• Ding, Y., , H. Wang, , and B. Wang, 2005: East Asian monsoon: East Asia. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 95–114.

    • Search Google Scholar
    • Export Citation
  • Haseler, J., 1982: An investigation of the impact at middle and high latitudes of tropical forecast errors. ECMWF Tech. Rep. 31, 42 pp.

  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S-K. Yang, , J. J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Kang, I-S., and Coauthors, 2002: Intercomparison of the climatological variations of Asian summer monsoon precipitation simulated by 10 GCMs. Climate Dyn., 19 , 383395.

    • Search Google Scholar
    • Export Citation
  • Kim, J-K., , Y-J. Bae, , and M-I. Lee, 2001: Global Climate Prediction System (GCPS) operating instruction (in Korean). Korea Meteorological Administration Long-Range Forecast Tech. Rep. 2000, 95 pp.

    • Search Google Scholar
    • Export Citation
  • Lau, N-C., , and B. Wang, 2005: Monsoon–ENSO interactions. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 299–312.

    • Search Google Scholar
    • Export Citation
  • Lee, E-J., , J-G. Jhun, , and C-K. Park, 2005: Remote connection of the northeast Asian summer rainfall variation revealed by a newly defined monsoon index. J. Climate, 18 , 43814393.

    • Search Google Scholar
    • Export Citation
  • Le Treut, H., , and Z. X. Li, 1991: Sensitivity of an atmospheric general circulation model to prescribed SST changes: Feedback effects associated with the simulation of cloud optical properties. Climate Dyn., 5 , 175187.

    • Search Google Scholar
    • Export Citation
  • Nigam, S., , C. Chung, , and E. Deweaver, 2000: ENSO diabatic heating in ECMWF and NCEP–NCAR reanalyses, and NCAR CCM3 simulation. J. Climate, 13 , 31523171.

    • Search Google Scholar
    • Export Citation
  • Numaguti, A., , M. Takahashi, , T. Nakajima, , and A. Sumi, 1995: Development of an atmospheric general circulation model. Reports of a New Program for Creative Basic Research Studies, Studies of Global Environment Change with Special Reference to Asia and Pacific Regions, Rep. I-3, CCSR, Tokyo, Japan, 27 pp.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., , and T. M. Smith, 1994: Improved global sea surface temperature analyses using optimum interpolation. J. Climate, 7 , 929948.

    • Search Google Scholar
    • Export Citation
  • Sperber, K. R., , and T. N. Palmer, 1996: Interannual tropical rainfall variability in general circulation model simulations associated with the atmospheric model intercomparison project. J. Climate, 9 , 27272750.

    • Search Google Scholar
    • Export Citation
  • Sumi, A., , N-C. Lau, , and W-C. Wang, 2005: Present status of Asian monsoon simulation. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 376–385.

    • Search Google Scholar
    • Export Citation
  • Xie, P., , and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Climatological mean precipitation (mm day−1) during JJA from (a) CMAP and (b) the GCM control run, and (c) the difference (mm day−1) between (a) and (b). (d) The difference between the control run and NCEP II reanalysis results for the JJA climatological mean vertically averaged diabatic heating rate. The contour interval is 0.5 K day−1.

  • View in gallery

    (a) Vertical profiles of JJA-mean total diabatic heating (K day−1) based on NCEP II reanalysis (solid line), model control run (dotted line), and ETOT (dashed line) averaged over the western Pacific. Difference between the control run and ETOT for (b) 250-hPa eddy geopotential height (m) and (c) precipitation (contours, in levels of −4, −3, −2, −1, −0.4, 0.4, 1, 2, 3, and 4 mm day−1) and 850-hPa wind (scale arrow at top right of panel; m s−1) during the JJA period. Differences in (b) and (c) that are significant at the 99.9% level are shaded (dark gray is positive; light gray is negative).

  • View in gallery

    JJA precipitation anomalies (mm day−1) averaged over northeastern Asia at 30°–50°N, 100°E–180°, based on the CMAP dataset (solid line), the control run (dotted line), and ETOT (dashed line) from 1979 to 1999.

