We present an analysis of a simulated climate projection covering the period 1971-2080 over the Korean Peninsula with a regional climate model (RegCM3) using a one-way double-nested system. The mother and nested domain cover the East Asian region at 60 km grid spacing and the Korean Peninsula at 20 km grid spacing, respectively. The mean climate state as well as the frequency and intensity of daily extreme events are investigated at various temporal and spatial scales, with a focus on surface air temperature and precipitation. Our analysis is primarily centered on the comparison of two 30-year climate periods between 2021-2050 and 2051-2080, in order to assess climatic changes at different time periods in response to anthropogenic forcings under the IPCC B2 emission scenario. Regarding the interdecadal variation of projected temperature and precipitation over the whole integration period, we find that the temperature change gradually increases by 3.2 K for the 2070s, with a persistently increasing trend. The projected precipitation in the future shows generally greater amounts than in the reference run despite significant interdecadal variation. A substantial increase (decrease) of hot (frost) spells is projected along with increasing of maximum and minimum temperature. Wet spells of long periods tend to be more frequent, accompanying the increase of precipitation amounts. The temperature changes are statistically significant at the 90% confidence level in all seasons and regions, indicating robustness of the projected change while the precipitation changes show low statistical confidence, especially summer season, due to the large natural variability. Based on the spatial distribution, although the general pattern between the mother and nested domain simulations shows similarity, more detailed structure over Korea is found in the nested domain simulation. By comparison of the change between 2021-2050 and 2051-2080 periods, the climate change signal of both temperature and precipitation becomes more pronounced in the late 21 st century as the greenhouse gas concentration is increased.
Precipitable water (PW) obtained from a permanent ground-based Global Positioning System (GPS) network on the Korean Peninsula was evaluated by using radiosonde measurements for two years. The GPS-derived PW is also validated using 5-min and 2-hourly troposphere products from International GNSS Service (IGS). The GPS-derived PW from our analysis and the IGS products were reasonably consistent with mean bias (Ours minus IGS products) of- 0.27 mm and a standard deviation of 0.78 mm. The PW derived from radiosonde measurement (radiosonde PW) shows dry bias relative to GPSderived PW, where the mean bias (GPS minus radiosonde) is 1.50 mm with a standard deviation of 2.45 mm. We also found systematic bias in radiosonde PW depending on the radiosonde launch time. At the site with 6-hourly high temporal resolution soundings, the bias is greater during day (0900 and 15 LST) than at night (2100 and 0300 LST) with a difference of 1.34 mm in the maximum. In addition, our results show the dependency of error statistics on the absolute amount of PW and seasons. The bias increased for large amounts of PW especially in the daytime. Standard deviation tended to increase under humid conditions, but does not vary with observation time. The possible bias in GPS-derived PW caused by un-modeled phase center variation is also simulated in this study. Most phase observation errors are attributed to multi-path effects caused by the structure of monuments, where the errors are relatively small against the dry biases found in radiosonde PW. Our results addressed the accuracy of GPS-derived PW with reasonable quality and showed strong temporal and spatial variability of PW due to geographic location. These results suggest that GPSderived PW produced by the Korean regional GPS network has great potential to be used for various meteorological applications and could also contribute global GPS climatology.
A statistical study was made for surface wind fields of tropical cyclones (TCs) at landfall on the main islands of Japan using the AMeDAS data. The analysis was performed for 70 TCs which made landfall on the southern part (south of 36°N) of the main islands during the period 1979-2004. The wind field within 200 km of each storm center was obtained by applying a composite analysis to hourly or tenminute observations for the four hours around landfall, with the direction of storm motion as the y-axis. Then the fuzzy c-means method (FCM), which is a kind of cluster analysis, was applied to classify the wind distribution. All but one of the resulting five patterns show stronger winds on the right side of the center than on the left, with different degrees of right-left asymmetry and the shapes of the strong wind area. The five patterns have some differences in the preferred season, central pressure, speed and direction of motion, and regions of landfall of storms. A supplementary analysis using the wind profiler data, available from 2001, reveals that the right-left asymmetry is roughly explained by the storm motion, although the storm-relative wind fields still have some asymmetry.
