155SOLA, 2012, Vol. 8, 155−159, doi:10.2151/sola.2012-038

AbstractDiurnal variations of Korean summertime (June−August)

precipitation in 2009 were investigated using hourly National Institute of Meteorological Research/Korea Meteorological Administration (NIMR/KMA) Forecast Research Laboratory (FRL) precipitation data that had high spatial (5 km by 5 km grid distance) and temporal (1 h) resolutions. Using the techniques of multiresolution analysis and Incomplete Gamma Function, NIMR/KMA FRL precipitation reanalysis data are produced from the observations of about 680 Automatic Weather Systems and reflec-tivity data from 10 radars over South Korea.

Three dominant modes of diurnal variations in 2009 summer precipitation over South Korea were indentified via the cyclosta-tionary EOF (CSEOF) technique. Nocturnal precipitation maxima were the result of rain band enhancement from instability due to radiative cooling at the cloud top during the nighttime. This precipitation over the central region of South Korea strengthened and moved rapidly southeastward during the nighttime and then dissipated in the southern coast area of Korea by mid morning. Over the north-central region of South Korea, the daytime prevail-ing precipitation was nearly stationary from early morning to late afternoon. The precipitation over the southern coastal periphery of Korea also moved slowly eastward from midnight to mid after-noon.

1. Introduction

Abnormal weather/climate is generally believed to be a rare phenomenon that occurs once in 30 years or more, but such phenomena have become more frequent. In particular, localized heavy rainfall for a short time has increased markedly in both frequency and intensity in some mid-latitude locations. Therefore, the diurnal cycle of precipitation has been considered to play an increasingly important role in regional weather and climate (e.g. Wallace 1975; Dai 2001; Yang and Slingo 2001). Furthermore, diurnal patterns in precipitation could be used to forecast rainfall events and to improve the physical processes of numerical models if there are typical daily rainfall cycles. Many studies in different regions show that there are distinct diurnal precipitation cycles during the warm-season, with rainfall peaks in the late afternoon or early morning, using surface station data (e.g. Wallace 1975; Dai et al. 1999; Dai 2001) and satellite data (e.g. Yang and Slingo 2001), with and without model simulation. The diurnal variations in precipitation have often been influenced by the temperature contrast induced from differential heating between land and sea, or high and low lands with elevation gradient, and changes in frictional drag associated with diurnal variation in static stability (e.g. Carbone et al. 2002; Wang et al. 2004).

In the case of East Asia, since topographical characteristics such as elevation gradients and land/sea distributions are complex, there are large regional differences in the time at which maximum

diurnal precipitation occurs, especially during the summer season. Analysis of diurnal variations in summer precipitation over contiguous China showed that peak precipitation occurred in late afternoon over southern inland and northeastern China, but at mid-night over most of the Tibetan Plateau and its eastern periphery (Yu et al. 2007; Zhou et al. 2008). To investigate the diurnal rainfall variations in regions with complex topography, Chen et al. (2012) stated that a combination of station observations and satellite products are essential.

The diurnal variations over Korea, which might be accompa-nied to those researches, were analyzed to be typified by two peak values in the early morning and afternoon (Ramage 1952; Lim and Kwon 1998). However, they had mainly been limited to using surface station data. In addition, there are multiple potential causes of summertime precipitation in South Korea. Representatively, the middle latitude and subtropical weather fronts, associated with a quasi-stationary moisture convergence zone (or band) between the tropical maritime air mass from south and the continental and maritime polar air masses from the north, make up one of the main rain-bearing synoptic systems during the summer rainy season in East Asia (Ding and Sikka 2006; Lee 2004). Localized heavy rainfalls with convective instability induced by increasing low level warm and moist air and the direct/indirect effect of trop-ical cyclone are also dominated in the summertime precipitation in South Korea (Lee 2004). In order to help resolve the spatial and temporal distribution and the diurnal variations in summer precipitation over South Korea in more detail, we used NIMR/KMA (National Institute of Meteorological Research/Korea Meteorological Administration) FRL (Forecast Research Labo-ratory) precipitation reanalysis data, which will be introduced in Section 2, in this study.

