Opposite trends in summer precipitation in South and North Korea

Authors

  • Yeonjoo Kim,

    1. Korea Adaptation Center for Climate Change, Korea Environment Institute, Eunpyeong-gu, Seoul, Korea
    Search for more papers by this author
    • [Correction added 23 April 2012 after original online publication: the name of the first author has been written out in full as ‘Yeonjoo Kim’.]

  • B. Kang,

    1. Department of Civil and Environmental Engineering, Dankook University, Yongin-si, Gyeonggi-do, Korea
    Search for more papers by this author
  • J. M. Adams

    Corresponding author
    1. School of Biological Sciences, Seoul National University, Gwanak-gu, Seoul 151, Korea
    • School of Biological Sciences, Seoul National University, Gwanak-gu, Seoul 151, Korea.
    Search for more papers by this author

Abstract

We analysed daily precipitation data at the rain gauge stations in North Korea over a period of 25 years from 1983 to 2007, and in South Korea over a period of 35 years from 1973 and 2007. We found a striking trend of decreasing summer precipitation across North Korea. By contrast, in South Korea, the trend is opposite: there is a major increase in summer precipitation. Also, the number of dry days in summer showed an increasing trend in North Korea and a decreasing trend in South Korea. For the number of days with heavy precipitation (i.e. days with above 50 mm/day daily precipitation) during summer, a decreasing trend was detected in North Korea, but no trend in South Korea. However, in South Korea, there was a significant increase of days with heavy precipitation over the whole year. These opposite trends in summer precipitation between North and South Korea were further confirmed using four global/regional satellite and rain gauge datasets of CPC Merged Analysis of Precipitation (CMAP), the Global Precipitation Climatology Project (GPCP), Precipitation REConstruction over the Land (PREC/L), and the Asian Precipitation-Highly Resolved Observation Data Integration Towards the Evaluation of Water Resources (APHRODITE). Copyright © 2011 Royal Meteorological Society

1. Introduction

Long-term trends in precipitation have been observed in recent decades in many regions, and different trends from region to region have been reported. Increased precipitation has been observed in eastern parts of America, northern Europe, and northern and central Asia, and decreased precipitation has been recorded in the Sahel, the Mediterranean, southern Africa and parts of southern Asia. (IPCC, 2007).

For the Korean Peninsula, a number of studies have examined historical precipitation patterns and trends (Ho, 2003; Chang and Kwon, 2007; Kim et al., 2009; Jung et al., 2010). Ho et al. (2003) have showed a sudden change in summer rainfall characteristics during the late 1970s in South Korea and suggested the observed sudden increase in the geopotential height at 700 hPa over mid-latitude Asia as a possible explanation for the increasing trend in precipitation. Chang and Kwon (2007) have investigated the spatial patterns of trends in the amount and intensity of total summer precipitation and the frequency of heavy precipitation for South Korea between 1973 and 2007, showing increasing trends in the amount of precipitation during summer months. Kim et al. (2009) have applied Bayseian changepoint analysis to detect a change point in the 30-year (1976–2005) time series of the annual maximum precipitation averaged over South Korea. They found that a single change occurred around 1997 in the area-averaged annual maximum precipitation, which is consistent with an abrupt increase in the intensity and frequency of heavy precipitation after 1997. However, whilst these studies are consistent in finding the increasing trends in precipitation, they are limited to investigating precipitation in South Korea only.

In this study, we investigated precipitation data observed at rain gauge stations in both South and North Korea to understand the spatial patterning of precipitation trends over the entire Korean Peninsula. We examined the precipitation amount, number of dry days and numbers of days with heavy precipitation, comparing trends in precipitation in South and North Korea. In addition, we analysed the global/regional gridded precipitation datasets to strengthen the validity of our findings with observations at stations.

