1. Introduction
A reanalysis dataset is a continually updating gridded dataset that represents the state of the earth’s atmosphere; it incorporates observations and numerical weather prediction (NWP) model outputs. When all available observation data and short-range forecasts are utilized, a frozen and combined data assimilation/forecast model produces an optimal and uniform quality analysis over a long period of time. Reanalysis projects have been conducted in operational centers such as the National Centers for Environmental Prediction (NCEP) (Kalnay et al. 1996; Kistler et al. 2001; Kanamitsu et al. 2002b), the European Centre for Medium-Range Weather Forecasts (ECMWF) (Uppala et al. 2005), and the Japan Meteorological Agency (Onogi et al. 2007). Reanalysis datasets have been extensively used by the weather and climate research communities.
Specific model variables such as the radiative flux have been utilized to obtain useful information on physical mechanisms in the atmosphere (Walsh et al. 2009). These long-term data, along with in situ observations, are particularly useful in expanding our knowledge of changes in atmospheric circulations and precipitation activities due to global warming (Cai and Kalnay 2005; Wang and Mehta 2008). However, spurious interdecadal changes caused by observational discontinuities have been found in reanalysis data (Kistler et al. 2001; Screen and Simmonds 2011). For example, the use of satellite data and its continuous refinement hinder the examination of climate variability and change. Screen and Simmonds (2011) explicitly documented a discontinuity in the 40-yr ECMWF Re-Analysis (ERA-40) that leads to significantly exaggerated warming in the Arctic. Such exaggerated warming is specifically due to the discontinuity of satellite data refinement in 1997. Thorne and Vose (2010) raised this critical issue and argued that a substantial rethinking of the current strategy for producing reanalysis products is required.
An alternative approach for overcoming the aforementioned drawbacks in long-term datasets is to use dynamical global downscaling. Developed by Yoshimura and Kanamitsu (2008, hereafter YK08), dynamical global downscaling is a global version of a spectral nudging technique used in a regional spectral model. YK08 proposed the global downscaling method and nudging configuration and performed downscaling experiments during 2001 using a T248 (about 50-km resolution) global model with the NCEP–Department of Energy (DOE) reanalysis (RA2; Kanamitsu et al. 2002b) as the large-scale forcing. An evaluation with high-resolution observations showed that the monthly averaged global temperature field and daily variability in the downscaled precipitation were significantly improved.
As an extension of the YK08 method, we generate new 31-yr global analysis data on the order of approximately 100 km. The data are forced by RA2 using dynamical global downscaling. One of the main objectives of this study is to demonstrate that dynamical global downscaling is capable of reproducing long-term global reanalysis data without expensive high-resolution data assimilation. Another goal of this work is to investigate the sensitivity of the physics in the downscaling. Brief descriptions of both the downscaling method and the employed model are given in section 2. An evaluation of nudging variables and the precipitation against global observations is presented in section 3. For a detailed examination of how downscaled fields fit to observations, a 1-month analysis (July 2001) is conducted. Although the results from other months are not shown, the capability of the downscaling method is similar throughout the year. A summary of the obtained results and conclusions are given in section 4.
2. Model and dynamical downscaling procedure
a. Global model
The model used in this study is a version of the NCEP Global Spectral Model (GSM; Kanamitsu et al. 2002a); its evolution has followed a path that is rather separate from the ongoing development of operational NCEP medium-range forecast models. Since 2000, the model has been used to modify message passing interface (MPI) computing methods and implement realistic physical processes at Yonsei University. The model has great flexibility with multiple platforms in either thread mode or message parallel mode (Park et al. 2008). It also has multiple options with regards to physics parameterization. The physics of the model include longwave and shortwave radiation, cloud–radiation interactions, planetary boundary layer (PBL) processes, shallow convection, gravity wave drag, simple hydrology, and vertical and horizontal diffusions. Detailed descriptions of the physics package used in this study are given in Table 1.