  • View in gallery

    Regression coefficients of JJA precipitation (contours; mm day−1) and wind at 850 hPa (scale arrows at top right of each panel; m s−1) onto the leading PC time series for northeast Asian precipitation based on the (a) observations, (b) control run, and (c) ETOT. (d) Difference between the control run and ETOT for regression coefficients of 850-hPa moisture transport (scale arrow at top right of panel; m s−1) and moisture convergence (contours; 10−8 s−1) onto the PC time series for northeast Asian precipitation based on their corresponding datasets.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 13 13 1
PDF Downloads 15 15 1

Improving the Northeast Asian Monsoon Simulation: Remote Impact of Tropical Heating Bias Correction

View More View Less
  • 1 APEC Climate Center, Busan, South Korea
© Get Permissions
Full access

Abstract

This study investigates the role of model tropical diabatic heating error on the boreal summer northeast Asian monsoon (NEAM) simulation given by a general circulation model (GCM). A numerical experiment is carried out in which the GCM diabatic heating is adjusted toward more realistic values in the tropics. It is found that the seasonal mean NEAM circulation and rainfall are improved in the GCM. This can be attributed to the reduced positive heating bias in the western Pacific Ocean around 10°–15°N in the model, which in turn leads to better-simulated low-level southerly winds over eastern Asia and more moisture supply to the NEAM region. The GCM’s ability in capturing the year-to-year variation of NEAM rainfall is also markedly improved in the experiment. These results show that the diabatic heating error over the western Pacific can be one reason for poor NEAM simulations in GCMs. The authors also suggest a simple method to reduce model heating biases that can be readily applied to dynamical seasonal prediction systems.

Corresponding author address: Kyong-Hee An, APEC Climate Center (APCC), National Pension Service, Busan Bldg., Yeonsan 2(i)-dong, Yeonje-Gu, Busan 611-705, South Korea. Email: khan@apcc21.net

Abstract

This study investigates the role of model tropical diabatic heating error on the boreal summer northeast Asian monsoon (NEAM) simulation given by a general circulation model (GCM). A numerical experiment is carried out in which the GCM diabatic heating is adjusted toward more realistic values in the tropics. It is found that the seasonal mean NEAM circulation and rainfall are improved in the GCM. This can be attributed to the reduced positive heating bias in the western Pacific Ocean around 10°–15°N in the model, which in turn leads to better-simulated low-level southerly winds over eastern Asia and more moisture supply to the NEAM region. The GCM’s ability in capturing the year-to-year variation of NEAM rainfall is also markedly improved in the experiment. These results show that the diabatic heating error over the western Pacific can be one reason for poor NEAM simulations in GCMs. The authors also suggest a simple method to reduce model heating biases that can be readily applied to dynamical seasonal prediction systems.

Corresponding author address: Kyong-Hee An, APEC Climate Center (APCC), National Pension Service, Busan Bldg., Yeonsan 2(i)-dong, Yeonje-Gu, Busan 611-705, South Korea. Email: khan@apcc21.net

1. Introduction

The Asian summer monsoon is one of the most energetic components of the earth’s climate system. Many operational centers are using dynamical models for seasonal climate predictions. However, simulating the monsoons is still among the most challenging tasks in climate modeling. For the summertime northeast Asian monsoon (NEAM), Kang et al. (2002) showed that many general circulation models (GCM) have difficulties in reproducing the NEAM subtropical rainband (the mei-yu–changma–baiu front). Model NEAM rainfall is likely to be sensitive to the physical parameterizations being used. On the other hand, some modeling works suggest that this problem can be alleviated by increasing the GCM resolution (Sumi et al. 2005).

We also note that the NEAM simulation can be influenced by model biases in remote locations. Haseler (1982) assessed the impact of tropical bias in the context of numerical weather prediction. By nudging the model tropical circulation toward analysis values, large improvement was seen in forecasts over the extratropics. Lee et al. (2005) examined the observed NEAM variability. Increased NEAM rainfall is found to be associated with the anomalous western Pacific Ocean anticyclone that brings more moisture to eastern Asia and increases midlatitude temperature gradient through stronger shortwave radiation. This illustrates that NEAM can be influenced by remote circulation elements in the tropics.