2. Methods

2.1. Rain gauge data

We examined daily precipitation data at 27 stations available for North Korea from 1983 to 2007, and at 60 stations in South Korea from 1973 to 2007. For North Korea, we used data provided by the World Meteorological Organization (WMO) to the Korea Meteorological Administration (KMA). For South Korea, the KMA has maintained the daily precipitation data of observatory stations since the early 1970s. Among the 76 stations run by the KMA, we chose 60 stations to exclude those stations which are missing more than 3 days per year and the stations established later than 1973. With regard to the quality of data in South Korea, Chang and Kwon (2007) analysed the station data with Worsley likelihood ratio, cumulative deviations, and the Bayesian procedure, with the result that only a few stations exhibited homogeneity with significance slightly less than 95% confidence level by one or two criteria.

We calculated the total precipitation (Ptot), the number of days without precipitation (Ndry), and the number of days in which the daily precipitation is greater than 50 mm (number of days with the heavy precipitation: Nheavy) for each season, i.e. for the spring with March, April and May (labelled MAM); the monsoon summer with June, July and August (JJA); the fall with September, October and November (SON); the winter with December, January and February (DJF). These three indices are calculated in the station level as well as in the region level.

To detect positive or negative trends for the three indices (Ptot, Ndry, and Nheavy), the non-parametric Kendall's τ test and linear regression are used here. Kendall's τ coefficient is a measure of rank correlation, i.e. the similarity of the orderings of the data when ranked by each of the quantities (Gilbert, 1987). It ranges from − 1 (−100%) to + 1 (+100%) for negative association (i.e. decreasing trend) and positive association (i.e. increasing trend). Here we use τB, which makes adjustment for ties (i.e. tied pair). This test has been applied to other studies examining hydroclimatic trends (Chang and Kwon, 2007; George, 2007). In addition, the slope of linear regression represents the steepness of the linear line, estimated with the least square approach.

2.2. Gridded precipitation data

To confirm the validity of our findings with observations at stations, we used four gridded precipitation datasets. They include two monthly global datasets of the CPC Merged Analysis of Precipitation (CMAP) and the Global Precipitation Climatology Project (GPCP), which are the merged products of gauge observations and various satellite estimates, and two land-only datasets of the Precipitation REConstruction over the Land (PREC/L) and the Asian Precipitation-Highly Resolved Observation Data Integration Towards the Evaluation of Water Resources (APHRODITE), which are based on rain gauge observations. Below, we provide a brief description of each dataset.

The CMAP dataset (Xie and Arkin, 1997) was created globally on a 2.5° latitude–longitude grid from 1979 to near the present by merging several kinds of data sources. Data sources included gauge observations, estimates from satellite observations of infrared (IR), outgoing longwave radiation (OLR), Microwave Sounding Unit (MSU), and microwave (MW) scattering and emission from the Special Sensor Microwave Imager (SSM/I), and precipitation forecast by NCEP/NCAR Reanalysis.

The GPCP monthly precipitation analysis (Adler et al., 2003) was constructed at 2.5° latitude–longitude resolution from 1979 to near the present by combining the precipitation information from multiple sources. SSM/I data, IR estimates were merged, Television and Infrared Observation Satellite (TIROS) Operational Vertical Sounder (TOVS), OLR measurements, and quality-controlled rain gauge data were merged based on hierarchical relations. Much of input data of GPCP and CMAP are in common, particularly over the land, but the merging algorithms of two datasets are different (Yin et al., 2004).

The PREC/L dataset (Chen et al., 2002) was constructed globally over land on 2.5°, 1.0°, and 0.5° latitude–longitude grids from 1948 to the present. The data were derived from gauge observations collected in the Global Historical Climatology Network (GHCN), and the Climate Anomaly Monitoring System (CAMS) dataset, using the Optimal Interpolation (OI) technique.

The APHRODITE dataset for 1958–2007 (Yatagai et al., 2009) was created on 0.50° and 0.25° latitude–longitude grids in a daily time step by collecting gauge observations over three different regions, including Monsoon Asia, Russia, and the Middle East. For this dataset, three categories of data were used: datasets from the Global Telecommunication System (GTS) network, precompiled datasets, and individual data collected by APHRODITE project. Here, the GTS over the entire globe was also used for the above-mentioned global datasets of CMAP, GPCP and PREC/L. The pre-compiled datasets include different sources such GHCN-Daily, GWEX Asian Monsoon Experiment.

