Composite maps of the zonal wind anomaly (m s -1 ) at 300 hpa for (a) the 7 AO+ years and (b) the 7 AO- years. The steady-state response of 300 hpa zonal wind (m s -1 ) to (c, d) the total forcings, (e, f) diabatic heating, and (g, h) nonlinear effects for (c, e, g) the 7 AO+ years and (d, f, h) the 7 AO- years. 573 Figure 12. Same as Fig. 11, but for SOI. 574 30
575 576 Figure 1. Winter climatological zonal wind speed (m s -1 ) at 300 hpa. 577 31
578 579 32
580 581 582 Figure 2. Composite maps of the zonal wind anomaly (m s -1 ) at 300 hpa for (a) strong monsoon years and (b) weak monsoon years. The hatchings indicate the areas exceeding the 90% significance level. 583 33
584 585 586 587 Figure 3. Three-dimensional wave activity fluxes, W, associated with the high-frequency disturbances for (a) strong monsoon years, (b) weak monsoon years, and (c) their difference (strong minus weak). The vectors indicate the horizontal components of W and the shading 34
588 589 indicates the vertical component of W at 300 hpa with units of m 2 s -2. The hatchings in (c) indicate the areas exceeding the 90% significance level. 35
590 591 592 593 Figure 4. W anomalies associated with the high-frequency disturbances for (a) strong and (b) weak monsoon years at 300 hpa in units of 10-6 m s -2. The hatchings indicate the areas exceeding the 90% significance level. 594 36
595 596 Figure 5. Same as Fig. 3, but for the low-frequency disturbances. 37
597 598 Figure 6. Same as Fig. 4, but for the low-frequency disturbances. 599 38
600 601 602 Figure 7. The diabatic heating anomaly for (a) strong and (b) weak monsoon winters at 500 hpa with units of 10-5 K s -1. The hatchings indicate the areas exceeding the 90% significance level. 603 39
604 605 606 Figure 8. The steady-state response of the zonal wind (m s -1 ) at 300 hpa to the total forcings for (a) strong and (b) weak monsoon winters. 607 40
608 609 610 611 612 Figure 9. The steady-state response of the zonal wind (m s -1 ) at 300 hpa to the forcing separated by (a, b) diabatic heating, (c, d) nonlinear effects due to the high-frequency disturbances, and (e, f) nonlinear effects due to the low-frequency disturbances for strong monsoon winters (left) and weak monsoon winters (right). 613 41
614 615 616 617 618 619 620 Figure 10. The response of 300 hpa zonal wind (m s -1 ) over (25--45 N, 80--180 E) to the total forcings (All), diabatic heating (Q ), nonlinear effects due to the high-frequency disturbances (Nh ), and nonlinear effects due to the low-frequency disturbances (Nl ) for the strong (left) and weak (right) monsoon years. Grey bars, red bars, orange bars, and blue bars show the response to the forcings over the entire globe (90 S--90 N), low-latitudes (20 S--20 N), mid-latitudes (20 -- 50 N), high-latitudes (50--80 N), respectively. 621 42
622 623 624 625 626 Figure 11. Composite maps of the zonal wind anomaly (m s -1 ) at 300 hpa for (a) the 7 AO+ years and (b) the 7 AO- years. The steady-state response of 300 hpa zonal wind (m s -1 ) to (c, d) the total forcings, (e, f) diabatic heating, and (g, h) nonlinear effects for (c, e, g) the 7 AO+ years and (d, f, h) the 7 AO- years. 43
627 628 Figure 12. Same as Fig. 11, but for SOI. 44