This study investigates the extent to which the NEAM simulation is affected by tropical heating bias in a GCM. For this purpose, a simple numerical experiment is carried out in which a constant is added to the total diabatic heating output in the tropics, which is “felt” by the dynamics of the model. The impact of heating adjustment is then inferred by comparing this integration with a control run. It will be demonstrated that a correct heating field in the tropics can be instrumental to better NEAM simulations in GCMs.

2. Model experiment and datasets

The model being used in this study is the Global Climate Prediction System atmospheric GCM developed by Seoul National University (Kim et al. 2001). It has a spectral resolution of T63 with 21 vertical levels in hybrid coordinate. The simplified Arakawa–Schubert scheme for cumulus convection (Numaguti et al. 1995) and the large-scale condensation scheme of Le Treut and Li (1991) are adopted. For the model experiment, a control run has been carried out during the June–August (JJA) periods of 1979–99, using observed monthly sea surface temperature (SST; Reynolds and Smith 1994). For each year, the GCM was initialized with atmospheric and land surface data on 29 May for that particular year taken from the National Centers for Environmental Prediction (NCEP)–U.S. Department of Energy Second Atmospheric Model Intercomparison Project (AMIP II) reanalysis data (Kanamitsu et al. 2002), and the model was integrated until 31 August. Next, the GCM is integrated in the same setting but with the total diabatic heating adjusted as follows:
i1520-0493-137-2-797-eq1
where Qobs and Qmodel are the climatological total diabatic heating from NCEP II reanalysis and the control run, respectively. Here, Qorg is the sum of heating for cumulus convection, large-scale condensation, shallow convection, vertical diffusion, shortwave radiation, and longwave radiation from the model. The adjusted heating Qnew is passed to the dynamical equations at each time step. Here, F is a cosine function that is equal to unity at 10°N, with the first zeros at 10°S and 30°N; it has zero value outside the 30°N–10°S belt. The α is a tuning parameter. Because of positive feedbacks, the GCM tends to generate extra heating in addition to the prescribed perturbation. We found that the model heating bias is reasonably reduced by setting α ∼ 25%. This experiment for total diabatic heating correction (hereinafter referred to as ETOT) is integrated using the same initial and boundary conditions as in the control run.

The observed precipitation data have been extracted from the Climate Prediction Center Merged Analysis of Precipitation (CMAP; Xie and Arkin 1997). This dataset utilizes in situ observations, infrared and microwave satellite observations, and the NCEP–National Center for Atmospheric Research reanalysis data to estimate pentad precipitation averaged over an area of 2.5° × 2.5° latitude–longitude. The atmospheric variable datasets such as those for the wind and geopotential height are obtained from the NCEP II reanalysis. Those observational datasets are compared with our model results.

Last, note that there could be some uncertainty in the NCEP reanalysis diabatic heating field. Nigam et al. (2000) compared the European Centre for Medium-Range Weather Forecasts and NCEP reanalyses and found a maximum difference in their vertically averaged heating of ∼ 0.3 K day−1 over the equatorial central Pacific Ocean. Taking that as an estimate of the error in the reanalysis products, its value is, however, seen to be much smaller than the typical heating bias in our model (with a maximum of 1.5 K day−1 in the western Pacific). Moreover, the model biases in the diabatic heating and precipitation fields are also consistent with each other (see section 3). These give us credence for using the NCEP reanalysis data for correcting at least the bulk of the heating bias in the model. In the next section, the effects of such heating correction on the summer monsoon simulations will be presented.

3. Results

We first compare the climatological precipitation from the GCM and observations. Figure 1 shows the mean precipitation in the tropics from the control run and CMAP data, as well as their difference, during the JJA season. Insufficient rainfall in the simulation can be found in tropical regions over the Atlantic Ocean, eastern Pacific, western Pacific near the Philippines, South China Sea, and the equatorial eastern Indian Ocean (Fig. 1c). The southern part of both the Indian subcontinent and Indochina also experience less rainfall in the GCM, when compared with observations. Positive model biases are found in the equatorial central to western Indian Ocean and the far western Pacific region around 10°–15°N. The latter, together with the negative bias in more southern locations, is a result of the double intertropical convergence zone (ITCZ) in the GCM (Fig. 1b).