Several meso-a-scale lows (MAL), which developed along the Baiu Front in the year of 2001 are studied by an observational analysis and a numerical experiment. The observational analysis utilizes the GANAL and RANAL data which have a horizontal resolution of 125 km and 20 km, respectively. The observed MALs are found to be categorized into two types: One is accompanied by an upper-level disturbance, and has a trough tilting westward with increasing height. The other is not accompanied by an upper-level disturbance, and has a trough tilting eastward with height. A MAL of the latter type that appeared on 20 June 2001 is examined in detail. This MAL had a horizontal pressure trough which runs in the southwest-northeast direction. It was associated with a cold temperature anomaly at the low-level to the east side of the low center, which is consistent with the eastward tilting trough in view of the hydrostatic balance. A numerical simulation using a 5 km-mesh Japan Meteorological Agency Non- Hydrostatic Model, without a cumulus parameterization, successfully reproduced the observed characteristics of the MAL. A budget analysis of the thermodynamic equation shows that the cold anomaly at the low-level in the east side of the MAL, is caused by an adiabatic cooling associated with the strong updraft induced by a condensational heating in the mid- and upper-levels.
In this study, the energy spectrum and the energy interactions of the atmospheric general circulation are analyzed, using the expansion in three dimensional normal mode functions for JRA-25 and ERA-40. The analytical solutions of the vertical structure equation are used for the expansion in the vertical direction. The problem of aliasing associated with the numerical solution is avoided by the present approach. According to the result of the energy spectrum, using the analytical vertical structure functions, it is found that the energy spectrum indicates a clear peak in the middle vertical modes, and the spectrum decreases monotonically at the higher order vertical modes. The energy interactions for lower order vertical modes are consistent with that by Tanaka and Kung (1988). It is found from the analysis of the energy interactions that there is another energy source region in the higher order vertical modes in the zonal field. It is also found from the energy flux analysis in the vertical domain that the atmospheric energy is converted from baroclinic component to barotropic component.
The Kiyokawa-dashi (strong local wind) was observed with a coherent Doppler lidar at the exit of the Mogami Valley, in Kiyokawa (38.80°N, 140.01°E, 20 m MSL) during the summers of 2003 and 2004. Used to define days on which the Kiyokawa-dashi wind blew were meteorological data obtained from four automatic weather stations installed at four observation sites: Kiyokawa, Karikawa (38.79 °N, 139.99°E, 15 m MSL), Mawadate (38.81°N, 139.92°E, 10 m MSL) and Hirono (38.83°N, 139.85°E, 5 m MSL). When the Kiyokawa-dashi blew, the synoptic patterns tended to be high-pressure in the east and low-pressure in the west. After the Kiyokawa-dashi was extracted, the data measured by the coherent Doppler lidar with a high spatial and temporal resolution were then utilized to investigate it. On the basis of the range height indicator and Plane position indicator scans, the coherent Doppler lidar was able to detect the three-dimensional dynamical structure of the Kiyokawa-dashi up to 8-10 km downstream from the Kiyokawa observation site. The coherent Doppler lidar clearly showed that a low-level critical layer (horizontal wind speed Approx 0 m s -1 ) existed at altitudes of 0.5-1.2 km, and that a strong easterly/southeasterly wind confined below the critical layer blows away about 10-km downstream of the Mogami Valley. Intensive upward/downward vertical motion under the critical layer was indicated by the velocity-azimuth display scan. The line-of-sight wind speeds obtained by high-frequency plane position indicator scans were high, not only near the exit of the Mogami Valley but also over the entire Shonai plain, and the distributions of the line-of-sight wind speeds were almost steady, indicating the existence of transient disturbances.