2. NIMR/KMA FRL precipitation reanalysis data

NIMR/KMA developed high resolution precipitation data with 5 km by 5 km resolution in order to be applied predominately in real-time forecasting over South Korea. KMA have deployed a high-density Automatic Weather System (AWS) network over South Korea to collect real-time observations of surface mete-orological parameters including rainfall. The precipitation data observed at ~680 AWS stations over South Korea (Fig. 1) were in-terpolated to 650 by 650 grid points at 1-km horizontal resolution using the method of multiquadric interpolation objective analysis (e.g. Nuss and Titley 1994). In addition, precipitation data derived from reflectivity data from ten radars, which cover all of South Korea including its territorial waters (cf. Fig. 1), were combined with AWS precipitation data through multiresolution analysis (e.g. Chin et al. 1998) and Incomplete Gamma Function, in order to set the same grid distributions. The Incomplete Gamma Function, which is a kind of cumulative probability density function, has been widely adopted for a range of studies (e.g. Blahak 2010; Lloyd-Hughes and Saunders 2002). Precipitation data are then produced through applying different weighting to the land and ocean areas. NIMR/KMA FRL precipitation reanalysis data are reproduced for case analysis of rainfall events and investigating climatic precipitation over South Korea.

Diurnal Variations of Summertime Precipitation in South Korea in 2009 Using Precipitation Reanalysis Data

Joon-Woo Roh, Yong Hee Lee, Ji-Eun Nam, and Kwan-Young ChungForecast Research Laboratory, National Institute of Meteorological Research,

Korea Meteorological Administration, Republic of Korea

Corresponding author: Joon-Woo Roh, Forecast Research Laboratory, National Institute of Meteorological Research, Korea Meteorological Administration, Seoul, 156-720, Republic of Korea. E-mail: neolegacy@gmail.com. ©2012, the Meteorological Society of Japan.

156 Roh et al., Diurnal Variations of Summertime Precipitation in South Korea

first Changma is active from late June to late July. The periods with stronger fluctuations in the three PC time series are mainly observed within the first Changma period.

Figure 3 shows the spatial distributions of the first CSEOF of NIMR/KMA FRL precipitation at 1-h intervals through a full di-urnal cycle (24 h). The first CSEOF of summer precipitation over South Korea explains 19.7% of the total variability. The variance explained by the first mode may look small, but this mode explains 61.3% of the total amount of precipitation. The amplitude of the diurnal cycle fluctuates around 0.27 mm with a standard deviation of 0.55 mm in this mode. The precipitation distributions from Fig. 3a to Fig. 3x explain the diurnal variation in the first mode of precipitation variability. Two precipitation bands at 0100 local standard time (LST) were widely distributed in the north-central area (~37°N−38°N) and in the central area (~35°N−37°N) of South Korea (Fig. 3a). A precipitation band with an orientation typical to the front originates in the north-central periphery of South Korea at 2000 LST (Fig. 3t), moves towards the southern coast of Korea until 0800 LST the next day (Fig. 3u−3x and 3a−3h), and then dissipates (Fig. 3i). This nocturnal precipitation can be assumed to be enhanced by instability due to nocturnal radiative cooling at the cloud top (Lin et al. 2000). The diurnal variation in summer precipitation seemed to correspond to the precipitation pattern of meso-scale convective systems (MCS), which have predominantly nocturnal rain, mostly after midnight,

3. Methods

In order to investigate diurnal variations of summertime pre-cipitation in South Korea in detail, the cyclostationary empirical orthogonal function (CSEOF) technique was employed (Kim et al. 1996; Kim and North 1997). In this technique, space-time data P(r, t ) are written as a linear combination of CSEOFs:

p(r, t ) = Σn LVn(r, t )PCn(t ), (1)where LVn(r, t ) and PCn(t ) are cyclostationary loading vectors, which are described time-dependent physical evolutions, and principal component (PC) time series, respectively. The CSEOF representation of the data is different from the EOF representation in that the loading vectors are time dependent and periodic, such that