It can also be seen that weaker-than-observed NEAM precipitation is produced in the GCM. Prominent negative rainfall bias covers the coastal regions of southeastern China, Korea, and the southern part of Japan, indicating a weak mei-yu–changma–baiu rainband in the control run. Inspection of the JJA mean circulation further reveals suppressed low-level southerly wind over subtropical eastern Asia and weak westerlies at about 30°N (not shown). These model deficiencies will be partly corrected in the heating experiment ETOT. We have also computed the difference between diabatic heating from the control run and reanalyses. The bias in the vertically averaged heating field (Fig. 1d) and the rainfall bias are consistent with each other. This gives some credence that reanalysis data can be used to adjust the model toward a more realistic basic state.

The heating adjustment experiment was carried out by putting α = 25%. Figure 2a shows the vertical profiles of the total diabatic heating based on reanalysis data, the control experiment, and ETOT averaged over 0°–15°N, 120°–140°E. Consistent with the rainfall bias, the control run gives insufficient heating in this western Pacific region. On the other hand, the ETOT heating is greatly improved. Over this domain, the vertically averaged heating bias from ETOT is −0.03 K day−1, as compared with −0.71 K day−1 from the control experiment. Other regions also see reduced heating bias in ETOT (∼23% reduction in the western Pacific around 20°N, and 53% within 0°–15°N from the eastern Indian Ocean to the South China Sea). The root-mean-square error (RMSE) for the vertically averaged diabatic heating is also reduced in ETOT. (RMSE between reanalysis and ETOT heating, averaged over the Indian Ocean and the western Pacific region of 10°S–30°N, 60°–160°E, is 0.95, as compared with 1.26 based on the control run.) However, some error is still present in the ETOT heating profile: the model heating is more concentrated at 400–500 hPa, whereas the reanalysis data show strong heating at even higher levels. Nevertheless, it will be shown that a better heating field leads to improved NEAM simulation in the GCM.

Figure 2b shows the JJA mean 250-hPa eddy geopotential height for ETOT minus the control run, and Fig. 2c gives the counterpart for precipitation and the 850-hPa wind. Enhanced rainfall is found over a broad region in the western Pacific around 0°–10°N, the South China Sea, Indochina, and the southern Indian subcontinent. Suppressed precipitation is present in the equatorial to southern Indian Ocean and the subtropical western Pacific. As expected, these rainfall changes act to reduce the model biases in the tropics (Fig. 1c).

Of interest is that the basic-state circulation outside the deep tropics is also changed in ETOT. It can be clearly seen that the NEAM precipitation is enhanced in ETOT relative to the control run. One reason for this improvement can be found in the low-level anticyclone southeast of the enhanced rainband. It is consistent with the Rossby wave response to suppressed convection in the western Pacific. Its western wind branch leads to enhanced southerlies over and to the east of coastal eastern Asia. There is thus stronger moisture supply, resulting in better NEAM rainfall in ETOT. Increased low-level westerlies over and to the south of the anomalous rainband can also be seen. The enhanced low-level jet also plays a role in organizing convection in the NEAM region (Ding et al. 2005).

The heating adjustment also has an impact on the extratropical circulation, as evidenced by the difference of eddy geopotential height between ETOT and the control run (Fig. 2b). One noteworthy feature is the couplet of positive and negative height anomalies over the coastal regions in eastern Asia. This circulation structure induced by modified tropical heating leads to a faster upper-level jet stream at about 40°N. This might also lead to better NEAM rainfall simulations. Overall, we can see that adjusted diabatic heating in the tropics affects the circulation in both local and remote locations. In particular, there are significant changes in the NEAM mean flow induced by modified heating in ETOT. The resulting anomalous circulation is conducive to stronger monsoon rainfall over eastern Asia, which partly overcomes the weaker NEAM in the control run.