In this study, we evaluated the seasonal precipitation (Ptot) for gridded datasets. While some of the datasets extended back to 1948, here we analysed the data from 1983 to 2007 except for PREC/L (1983 ∼ 2006), to be consistent with the rain gauge station data. We chose to use the data at the finest grids when multiple datasets were available, and for PREC/L and APHRODIE we used the data at a 0.5° and 0.25° grid, respectively. PREC/L data at the 0.5° grid resolution were available only up to July of 2007.

3. Results

3.1. Amount of precipitation

Annual precipitation averaged over the 27 stations in North Korea (1983 ∼ 2007) shows a decreasing trend with the linear regression slope of − 9.3 mm/yr and the Kendall's τB of − 24.7% (Table I and Figure 1). The decreasing trends are pronounced in spring and summer precipitation with the linear regression slope of − 5.3 and − 16.1 mm/yr, and Kendall's τB of − 58.7 and − 52.7%, respectively (Figure 2 for summer; not showed for spring). In contrast, annual precipitation averaged over the 60 stations in South Korea (1973 ∼ 2007) shows an increasing trend with the linear regression slope of 8.5 mm/yr and the Kendall's τB of 23.4%. Such an increasing trend is mostly attributable to summer precipitation, showing a similar increasing trend with the linear regression slope of 7.3 mm/yr and the Kendall's τB of 29.1%. However, unlike North Korea, the spring precipitation in South Korea did not show any significant trends.

Figure 1.

Annual precipitation across (a) North Korea (1983 ∼ 2007), and (b) South Korea (1973 ∼ 2007) (bar graph), the linear regression line of summer precipitation (solid line), and the average line (dashed line)

Figure 2.

Summer precipitation for JJA (June, July and August) (a) across North Korea (1983 ∼ 2007), and (b) South Korea (1973 ∼ 2007) (bar graph), the linear regression line of summer precipitation (solid line), and the average line (dashed line)

Table I. Kendall τB and slope from the linear regression for precipitation data in South and North Korea
 Kendall τBSlope from the linear regression
 Ptot (%)Ndry (%)Nheavy (%)Ptot (mm/yr)Ndry (#/yr)Nheavy (#/yr)
  • a

    Statistically significant trends of Kendall τB at the 90% confidence level.

North Korea (1983–2007)Annual− 24.7a47.3a− 15.0− 9.31.31− 0.023
 MAM− 58.7a65.1a− 41.2a− 5.30.56− 0.013
 JJA− 52.7a52.2a− 45.4a− 16.10.75− 0.015
 SON15.30.330.1a11.1− 0.170.017
 DJF7.319.823.10.90.170.003
South Korea (1973–2007)Annual23.4a1.231.6a8.50.030.069
 MAM− 2.93.4− 4.4− 0.70.01− 0.002
 JJA29.1a− 14.821.77.3− 0.120.001
 SON12.34.08.81.90.050.003
 DJF− 3.918.316.5− 0.20.090.001

Kendall's τB of precipitation at each rain gauge station clearly presents contrasting trends between South and North Korea (Figure 3). At most stations in North Korea, the statistically significant and decreasing trends (i.e. negative τB) are observed in spring and summer (27 and 26 among 27 stations in MAM and JJA, respectively). In South Korea, 56 among 60 stations show the increasing trends (i.e. positive τB), which are statistically significant at 18 among those 56 stations in summer. We found that those 18 stations are located in the northern part of South Korea, in general. In addition, we also found increasing trends in the fall precipitation (SON) at most stations in North Korea although only a few stations show statistically significant trends.

Figure 3.

Kendall's τB for seasonal precipitation during (a) Spring (MAM), (b) Summer (JJA), (c) Fall (SON), and (d) Winter (DJF) in each station (1983–2007). Statistically significant trends at the 90% confidence level are represented with filled circles, and otherwise with open circles

Figure 4 presents the Kendall's τB for the summer precipitation (JJA) in CMAP, GPCP, PREC/L and APHRODITE data. All four datasets reveal decreasing trends in the northern part of the Korean Peninsula and the increasing trends in the southern part. In addition, three datasets except PREC/L show the transition between the decreasing and increasing trends around the border between South and North Korea. The magnitude of Kendall's τB, ranging from ∼− 30 to ∼30%, is relatively small compared to that from the rain gauge data. For the other seasons, any dominant trends in either North or South Korea are detected (not shown).