LVn(r, t ) = LVn(r, t+d ). (2)where d is the nested period. In this research, the physical evolu-tion with a nested period of 24 h exhibits the long-term amplitude fluctuations described in the PC time series. CSEOF analysis has been conducted for this summer period (June, July, and August; 2208 h) using 5 km NIMR/KMA FRL precipitation reanalysis data from the spatial domain of South Korea (123.361°E−131.468°E by 32.0202°N−39.4249°N; 135 point by 165 point).

4. Three distinctive diurnal variations in summer-time precipitation for different regions

Each mode of diurnal variability in summertime precipitation over South Korea in 2009 was analyzed via CSEOF analysis. CSEOF analysis of 1-h intervals of summertime precipitation re-vealed three principal modes of variability, which together explain about 44% of the total variability. Figure 2 shows the PC time series of the first to the third CSEOF. The primary component of Korean summertime rainfall, which accounts for about three-fourths of the annual rainfall in Korea, is caused by Changma, which is associated with the East Asian Summer Monsoon from late June until late August (Ding 2004; Qian et al. 2002). The

Fig. 1. The AWS networks (red dots) and the areal coverage of the 10 radars for precipitation estimates (white region with blue curved line as the boundary).

Fig. 2. PC time series of the first (a), second (b), and third (c) CSEOF of diurnal variations in 2009 summertime precipitation in South Korea.

157SOLA, 2012, Vol. 8, 155−159, doi:10.2151/sola.2012-038

Fig. 3. The first CSLV of 1-hourly precipitation (1 mm contours) over South Korea in summer 2009 from (a) 0100 LST (1600 UTC) to (x) 2400 LST (1500 UTC) at 1-h intervals.

158 Roh et al., Diurnal Variations of Summertime Precipitation in South Korea

and probably arise from the upscale growth of late afternoon con-vection (Nesbitt and Zipser 2003). On the other hand, the precipi-tation band over the north-central area, which weakens near 0300 LST (Fig. 3c), strengthens from 0800 LST, lasts through daytime, and then dissipates at 1800 LST without much movement. In the first CSEOF, both the migratory precipitation band from the north-central region to the southern coast of Korea during the nighttime and the quasi-stationary precipitation area over the north-central area of South Korea in the daytime are explained. These two peaks in the diurnal precipitation cycle are consistent with those in the domain between the Yangtze river and the Yellow River valley (around 110°−120°E, 30°−40°N) (Bao et al. 2011; Yu et al. 2007a; Yu et al. 2007b). The two peaks of summer

precipitation in the area between the Yangtze River and the Yellow River valley had been firstly identified by Yu et al. (2007a), Yu et al. (2007b), and Zhou et al. (2008).

The second CSEOF of the diurnal cycle in 2009 summertime precipitation in South Korea is shown in Fig. 4. This mode ex-plains ~13.3% of the total variability. Relatively strong diurnal precipitation activity of this mode was seen in late June through mid July (Fig. 2b). The zonal type of precipitation at ~35°N, which increases from ~0100 LST (Fig. 4a), lasts and covers much of the southern coastal area of Korea, and then dissipates at ~1600 LST (Fig. 4f). The strong precipitation in the second mode was most marked during the morning. As the spatial distributions show that the onset of precipitation occurred in southwest South

Fig. 4. The second CSLV of 1-h precipitation anomalies (1 mm contours) over South Korea in summer 2009 from (a) 0100 LST (1600 UTC) to (x) 2400 LST (1500 UTC) at 3-h intervals.

Fig. 5. The third CSLV of precipitation anomalies, organized as Fig. 4.