Physics package for the reanalysis and downscaled data.
b. Dynamical global downscaling
c. Integration
The downscaling experiment is performed with the T126 (about 100 km) and 28 sigma levels as the downscaling model, along with 6-hourly snapshots of prognostic variables from RA2 (resolution of T62, about 200 km) and 28 levels as the lateral forcing. The simulations were forced by the sea surface temperatures (SSTs) obtained from observations with a resolution of 1° (Reynolds and Smith 1994) and National Center for Atmospheric Research (NCAR) ice data.
Downscaling is formed in three streams: 1979–89, 1987–99, and 1998–2011. To avoid a spinup of soil moisture, the three streams overlap over 2-yr periods, 1987–88 and 1997–98. Therefore, the downscaled data cover 31 years from January 1980 to February 2011. For convenience, RA2 and the downscaled datasets in this study are referred to as RA62 and DA126, respectively.
3. Results
To examine the accuracy of primitive variables such as the wind, temperature, and relative humidity in the downscaled data, each downscaled variable and the corresponding radiosonde observation data (RAOBs) are compared for July 2001. The RAOBs from the University of Wyoming Web site (UWYO; http://weather.uwyo.edu/) are processed via simple quality control. For better verification, the four grid points nearest to the observations in the analysis data are interpolated to the location of the radiosonde stations. The analysis is performed over two representative regions: East Asia (20°–50°N, 100°~150°E) and North America (20°–50°N, 120°~60°W).
As shown in Fig. 1, the bias in the temperature magnitude for the downscaled dataset (designated by dashed lines) is quite similar to that in the reanalysis over both regions. In addition, the wind field over 900 hPa is significantly improved. On the other hand, the humidity profiles exhibit severe dry biases with a maximum around 850 hPa, and the ranges for the downscaled data are greater than those for the reanalysis. Drying within the entire troposphere is related to the warm bias of DA126. However, the temperature bias is relatively small, which may be due to the fact that the model physics have a more profound effect on the moisture fields than the temperature fields (Hong and Pan 1996). Considering that the humidity is biased wet for the model (Wang et al. 2002), DA126 can lessen the systematic wet bias in the reanalysis data. This result is confirmed because the large-scale variables in the reanalysis were successfully retained in the global downscaling.
The quality of the downscaled precipitation is also examined. The Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) data (Xie and Arkin 1997), with a 2.0° resolution, and the Global Precipitation Climatology Project (GPCP), which provides 1° daily precipitation estimates over both land and ocean (Huffman et al. 2001), are used to verify the global distribution and zonal mean amounts of precipitation. For a quantitative evaluation of the land precipitation, the CPC unified gauge analysis (Chen et al. 2008), which covers the entire globe at a daily interval, is used.
The annual variations in the analyzed and observed precipitation over the lands of East Asia and North America are shown in Fig. 2. In contrast to RA62, which overestimated the precipitation over all of the years, the precipitation amounts of DA126 closely represent the ranges of the observed precipitation. The domain-averaged amount of precipitation and the root-mean-square error (RMSE) also reveal an overall improvement (not shown). The temporal correlations for RA62 and DA126 against the CPC are 0.56 and 0.43 over East Asia, which are slightly lower than the corresponding values for the reanalysis. The temporal correlations for RA62 and DA126 over North America are 0.37 and 0.46, respectively. It should be noted that the variabilities in the reanalysis and the downscaled precipitation are highly correlated even though the two datasets were generated by different physics. Correlation coefficients between RA62 and DA126 are 0.87 and 0.84 for East Asia and North America, respectively. This suggests that the annual variability in the modeled precipitation may be mostly due to large-scale fields.
Another experiment is conducted to investigate the cause of the improvement in the downscaled precipitation. This experiment, referred to as DA62, is downscaled using a physics package and nudging configuration that are identical to those used for DA126, but with different resolutions (T62, resolution of about 200 km). The experiment is integrated for 2 months from June 2001 to give a 1-month spinup time, and the analysis is conducted only for July 2001. The 1-month averaged zonal mean precipitation data from all downscaled datasets are obtained and compared with the observed precipitation (Fig. 3). The two downscaled datasets are found to be closer to the observations than the reanalysis data. The intertropical convergence zone (ITCZ) intensity, which is exaggerated in the reanalysis precipitation, is reduced in all downscaled datasets. On the other hand, the DA126 results are similar to those of the additional experiment (DA62). It should be noted that the physics in RA62 are different from those of the downscaled data (see Table 1). This demonstrates that the precipitation is more sensitive to the model physics than the model resolution.