In this paper we present evidence that Asian dusts can become more absorbing aerosols when mixed with pollutants. Spectral direct/diffuse solar radiations were measured on April 7, 2000 at Anmyeon-Do, Korea, when a narrow and heavy dust band moved through the Anmyeon measurement site. Lowest SSA values (about 0.91) are found within the dust band, while comparatively high values (up to 0.97) are found outside over the non-dust area at about 300 km away from the dust band. Considering that the air masses passed through the Anmyeon site took nearly the same path through the Beijing area, obtained results strongly suggest that there may be interactions between air pollutants and mineral dust aerosols that lead to the decreased SSA and thus to the more solar absorption by dust aerosols. In other words, Asian dusts tend to become darkened after mixing with pollutant particles produced over the industrial /urban area of China.
This paper examines true oscillations in the Northern Hemisphere winter by using independent component analysis (ICA). ICA can distinguish between true and apparent oscillations under the assumption that the true oscillations are mutually independent. Particular attention is paid to the Arctic Oscillation (AO). For this purpose, the NCEP-NCAR reanalysis data (NCEP-NCAR data) and the data of the present climate experiment of the Meteorological Research Institute (MRI data) are used. There may be certain issues in ICA applied to meteorology: the selection of data periods, treatment of noise, and relationship between the number of dimensions in phase space and the number of independent components. We make several considerations and proposals about these issues. ICA should be performed for periods for which the variance is almost the same. Since independent components after whitening are not necessarily uncorrelated with each other under the existence of noise, and the relation between the dimension of phase space and the number of independent components cannot be predetermined, we propose the method for seeking independent components by kurtosis as the most relevant in meteorology. On the basis of the above considerations and proposals, ICA is performed on the NCEP-NCAR data. Independent components are found for the sea level pressure (SLP) and 500 hPa height (Z500) fields. They are the North Atlantic Oscillation (NAO) and the Pacific-North American Oscillation (PNA). Thus, the AO is an apparent mode derived from them. However, since the period of the data is too short, statistical significance cannot be obtained. Then, ICA is performed on the MRI data. Also in this data, the NAO and the PNA-like oscillation are independent for both the SLP and Z500, where the PNA-like oscillation implies that its pattern is somewhat different from the observed PNA pattern. It can be concluded again that the AO is not independent, but this time with statistical significance.
Future projection of the Arctic Oscillation (AO) signature and its significance towards the northern hemispheric surface temperature trend have been examined using 20 state-of-the-art Atmosphere-Ocean General Circulation Model (AOGCM) outputs forced under the IPCC SRES-A1B and 20C3M emission scenario. Models are mostly successful in simulating the observed AO structure and the corresponding surface temperature variability. It is found that while AO exhibits a large positive trend, especially during the autumn season with a relatively smaller trend during the winter and spring seasons. In all seasons the interannual variance in AO remains the same for both scenarios. These features in the timeseries leads to two distinct patterns of temperature variability. One is the "polar amplification" pattern due to the long-term anthropogenic forcing, which is much larger in its amplitude. Another is related to the natural variability of the AO, which is confined over the land surface and is marked by a dipole pattern of temperature between the Eurasian continent and Greenland. It is argued that the gradual trend in the AO is not a result of enhanced natural variability of the AO dynamics itself, but rather a result of the large anthropogenic forced linear trend projected onto the mean climatological state of the Arctic region. Distinguishing these two patterns of warming is crucial for detecting the signal of future global warming trend over the Eurasian continent and other regions.
The dynamics and predictability of a downward migration event of the negative Northern Annular Mode (NAM) anomaly following a stratospheric sudden warming (SSW) occurring in January 2003 are examined using the operational 1-month ensemble forecast data set provided by the Japan Meteorological Agency. It is found that the predictable period of the tropospheric negative NAM anomaly is at most 6 days, which is much shorter than that of the precedent SSW event. The tropospheric NAM anomaly is caused by the E-P flux convergence associated with zonal wavenumber 2 planetary waves at the tropopause level, of which propagating property is affected by the zonal-mean zonal wind there. Our results suggest that even large stratospheric circulation changes associated with the SSW event have only limited influence on the predictability of the tropospheric circulation.