159SOLA, 2012, Vol. 8, 155−159, doi:10.2151/sola.2012-038

Korea and the precipitation terminated in southeastern South Korea, the precipitation band moved slowly eastward. This mode of precipitation may be connected to an early morning peak in the variation in mean precipitation averaged over the region between the Yangtze River and the Yellow River Valley (110°E−120°E, 30°N−40°N) (Yu et al. 2007).

Figure 5 shows the diurnal evolution of 1-h precipitation in the third CSEOF. The third mode explains ~10.5% of the total variability. The strongest diurnal precipitation occurred on 14 July. In fact, an hourly rainfall of over 50−60 mm was recorded within the narrow region of the precipitation band shown in Figs. 5h, 5a, and 5b. This southwestwardly moving precipitation corresponds to the precipitation in the central region in the first CSEOF. Thus, detailed diurnal variations in summer precipitation in Korea could be identified through NIMR/KMA FRL reanalysis data with high spatial and temporal resolution.

5. Summary and discussions

Here, we used NIMR/KMA FRL precipitation reanalysis data based on high resolution AWS and radar data in order to inves-tigate diurnal variations in summer precipitation in more detail. The diurnal variations in summertime precipitation over South Korea in 2009 were successfully identified using the CSEOF technique. Via CSEOF analysis of 1-h summertime precipitation, three principal modes of variability, which together explain about 44% of the total variability, were obtained. As the evolution of the precipitation over South Korea was seen in the CS loading vector during 24 h, the results reveal detailed and interesting spatial and temporal features. Three prevailing features of diurnal variation in precipitation were identified for South Korea.

In the first CSEOF, which explains 61.3% of the total magni-tude of summer precipitation in 2009, the predominant nocturnal precipitation that occurred from evening to early morning and the daytime precipitation that occurred from early morning to late afternoon were found. While the nocturnal precipitation moved rapidly from the north-central area to the southern coast of South Korea, the daytime precipitation was nearly stationary over the north-central area of South Korea. The precipitation over the southern coastal periphery of South Korea from midnight to mid afternoon was extracted from the second CSEOF. The prevailing nocturnal precipitation band seemed to be enhanced by instability due to nocturnal radiative cooling at the cloud top (Lin et al. 2000). Considering occurrence time of peak precipitations, the precipitation over South Korea can be seen to be associated with diurnal variations in precipitation over the domain between the Yangtze River and the Yellow River valley (Yu et al. 2007a; Yu et al. 2007b; Zhou et al. 2008). In a sequel paper, we will study and address the relationship between the diurnal variation of precipitation in South Korea and that in the Yangtze River and the Yellow River valley.

Acknowledgements

This work was supported by NIMR-2012-B-1.

References

Bao, X., F. Zhang, and J. Sun, 2011: Diurnal variations of warm- season precipitation east of Tibetan Plateau over China. Mon. Wea. Rev., 139, 2790−2810, doi:10.1175/MWR-D-11-00006.1.

Blahak, U., 2010: Efficient approximation of the incomplete gamma function for use in cloud model applications. Geosci. Model Dev., 3, 329−336.

Carbone, R. E., J. D. Tuttle, D. A. Ahijevych, and S. b. Trier, 2002: Inferences of predictability associated with warm season precipitation episode. J. Atmos. Sci., 59, 2033−2056.

Chen, H. M., W. H. Yuan, J. Li, and R. C. Yu, 2012: A possible cause for different diurnal variations of warm season rainfall

as shown in station observations and TRMM 3B42 data over the Southeastern Tibetan Plateau. Adv. Atmos. Sci., 29, 193− 200, doi:10.1007/s00376-011-0218-1.

Chin, T. M., R. F. Milliff, and W. G. Large, 1998: Basin-scale, high-wave number sea surface wind fields from a multiresolu-tion analysis of scatterometer data. J. Atmos. Ocean. Tech., 15, 741−763.

Dai, A., 2001: Global precipitation and thunderstorm frequencies. Part II: Diurnal variations. J. Clim., 14, 1112−1128.

Dai, A., F. Giorgi, and K. E. Trenberth, 1999: Observed and model simulated diurnal cycles of precipitation over the contiguous United States. J. Geophys. Res., 104, 6377−6402.