The observed global distribution of the analyzed precipitation in July 2001 and the corresponding observations are shown in Fig. 4. The reanalysis precipitation is characterized by wetness in most continental regions, with the exception of some regions of central Africa and the northern part of East Asia (Fig. 4b). However, the biases are obviously reduced in Eurasia, India, and North America by dynamical downscaling (Figs. 4d,f). Such an improvement is apparent regardless of the resolution. Furthermore, the quantities of precipitation for the main rainband, including the ITCZ over the ocean, are realistic. The correlation coefficients between each dataset (RA62, DA126, and DA62) and the GPCP are all greater than 0.7, which indicates that the overall spatial pattern is reproduced well by all datasets. The results confirm that the downscaled data may not only retain the accuracy of large-scale fields but also correct erroneous precipitation amounts that existed in the original reanalysis.
4. Summary and conclusions
A 31-yr downscaled dataset is created using a T126L28 (about 100-km resolution) global model forced by RA2 with T62L28 as the large-scale forcing. The fit of the dynamical global downscaled dataset to observations is also investigated. A comparison of the downscaled dataset to radiosonde observations reveals improvements with respect to temperature and wind. The improvement in the downscaled precipitation is more prominent. The spatial distribution of the downscaled precipitation adequately reproduces that from observations. In addition, regional details over land and tropical oceans are better matched with observations because of the improved physics.
The datasets from the dynamical global downscaling are supposed to describe large-scale flow consistent with near-perfect circulation in the reanalysis, yet improvements on synoptic or smaller scales including precipitation are achieved because of the model at higher resolutions with physics. This indicates that the downscaled dataset can overcome both the systematic biases in reproducing prognostic components in a model and inhomogeneities introduced by the increasing amount of observations. Inhomogeneous differences in the sea level pressure between RA2 and DA126 over the data-sparse regions, which showed up when new data became available (shown in Figs. 1 and 2 of Kistler et al. 2001), confirm that the artificial trends of the conventional reanalysis data are significantly alleviated in the downscaled dataset (not shown). Therefore, the dataset employed in this work can be utilized for long-term climate research related to recent climate shifts. While the reanalyzed precipitation may have some bias due to both inadequacies in the physical parameterizations of the forecast model used in the reanalysis and the observational errors of large-scale variables, it can provide some important information on global climate change without repeated high-cost data assimilation.
Acknowledgments
This work was funded by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and Technology. The authors would like to acknowledge the support from the KISTI supercomputing center through the strategic support program for the supercomputing application research (KSC-2010-G2-0001).
REFERENCES
Alpert, J. C., M. Kanamitus, P. M. Calpan, J. G. Sela, G. H. White, and E. Kalnay, 1988: Mountain induced gravity wave drag parameterization in the NMC medium-range forecast model. Preprints, Eighth Conf. on Numerical Weather Prediction, Baltimore, MD, Amer. Meteor. Soc., 726–733.
Byun, Y.-H., and S.-Y. Hong, 2007: Improvements in the subgrid-scale representation of moist convection in a cumulus parameterization scheme: The single-column test and impact on seasonal prediction. Mon. Wea. Rev., 135, 2135–2154.
Cai, M., and E. Kalnay, 2005: Can reanalysis have anthropogenic climate trends without model forcing? J. Climate, 18, 1844–1849.
Campana, K. A., Y.-T. Hou, K. E. Mitchell, S.-K. Yang, and R. Cullather, 1994: Improved diagnostic cloud parameterization in NCEP’s global model. Preprints, 10th Conf. on Numerical Weather Prediction, Portland, OR, Amer. Meteor. Soc., 324–325.
Charnock, H., 1955: Wind stress on a water surface. Quart. J. Roy. Meteor. Soc., 81, 639–640.