In this study, EOF analyses are conducted for the barotropic component of the atmosphere in the Northern Hemisphere for each season to investigate the dominant low-frequency variability in the atmosphere. It shows that the structure of the Arctic Oscillation (AO) appears in winter and spring as a dominant mode. In summer and fall, however, the dominant modes are not the annular pattern, but show more localized structures. The seasonal variation of the dynamical SVD-1 mode is analyzed using the climate basic states based on the singular eigenmode theory by Tanaka and Matsueda (2005) to compare with the observed EOF-1 for each season. It is found that the AO-like structure appears robustly in DJF when hyper diffusion is used in the model. The AO-like pattern obtained by observed EOF-1 appears in MAM when the viscosity is changed to the bi-harmonic diffusion. But the structures obtained as the SVD-1 in JJA and SON are different from that of the observed EOF-1. From this result, we may conclude that the AO in winter is explained by the singular eigenmode theory. However, the dominant low-frequency variability in other seasons are unlikely to be the singular eigenmode of the dynamical system explained by the SVD analysis.
Drifting buoys deployed near the North Pole in the 2000s revealed that sea-ice export from the Arctic Ocean through the Fram Strait was large in 2005. The relationship between ice drift and sea-level pressure (SLP) showed that the mean SLP from June to September is conducive to continuous sea-ice drift offshore from Siberia. A record minimum of the ice extent in 2005 is partly explained by the strongest SLP gradient across the Transpolar Drift Stream for the recent 27 years from 1979. The SLP pattern regressed on the linear trends of ice extent is characterized by cyclonic circulation anomalies along the Eurasian coast, which tend to enhance the SLP gradient across the Transpolar Drift Stream. The large export of freshwater and sea ice in 2005 could increase the salinity of the upper ocean and lead to development of the mixed layer and suppression of sea-ice growth during the subsequent winter.
The effect of the Quasi-Biennial Oscillation (QBO) and the El Niño Southern Oscillation (ENSO) on the 11-year solar cycle modulation of the winter-mean North Atlantic Oscillation (NAO) is examined through analysis of observational data from 1958 to 2000. It is found that the solar cycle modulation of the NAO is more strongly enhanced in the westerly phase of the 50-hPa QBO wind and the cold phase of ENSO, although separation of these effects is statistically difficult. On these phases, the signal of the winter-mean NAO extends more to the upper stratosphere and summer-AO reappears more strongly in high solar years, whereas the signal is weaker throughout in low solar years.
Inter-basin links between the North Pacific and North Atlantic in the wintertime upper-tropospheric circulation over the extratropical Northern Hemisphere, including their structure, dominance and seasonal dependence, are examined based on the structures of the leading EOFs for monthly height anomalies for the second half of the last century. Dominant variability in the upper troposphere is found to reflect the corresponding modulations in the strength of the inter-basin dynamical link between the North Pacific and North Atlantic. The first and second EOFs for the upper troposphere represent hemispheric patterns, showing the coherent variability between the North Pacific and North Atlantic. They essentially reflect the respective out-of-phase and in-phase relationships between the Icelandic low (IL) and Aleutian low (AL), as upper-level manifestations of the AL-IL seesaw and the cold ocean-warm land (COWL) pattern, respectively. Dominance of the AL-IL seesaw in the upper-tropospheric leading EOF is a manifestation of a dynamical linkage between the North Atlantic Oscillation and Pacific-North
The purpose of this study is to analyze the possible impact of solar activity and geomagnetic activity upon the Arctic Oscillation (AO) as an external forcing to the atmosphere. Maximum and minimum periods are defined for the solar flux density at 10.7 cm wavelength (F10.7) and the amplitude antipodal index (aa index). In each period, the monthly mean of barotropic height anomaly, zonal-mean wind anomaly, and zonal-mean temperature anomaly are calculated, and the difference between the maximum and minimum period is analyzed. As a result, the structure of the AO is found in winter when the composite of anomaly is analyzed for each month. Significance test shows that the aa index has wider significant area than that of F10.7. According to the result for the zonal-mean temperature anomaly, a heating area is found in the lower stratosphere for both indices. But in the case of the aa index, the characteristic heating or cooling areas are stronger during the maximum period than that for the minimum period in the vicinity of 1 hPa. There is a possibility that the upper atmospheric differences in temperature can be a trigger of the excitation of the AO in the lower atmosphere.