Figure 4.

Kendall's τB for summer precipitation during JJA in gridded datasets of (a) CMAP (at a 2.5° latitude/longitude grid), (b) GPCP (at a 2.5° grid), (c) PREC/L (at a 0.5° grid), and (d) APHRODITE (at a 0.25° grid). CMAP, GPCP, and APHRODITE data are from 1983 to 2007, and PREC/L data are from 1983 to 2006. Grids with statistically significant trends at the 90% confidence level are represented with crosses, and otherwise with dots

3.2. Number of dry days

The number of dry days (i.e. number of days without precipitation) also shows contrasting trends between South and North Korea, in the summer (Table I and Figure 5). The number of dry days in North Korea shows an increasing trend, with the linear regression slope of 0.75 #/yr and the Kendall's τB of 52.2% between 1983 and 2007; in South Korea, the decreasing trends with the linear regression slope of − 0.12 (−0.20) #/yr and the Kendall's τB of − 14.8 (−23.0) % between 1973 and 2007 (1983 and 2007).

Figure 5.

Kendall's τB for the number of dry days during (a) Spring (MAM), (b) Summer (JJA), (c) Fall (SON), and (d) Winter (DJF) in each station (1983–2007). Statistically significant trends at the 90% confidence level are represented with filled circles, and otherwise with open circles

Examining the different seasons in Figure 5, we found that in North Korea the number of dry days has been increasing markedly in the spring and summer, and to a lesser extent in the winter. Dry days show decreasing trends in the summer and increasing trends in the winter in South Korea. In North Korea, most of the stations (all, except one station in the spring) exhibit statistically significant increasing trends in the spring and summer.

3.3. Number of days with heavy precipitation

Examining the number of heavy precipitation (here the heavy precipitation is defined as the daily precipitation with above 50 mm/day) (Table I and Figure 6), we found no significant annual trends in North Korea. We, however, found the decreasing trends clearly with the Kendall's τB of − 41.2 and − 45.4% in the spring and summer, respectively, and the increasing trends with the τB of 30.1% in the fall. In South Korea, the annual number of heavy precipitation days has increased with the Kendall's τB of 31.6% (22.0%) from 1973 (1983) to 2007 and such increasing trends are found in 54 among 60 stations in South Korea. Here, only the annual trend is significant, while none of seasonal trends are significant.

Figure 6.

Kendall's τB for the number of days with the heavy precipitation during (a) Spring (MAM), (b) Summer (JJA), (c) Fall (SON), and (d) Winter (DJF) in each station (1983–2007). Statistically significant trends at the 90% confidence level are represented with filled circles, and otherwise with open circles

4. Discussion

While a trend of increasing summer precipitation in South Korea has been reported in previous studies, the present study is the only one which has presented opposite precipitation trends in both South and North Korea, focusing on the station data and country averages. The plight of North Korea, beset by repeated famines and persistent food shortages, is well known (Haggard and Noland, 2005). In this context, a striking trend in precipitation is of great humanitarian and geopolitical importance.

4.1. Reliability of the trends

Could these trends be the spurious result of changes in sampling methodology, or changes in the degree of attention to sampling? In South Korea, sampling methodology has remained highly standardized and the data are likely reliable. In North Korea, any change in methodology or reliability in sampling would be expected to affect winter and autumn recorded precipitation as much as spring and summer figures; yet there is no trend in winter or autumn precipitation in North Korea during the same time period. It is also striking that all rain gauge stations in North Korea show the same trend, whereas a trend towards patchy sampling would be expected to result in different stations showing divergent trends. All this suggests that the trend in precipitation seen in the North is also likely to be genuine.