Chen, M., W. Shi, P. Xie, V. B. S. Silva, V. E. Kousky, R. W. Higgins, and J. E. Janowiak, 2008: Assessing objective techniques for gauge-based analyses of global daily precipitation. J. Geophys. Res., 113, D04110, doi:10.1029/2007JD009132.
Chou, M.-D., and K.-T. Lee, 1992: A solar radiation model for use in climate studies. J. Atmos. Sci., 49, 762–772.
Chou, M.-D., and K.-T. Lee, 1996: Parameterizations for the absorption of solar radiation by water vapor and ozone. J. Atmos. Sci., 53, 1203–1208.
Chou, M.-D., and M. J. Suarez, 1999: A solar radiation parameterization (CLIRAD-SW) for atmospheric studies. NASA Tech. Memo NASA/TM-1999-104606, Vol. 15, 38 pp.
Chou, M.-D., and K.-T. Lee, 2005: A parameterization of the effective layer emission for infrared radiation calculations. J. Atmos. Sci., 62, 531–541.
Chou, M.-D., K.-T. Lee, S.-C. Tsay, and Q. Fu, 1999: Parameterization for cloud longwave scattering for use in atmospheric models. J. Climate, 12, 159–169.
Chun, H.-Y., and J.-J. Baik, 1998: Momentum flux by thermally induced internal gravity waves and its approximation for large-scale models. J. Atmos. Sci., 55, 3299–3310.
Ek, M. B., K. E. Mitchell, Y. Lin, E. Rogers, P. Grunmann, V. Koren, G. Gayno, and J. D. Tarpley, 2003: Implementation of Noah and surface model advances in the National Centers for Environmental Prediction operational mesoscale Eta model. J. Geophys. Res., 108, 8851, doi:10.1029/2002JD003296.
Fels, S. B., and M. D. Schwarzkopf, 1975: The simplified exchange approximation: A new method for radiative transfer calculations. J. Atmos. Sci., 32, 1475–1488.
Hong, S.-Y., 2010: A new stable boundary-layer mixing scheme and its impact on the simulated East Asian summer monsoon. Quart. J. Roy. Meteor. Soc., 136, 1481–1496.
Hong, S.-Y., and H.-L. Pan, 1996: Nonlocal boundary layer vertical diffusion in a medium-range forecast model. Mon. Wea. Rev., 124, 2322–2339.
Hong, S.-Y., and H.-L. Pan, 1998: Convective trigger function for a mass flux cumulus parameterization scheme. Mon. Wea. Rev., 126, 2599–2620.
Hong, S.-Y., H.-M. H. Juang, and Q. Zhao, 1998: Implementation of prognostic cloud scheme for a regional spectral model. Mon. Wea. Rev., 126, 2621–2639.
Hong, S.-Y., Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package with an explicit treatment of entrainment processes. Mon. Wea. Rev., 134, 2318–2341.
Hong, S.-Y., J. Choi, E.-C. Chang, H. Park, and Y.-J. Kim, 2008: Lower tropospheric enhancement of gravity wave drag in a global spectral atmospheric forecast model. Wea. Forecasting, 23, 523–531.
Huffman, G. J., R. F. Adler, M. M. Morrissey, D. T. Bolvin, S. Curtis, R. Joyce, B. McGavock, and J. Susskind, 2001: Global precipitation at one-degree daily resolution from multi-satellite observations. J. Hydrometeor., 2, 36–50.
Jeon, J.-H., S.-Y. Hong, H.-Y. Chun, and I.-S. Song, 2010: Test of a convectively forced wave drag parameterization in a general circulation model. Asia-Pac. J. Atmos. Sci., 46, 1–10.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471.
Kanamaru, H., and M. Kanamitsu, 2007: Scale-selective bias correction in a downscaling of global analysis using a regional model. Mon. Wea. Rev., 135, 334–350.
Kanamitsu, M., and Coauthors, 2002a: NCEP dynamical seasonal forecast system. Bull. Amer. Meteor. Soc., 83, 1019–1037.