The predictability of a minor warming and the subsequent major warming in the winter of 2003/04 is examined in comparison with that of the major warming without preceding minor warmings in December 2001 by the use of operational ensemble 1-month forecast data produced by the Japan Meteorological Agency. In the case of the major warming in December 2001 caused by amplified zonal wavenumber 1 planetary waves, the predictable period based on zonal mean temperatures in the polar stratosphere is estimated to be at least 16 days, while the warmings in the winter of 2003/04 are predictable at most 9 days in advance. Such relatively reduced predictability for the latter warmings is considered due to the rather complicated time evolution of the warming episodes with a significant contribution of smallerscale planetary waves during the period prior to the warmings.
The leading empirical orthogonal function (EOF) of the geopotential height is called the Northern Annular Mode (NAM). Stratosphere-troposphere connections seen in the multilevel NAM signal are often considered as due to a propagation of the annular structure. However, the tropospheric leading EOF contains not just an annular structure, but also a wave structure. This indicates that the NAM index is sensitive to changes in wave activity. A previous study showed that an apparent downward propagation of the NAM index can appear as a result of a reflection of wave 1 due to a stratospheric sudden warming. If this result is correct, we may also expect to find a reverse case. Here we show a case when an amplification of a tropospheric wave 1 leads to a stratospheric warming, an apparent upward propagation of the "Annular" mode signal, i.e., change in the amplitude of the NAM index appears due to a projection on the wave component of the tropospheric NAM.
In this study, teleconnectivity and one point correlations are analyzed for the barotropic component of the atmosphere represented by a simple barotropic model (called barotropic S-model), and the results are compared with a barotropic component calculated from the NCEP/NCAR reanalysis. It is confirmed that the teleconnection of the PNA pattern is simulated reasonably well by the barotropic S-model. The question is focused on whether the Arctic Oscillation (AO) would appear by the one point correlation in the model atmosphere. According to the result, it is found that the one point correlation centered at the Azores High is not restricted within the Atlantic sector like the NAO, but spreads to entire midlatitudes surrounding the Arctic. The positive correlations in the Pacific sector are statistically significant at the 5% confidence level for both the model and observations. From the result, we may suggest that the dynamical AO mode is embedded in the statistical EOF-1 even for the observed atmosphere. The reason for the enhanced correlation in the barotropic component compared with the observed sea level pressure may be that the former is controlled purely by the barotropic dynamics, while the latter is influenced by the baroclinic component of the atmosphere.
Long time-scale variations of the climate system can be induced by forcings loaded onto the stratosphere, such as increase of carbon dioxide, solar activity, and volcanic eruption, which is possibly modulated by the interaction with the oceans. Spatial structure of the atmospheric response resembles the Northern Annular Mode (NAM) in an experiment of a coupled atmosphere-ocean general circulation model with a westerly momentum forcing in the winter midlatitude stratosphere. The atmospheric response further induces distinctive patterns of the sea surface temperature (SST) change in the Atlantic and the Pacific, which are reminiscent of the observed decadal to interdecadal variability dominant in each basin. Investigation of oceanic feedback with the atmospheric model forced by the SST change suggests that the oceanic response has positive feedback effects on the NAM-like atmospheric structure, implying enforcement or maintenance of the response to the stratospheric forcing. The positive feedback effect in the Atlantic is significant in winter, while that in the Pacific is significant in spring.