Besides its impact on the model basic state, modified diabatic heating in ETOT also brings about a substantial improvement on the simulated NEAM variability. Figure 3 gives the time series of JJA rainfall, averaged over the domain of 30°–50°N, 100°E–180°, based on the CMAP dataset, the control run, and ETOT. It is found that the ETOT rainfall over NEAM becomes much better correlated with observations, in comparison with the control run. The correlation coefficient between CMAP data and the control experiment is 0.12, and it is 0.47 between CMAP and ETOT. More sophisticated indices for the northeast Asian summer monsoon rainfall variability—namely, the northeast Asian summer rainfall anomaly (NEARA) index and the northeast Asian summer monsoon index (NEASMI; see Lee et al. 2005)—are also computed. The correlation between the observations and ETOT is 0.313 for NEARA and 0.151 for NEASMI, and the control run gives correlation coefficients of 0.283 for NEARA and −0.026 for NEASMI. Improved NEAM variability is likely due to an improved basic state in the model environment. This notion is consistent with the results of Sperber and Palmer (1996), who showed that GCMs with more realistic climatology can better capture monsoon variability. It is worth mentioning that during the JJA seasons of 1983 and 1998 when the NEAM precipitation was observed to be much stronger than normal, the control run produces very small or even negative rainfall anomalies. These extremely wet seasons are due to the presence of strong El Niño episodes during the preceding winters (Lau and Wang 2005). On the other hand, ETOT gives much better results with large and positive rainfall anomalies in these years.

Last, we compare the dominant circulation patterns associated with NEAM rainfall variability in the model runs with that based on observations. An empirical orthogonal function analysis for the JJA precipitation is performed, for the 1979–99 period. Following Lee et al. (2005), the analysis domain is 20°–50°N, 100°E–180°. Figures 4a–c show the regression coefficients of precipitation and the 850-hPa wind onto the leading principal component (PC) time series for the observations, the control run, and ETOT, respectively. The observed spatial pattern of the leading NEAM mode indicates a reinforced mei-yu–changma–baiu front with rainfall maxima over eastern China, part of the Korean peninsula to southern Japan, as well as an elongated zone extending eastward into the open ocean. Anomalous low-level westerlies over the southern edge of the enhanced rainband around 30°N can also be seen. The anomalous rainfall pattern in the control run is mainly confined over the continental landmass around coastal eastern Asia (Fig. 4b). A relatively weak rainband is found over the ocean. This latter feature and the low-level westerly flow are displaced to the north relative to observations. The above deficiencies in the control run are found to be partially corrected in ETOT. In particular, the ETOT result shows a stronger rainband over the western North Pacific. The extended NEAM rainband, together with anomalous westerlies, are found in more southern locations, when compared with the control experiment. Increased precipitation anomalies over part of Korea and southern Japan can also be discerned in the ETOT simulation.

The low-level moisture transport associated with the leading NEAM mode for the control run and ETOT are also compared. Figure 4d gives the difference between regression of the 850-hPa moisture flux onto the leading PC from ETOT and control run data. The difference of moisture flux convergence is also shown (contours in Fig. 4d). Anomalous convergence of moisture flux is present in the 30°–40°N latitudinal band covering eastern China, Korea, and southern part of Japan. East of about 140°E, there is also anomalous moisture transport from the south. Comparison with the model’s 850-hPa wind patterns (Figs. 4b,c) suggests that the latter is due to the reduced low-level northerly wind over the subtropical western Pacific region in ETOT. This results in more moisture supply over the western North Pacific in the ETOT simulation, when compared with the control run. Overall, these features of enhanced moisture flux convergence are consistent with improved precipitation in the NEAM mode from ETOT.

4. Conclusions

We have investigated the impact of tropical heating on the northeast Asia summer monsoon simulation in a simple GCM experiment. First, the model bias of total diabatic heating is determined from a control run. In ETOT, a heating perturbation is added to the GCM in the tropics, according to the climatological heating bias. Both ETOT and the control run are carried out for the JJA periods in 1979–99 using the same observed SST conditions.