Kanamitsu, M., W. Ebisuzaki, J. Woollen, S.-K. Yang, J. J. Hnilo, M. Fiorino, and G. L. Potter, 2002b: NCEP-DOE AMIP-II reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 1631–1643.
Kim, E.-J., and S.-Y. Hong, 2010: Impact of air–sea interaction on East Asian summer monsoon climate in WRF. J. Geophys. Res., 115, D19118, doi:10.1029/2009JD013253.
Kim, Y.-J., and A. Arakawa, 1995: Improvement of orographic gravity-wave parameterization using a mesoscale gravity wave model. J. Atmos. Sci., 52, 1875–1902.
Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-Year Reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82, 247–267.
Louis, J. F., 1979: A parametric model of vertical eddy fluxes in the atmosphere. Bound.-Layer Meteor., 17, 187–202.
Noh, Y., W. G. Cheon, S.-Y. Hong, and S. Raasch, 2003: Improvement of the K-profile model for the planetary boundary layer based on large eddy simulation data. Bound.-Layer Meteor., 107, 401–427.
Onogi, K., and Coauthors, 2007: The JRA-25 reanalysis. J. Meteor. Soc. Japan, 85, 369–432.
Pan, H.-L., and L. Mahrt, 1987: Interaction between soil hydrology and boundary layer developments. Bound.-Layer Meteor., 38, 185–202.
Pan, H.-L., and W.-S. Wu, 1995: Implementing a mass flux convection parameterization package for the NMC Medium-Range Forecast Model. NMC Office Note 409, 40 pp. [Available from NCEP/EMC 5200 Auth Road, Camp Springs, MD 20746.]
Park, H., and S.-Y. Hong, 2007: An evaluation of a mass-flux cumulus parameterization scheme in the KMA global forecast system. J. Meteor. Soc. Japan, 85, 151–168.
Park, H., S.-Y. Hong, and H. B. Cheong, 2008: Implementing a double Fourier series dynamic core into a global atmospheric model. Extended Abstracts, 20th Conf. on Climate Variability and Change, New Orleans, LA, Amer. Meteor. Soc., 15.B2. [Available online at http://ams.confex.com/ams/88Annual/techprogram/paper_134805.htm.]
Reynolds, R. W., and T. M. Smith, 1994: Improved global sea surface temperature analyses using optimum interpolation. J. Climate, 7, 929–948.
Screen, J. A., and I. Simmonds, 2011: Erroneous Arctic temperature trends in the ERA-40 reanalysis: A closer look. J. Climate, 24, 2620–2627.
Thorne, P. W., and R. S. Vose, 2010: Reanalyses suitable for characterizing long-term trends: Are they really achievable? Bull. Amer. Meteor. Soc., 91, 353–361.
Tiedtke, M., 1984: The sensitivity of time-mean large-scale flow to cumulus convection in the ECMWF model. Proc. Workshop on Convection in Large-Scale Numerical Models, Reading, United Kingdom, ECMWF, 297–316.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 2961–3012.
Walsh, J. E., W. L. Chapman, and D. H. Portis, 2009: Arctic cloud fraction and radiative fluxes in atmospheric reanalyses. J. Climate, 22, 2316–2334.
Wang, H., and V. M. Mehta, 2008: Decadal variability of the Indo-Pacific warm pool and its association with atmospheric and oceanic variability in the NCEP–NCAR and SODA reanalyses. J. Climate, 21, 5545–5565.
Wang, J., H. L. Cole, D. J. Carlson, E. R. Miller, K. Beierle, A. Paukkunen, and T. K. Laine, 2002: Corrections of humidity measurement errors from the Vaisala RS80 radiosonde—Application to TOGA COARE data. J. Atmos. Oceanic Technol., 19, 981–1002.
Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical outputs. Bull. Amer. Meteor. Soc., 78, 2539–2558.
Yhang, Y.-B., and S.-Y. Hong, 2008: Improved physical processes in a regional climate model and their impact on the simulated summer monsoon circulations over East Asia. J. Climate, 21, 963–979.
Yoshimura, K., and M. Kanamitsu, 2008: Dynamical global downscaling of global reanalysis. Mon. Wea. Rev., 136, 2983–2998.