Ding, Y. H., 2004: Seasonal march of the East Asian summer monsoon. The East Asian Monsoon, vol.2, C.-P. Chang, Ed., World Sci. Publ., 3−53.

Ding, Y. H., and Sikka, 2006: Synoptic systems and weather. The Asian Monsoon, B. Wang, Ed., Praxis. Springer, U. K., 131− 201.

Kim, K.-Y., and G. R. North, 1997: EOFs of harmonizable cyclos-tationary processes. J. Atmos. Sci., 54, 2416−2427, doi:10.1175 /1520-0469(1997)054<2416:EOHCP>2.0.CO;2.

Kim, K.-Y., G. R. North, and J. Huang, 1996: EOFs of one- dimensional cyclostationary time series: Computation, exam-ples and stochastic modeling. J. Atmos. Sci., 53, 1007−1017, doi:10.1175/1520-0469(1996)053<1007:EOODCT>2.0.co;2.

Lee, D.-K., 2004: Summertime climatology, Korean Climatology, H.-S. Chung, Ed., Meteorol. Res. Inst., Korea Meteorol. Admin., 92−99.

Lim, G.-H., and H.-J. Kwon, 1998: Diurnal variations of precipi-tation over South Korea and its implication. J. Korean Meteor. Soc., 34, 222−237.

Lin, X., D. A. Randall, and L. D. Fowler, 2000: Diurnal variability of the hydrologic cycle and radiative fluxes: Comparisons between observations and a GCM. J. Clim., 13, 4159−4179.

Lloyd-Hughes, B., and M. A. Saunders, 2002: A drought climatol-ogy for Europe. Int. J. Climatol., 22, 1571−1592.

Nesbitt, S. W., and E. J. Zipser, 2003: The diurnal cycle of rainfall and convective intensity according to three years of TRMM measurements. J. Clim., 16, 1456−1475.

Nuss, W. A., and D. W. Titley, 1994: Use of multiquadric inter-polation for meteorological objective analysis. Mon. Weather. Rev., 122, 1611−1631.

Ramage, C. S., 1952: Diurnal variation of summer rainfall over east China, Korea, and Japan. J. Meteor., 9, 83−86.

Yu, R., T. Zhou, A. Xiong, Y. Zhu, and J. Li, 2007a: Diurnal variations of summer precipitation over contiguous China. Geophys. Res. Lett., 34, L01704, doi:10.1029/2006GL028129.

Yu, R., Y. Xu, T. Zhou, and J. Li, 2007b: Relation between rainfall duration and diurnal variation in the warm season precipitation over central eastern China. Geophys. Res. Lett., 34, L13703, doi:10.1029/2007GL030315.

Qian, W. H., H.-S. Kang, and D.-K. Lee, 2002: Distribution of seasonal rainfall in East Asian monsoon region. Theor. Appl. Climatol., 73, 151−168, doi:10.1007/s00704-002-0679-3.

Wallace, J. M., 1975: Diurnal variations in precipitation and thunderstorm frequency over the conterminous United States. Mon. Weather Rev., 103, 406−419.

Wang, C. C., G. T. J. Chen, and R. E. Carbone, 2004: A clima-tology of warm-season cloud patterns over east Asia based on GMS infrared brightness temperature observations. Mon. Weather Rev., 132, 1606−1629.

Yang, G.-Y., and J. Slingo, 2001: The diurnal cycle in the tropics. Mon. Weather Rev., 129, 784−801.

Zhou, T., R. Yu, H. Chen, A. Dai, and Y. Pan, 2008: Summer pre-cipitation frequency, intensity, and diurnal cycle over China: a comparison of satellite data with rain gauge observations. J. Clim., 21, 3997−4010, doi:10.1175/2008JCLI2028.1.

Manuscript received 10 July 2012, accepted 1 November 2012SOLA: http://www. jstage. jst.go. jp/browse/sola