Our results show that, along with reduced GCM tropical heating error, the NEAM summer rainfall simulation is also improved. In the model control run, there is excessive rainfall over the subtropical western Pacific related to the split ITCZ in the model. Associated with this erroneous feature is suppressed southerly flow in subtropical eastern Asia, resulting in less moisture supply from the south to the midlatitudes. This is one reason for the weak NEAM rainfall in the model. In ETOT, these problems are partly alleviated by reducing the diabatic heating over the western North Pacific. The Rossby wave response to suppressed heating leads to increased low-level southerlies near coastal eastern Asia and westerlies in the NEAM region. The stronger moisture flux from the south and enhanced low-level westerly flow in the midlatitudes help to correct the NEAM rainfall in the ETOT simulation.

Besides a more realistic mean circulation, the NEAM interannual variability is also markedly improved in ETOT relative to the control experiment. This result is particularly noteworthy, because our heating adjustment involves the same amount of perturbation during the whole course of model integration. In other words, no information regarding the interannual variation of the tropical circulation is provided in the experiment. The correlation coefficient between ETOT and observed JJA rainfall, averaged over the NEAM region of 30°–50°N, 100°E–180°, is 0.47, as compared with 0.12 from the control run. Further analysis indicates that the dominant mode of NEAM variability is better captured in ETOT. In particular, ETOT gives an enhanced mei-yu–changma–baiu rainband that extends well to the western Pacific region east of Japan, accompanied by a low-level westerly jet feature on the southern edge of the rainband. These circulation features resembles those from the observed NEAM mode. On the other hand, the control run shows a leading NEAM mode rainfall that is very weak over the ocean and is displaced to the north. Consistent with the improved precipitation around 30°–40°N, there is stronger moisture flux convergence in the same latitudinal band associated with the NEAM mode in ETOT. Overall, the pattern of NEAM variability is better captured, and its year-to-year variation is better reproduced, in the ETOT simulation relative to the control run. This is likely due to the improved basic state in the model environment caused by heating adjustment in ETOT.

The results in this study highlight the important role of a correct tropical heating field in the GCMs’ ability to successfully reproduce the NEAM. Our model experiments illustrate clearly that reduced diabatic heating bias in the tropics can bring substantial improvement of the NEAM rainfall simulation. There is currently much effort in simulating the Asian monsoon using very high resolution models (Sumi et al. 2005). Our results indicate that NEAM simulations in GCMs can be sensitive to the model diabatic heating in remote tropical western Pacific locations, in addition to model resolution or convective parameterization.

Last, our findings suggest a technique to improve dynamical seasonal predictions using climate models. They key is to correct the diabatic heating in a GCM by perturbing the heating field according to its climatological bias. Because no a priori information of the tropical circulation is needed, in principle this setting can be used for a seasonal prediction system. This will serve to adjust the model toward a more realistic mean state, which might lead to better skill of climate forecasts. Further studies are needed to test the feasibility of this method and its benefits in dynamical seasonal predictions.

Acknowledgments

We gratefully acknowledge the Seoul National University (SNU) Climate Dynamics Lab (CDL) group for making their model available to us and thank the anonymous reviewers for their helpful comments and suggestions. This work was supported by the Korea Meteorological Administration Research and Development Program.

REFERENCES

  • Ding, Y., , H. Wang, , and B. Wang, 2005: East Asian monsoon: East Asia. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 95–114.

    • Search Google Scholar
    • Export Citation
  • Haseler, J., 1982: An investigation of the impact at middle and high latitudes of tropical forecast errors. ECMWF Tech. Rep. 31, 42 pp.

  • Kanamitsu, M., , W. Ebisuzaki, , J. Woollen, , S-K. Yang, , J. J. Hnilo, , M. Fiorino, , and G. L. Potter, 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Search Google Scholar
    • Export Citation
  • Kang, I-S., and Coauthors, 2002: Intercomparison of the climatological variations of Asian summer monsoon precipitation simulated by 10 GCMs. Climate Dyn., 19 , 383395.

    • Search Google Scholar
    • Export Citation
  • Kim, J-K., , Y-J. Bae, , and M-I. Lee, 2001: Global Climate Prediction System (GCPS) operating instruction (in Korean). Korea Meteorological Administration Long-Range Forecast Tech. Rep. 2000, 95 pp.

    • Search Google Scholar
    • Export Citation
  • Lau, N-C., , and B. Wang, 2005: Monsoon–ENSO interactions. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 299–312.