Furthermore, our analysis using multiple global and regional gridded datasets, including satellite-based estimates as well as gauge-based observations, consistently present the same opposite trends in summer precipitation. The results with the gridded datasets mimic our results with station data, including the decreasing and increasing trends in the northern and southern part of the Korean Peninsula, respectively, as well as (to some extent) the location of the transition. These gridded datasets are constructed with multiple sources of data, including some of the rain gauge data we used in this study. However, it should be noted that different data-treatment approaches are used, and that the satellite-based products (CMAP and GPCP) do merge multiple data sources with different characteristics; the satellite-based data are combined to construct the global datasets, and gauge data are used to validate them only thereafter, and thus, they are to some extent independent of the rain gauge data.

However, it is noted that significant decreasing trends in the spring (rather than summer) in North Korea are detected only in rain gauge data, not in the gridded datasets. One may surmise that the magnitude of spring precipitation is relatively small compared to that of summer monsoon precipitation in the region and it is therefore more difficult to detect its trend at such a coarse spatial resolution of gridded data. Indeed, the trends in summer precipitation are weaker in the gridded datasets than in the rain gauge data. However, such details are not investigated in this study and should be subject to further research.

4.2. Causes of the trends

Figure 7 shows that the decreasing trend of summer precipitation in the northern part of the Korean Peninsula extends to northeastern China, and the increasing trend in the southern part of the Korean Peninsula extends to southern China. This suggests that the opposite trends in the Korean Peninsula may be merely the part of large-scale trends across East Asia. In this context, several hypotheses have been investigated to reveal the causes of observed trends. Large-scale atmospheric phenomena have been investigated with focus on the location of Meiyu front in China, which is the Chagma front in Korea (e.g. Yu et al., 2004; Hirota et al., 2005). Also, the impact of anthropogenic changes on climate, such as the impact of black carbon aerosols (Surabi M et al., 2002) as well as land cover-climate interactions (Zhang et al., 2009), have been suggested.

Figure 7.

Kendall's τB for summer precipitation during JJA in CMAP dataset from 1983 to 2006 at a 2.5° latitude/longitude grid over East Asia (100 ∼ 150°E and 20 ∼ 50°N). Grids with statistically significant trends at the 90% confidence level are represented with crosses, and otherwise with dots

Yu et al. (2004) have suggested a distinctive strong Tropospheric Cooling Trend (TCT) in East Asia during July and August as a cause of increased droughts in northern China and floods in the Yangtze River Valley, and showed observational evidence that the TCT is linked to the stratosphere temperature changes and the interaction between the troposphere and stratosphere. Accompanying the TCT, the upper level westerly jet stream over East Asia shifts southward and the East Asian summer monsoon weakens, which in turn leads to observed contrasts between northern China and the Yangtze River Valley. Surabi M et al. (2002) have investigated possible aerosol contributions to the observed trends in China and India. Using a global climate model, they found that the precipitation and temperature changes were comparable to those observed only if the aerosols include a large portion of absorbing black carbon. They then argued that absorbing aerosols heated the air and affected the large-scale circulation and hydrologic cycle.

Although, as suggested above, the trends in the Korean Peninsula might be explained in the context of large-scale trends, it is still interesting to consider why we find the transition between the opposite trends around the border between South and North Korea in multiple precipitation datasets. One might hypothesize that a sudden abrupt barrier in the land cover would be able to determine the position of the transition. Whereas South Korea's forests have been extensively replanted and become larger and leafier each year, North Korea is suffering from severe ongoing deforestation (Lee et al., 1997; Youn, 2009). Possible causes of changes in precipitation change could be alteration of surface albedo, roughness or evapotranspiration (Foley et al., 2005). Clearly, this is a matter which requires regional climate modelling studies which include vegetation–climate interactions.

Acknowledgements

This study was supported by funding from the National Institute of the Environmental Research to B. Kang, and from the Korea Environment Institute (BA2010-04) to Y. Kim. The authors would like to acknowledge the Korea Meteorological Administration (KMA) and World Meteorological Organization (WMO) for providing the daily precipitation data at the stations. The authors would also like to acknowledge the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA, for providing CMAP, GPCP and PREC/L global monthly precipitation data at their Web site at http://www.esrl.noaa.gov/psd, and the APHRODITE's project team for providing the daily precipitation data over Monsoon Asia at the their Web site http://www.chikyu.ac.jp/precip.

Ancillary