    • Search Google Scholar
    • Export Citation
  • Lee, E-J., , J-G. Jhun, , and C-K. Park, 2005: Remote connection of the northeast Asian summer rainfall variation revealed by a newly defined monsoon index. J. Climate, 18 , 43814393.

    • Search Google Scholar
    • Export Citation
  • Le Treut, H., , and Z. X. Li, 1991: Sensitivity of an atmospheric general circulation model to prescribed SST changes: Feedback effects associated with the simulation of cloud optical properties. Climate Dyn., 5 , 175187.

    • Search Google Scholar
    • Export Citation
  • Nigam, S., , C. Chung, , and E. Deweaver, 2000: ENSO diabatic heating in ECMWF and NCEP–NCAR reanalyses, and NCAR CCM3 simulation. J. Climate, 13 , 31523171.

    • Search Google Scholar
    • Export Citation
  • Numaguti, A., , M. Takahashi, , T. Nakajima, , and A. Sumi, 1995: Development of an atmospheric general circulation model. Reports of a New Program for Creative Basic Research Studies, Studies of Global Environment Change with Special Reference to Asia and Pacific Regions, Rep. I-3, CCSR, Tokyo, Japan, 27 pp.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., , and T. M. Smith, 1994: Improved global sea surface temperature analyses using optimum interpolation. J. Climate, 7 , 929948.

    • Search Google Scholar
    • Export Citation
  • Sperber, K. R., , and T. N. Palmer, 1996: Interannual tropical rainfall variability in general circulation model simulations associated with the atmospheric model intercomparison project. J. Climate, 9 , 27272750.

    • Search Google Scholar
    • Export Citation
  • Sumi, A., , N-C. Lau, , and W-C. Wang, 2005: Present status of Asian monsoon simulation. The Global Monsoon System: Research and Forecast, WMO/TD-1266, C.-P. Chang, B. Wang, and N.-G. G. Lau, Eds., WMO, 376–385.

    • Search Google Scholar
    • Export Citation
  • Xie, P., , and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78 , 25392558.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Climatological mean precipitation (mm day−1) during JJA from (a) CMAP and (b) the GCM control run, and (c) the difference (mm day−1) between (a) and (b). (d) The difference between the control run and NCEP II reanalysis results for the JJA climatological mean vertically averaged diabatic heating rate. The contour interval is 0.5 K day−1.

Citation: Monthly Weather Review 137, 2; 10.1175/2008MWR2612.1

Fig. 2.
Fig. 2.

(a) Vertical profiles of JJA-mean total diabatic heating (K day−1) based on NCEP II reanalysis (solid line), model control run (dotted line), and ETOT (dashed line) averaged over the western Pacific. Difference between the control run and ETOT for (b) 250-hPa eddy geopotential height (m) and (c) precipitation (contours, in levels of −4, −3, −2, −1, −0.4, 0.4, 1, 2, 3, and 4 mm day−1) and 850-hPa wind (scale arrow at top right of panel; m s−1) during the JJA period. Differences in (b) and (c) that are significant at the 99.9% level are shaded (dark gray is positive; light gray is negative).

Citation: Monthly Weather Review 137, 2; 10.1175/2008MWR2612.1

Fig. 3.
Fig. 3.

JJA precipitation anomalies (mm day−1) averaged over northeastern Asia at 30°–50°N, 100°E–180°, based on the CMAP dataset (solid line), the control run (dotted line), and ETOT (dashed line) from 1979 to 1999.

Citation: Monthly Weather Review 137, 2; 10.1175/2008MWR2612.1

Fig. 4.
Fig. 4.

Regression coefficients of JJA precipitation (contours; mm day−1) and wind at 850 hPa (scale arrows at top right of each panel; m s−1) onto the leading PC time series for northeast Asian precipitation based on the (a) observations, (b) control run, and (c) ETOT. (d) Difference between the control run and ETOT for regression coefficients of 850-hPa moisture transport (scale arrow at top right of panel; m s−1) and moisture convergence (contours; 10−8 s−1) onto the PC time series for northeast Asian precipitation based on their corresponding datasets.

Citation: Monthly Weather Review 137, 2; 10.1175/2008MWR2612.1

Save