Cite this Article
ZONG Xue-Mei, WANG Pu-Cai, XIA Xiang-Ao, 2013: Variability and Long-Term Trend of Total Cloud Cover in China Derived from ISCCP, ERA-40, CRU3, and Ground Station Datasets. Atmospheric and Oceanic Science Letters, 6(3): 133-137  
Permissions
Copyright©2013, Institute of Atmospheric Physics, Chinese Academy of Sciences
Variability and Long-Term Trend of Total Cloud Cover in China Derived from ISCCP, ERA-40, CRU3, and Ground Station Datasets
ZONG Xue-Mei*, WANG Pu-Cai, XIA Xiang-Ao
Key Laboratory of the Middle Atmosphere and Global Environmental Observation (LAGEO), Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China
* Corresponding author: ZONG Xue-Mei, E-mail:zongxm@mail.iap.ac.cn
Fund:This work is supported by the “Strategic Priority Research Program” of the Chinese Academy of Sciences (XDA05100300), the National Basic Research Program of China (2013CB955801), and the National Natural Science Foundation of China (41175030).
Abstract

Total Cloud Cover (TCC) over China determined from four climate datasets including the International Satellite Cloud Climatology Project (ISCCP), the 40-year Re-Analysis Project of the European Centre for Medium-Range Weather Forecasts (ERA-40), Climate Research Unit Time Series 3.0 (CRU3), and ground station datasets are used to show spatial and temporal variation of TCC and their differences. It is demonstrated that the four datasets show similar spatial pattern and seasonal variation. The maximum value is derived from ISCCP. TCC value in North China derived from ERA-40 is 50% larger than that from the station dataset; however, the value is 50% less than that in South China. The annual TCC of ISCCP, ERA-40, and ground station datasets shows a decreasing trend during 1984-2002; however, an increasing trend is derived from CRU3. The results of this study imply remarkable differences of TCC derived from surface and satellite observations as well as model simulations. The potential effects of these differences on cloud climatology and associated climatic issues should be carefully considered.

Key words: total cloud cover; ISCCP; ERA-40; CRU3; ground station dataset
1 Introduction

Clouds are key elements in climate systems and play a significant role in the energy budget and water cycle of the earth-atmosphere system. In addition, clouds have a significant effect on the radiative energy budget. By reflecting incoming solar radiation back to space, they cool and heat the earth-atmosphere system by blocking upwelling long wave radiation from the Earth’s surface and underlying warmer atmosphere toward space. The energy budget of the earth-atmosphere system is determined by two mechanisms including the gain and loss of energy. Cloud in homogeneous radiative heating is a major driving force in atmospheric circulation that can also affect the changes of convection and precipitation. In its fourth assessment report, the Intergovernmental Panel on Climate Change ( IPCC, 2007) cloud feedback remainsan uncertain source in global climate change. Therefore, an accurate understanding of the temporal and spatial distribution of the total cloud cover is the basis of research for atmospheric radiation, energy and water cycles, atmospheric environment monitoring, and climate change ( Harrison et al. , 1990; Ronald et al. , 2007).

Cloud climatology in the pre-satellite period originates from ground observations. In addition, satellite cloud observational data have been widely used since 1980s to build cloud climate datasets, the most popular of which was established by the International Satellite Cloud Climatology Project (ISCCP). These two types of cloud climate information are widely used in various climate studies ( Rossowand Garder, 1993; Weng and Han, 1998; Kaiser, 1998, 2000; Ding et al. , 2005; Wang and Wang. , 2009). Moreover, some researchers use cloud data to study the characteristics of cloud cover distribution provided by the latest satellite instruments such as and CloudSat (Cloud Satellite) ( Liu et al. , 2009; Wang et al. , 2010); however, their short periods make these type sun suitable for climatic studies. Moreover, synthetic cloud climate information obtained by assimilation observations such European Reanalysis (ERA) and the United States National Centers for Environmental Prediction (NCEP) reanalysis is also applied in model studies ( Kalnay et al. , 1996; Jung et al. , 2006).

Many studies have been conducted on the temporal and spatial distribution of cloud cover; however, most include done or two climatic cloud datasets. The trends derived from these types of studies often differ significantly. In this paper, three global gridded datasets and one additional dataset derived from observations of more than 600 ground stations in China are used to study the temporal and spatial variations of total cloud cover over China. Trends in the past 20 years revealed by these four datasets are compared, and potential differences are emphasized. In addition, the implications of using these datasets for climate studies are discussed.

2 ISCCP, ERA-40, CRU3, and ground station datasets

The three global gridded datasets of cloud climate information used in this paper are provided by ISCCP, the 40-year Re-Analysis Project of the European Centre for Medium-Range Weather Forecasts (ECMWF; ERA-40), and the Climate Research Unit Time Series 3.0 (CRU3). One additional dataset is derived from observations of more than 600 ground stations in China.

Developed in 1982 by the World Climate Research Plan, the ISCCP has been executed for more than 20 years and is the first to provide the global systemic cloud climate information ( Rossow and Schiffer, 1999). The ISCCP climatological summary product called D2 (D2 is the climatological summary product of ISCCP. It consists of monthly averages of several properties of clouds) from July 1983 to 2008 is used in this study at a spatial resolution of 2.5°×2.5°. Available from ISCCP website, 130 parameters are used that include total cloud cover consisting of monthly cloud cover and frequency; marginal cloud cover; cloud top pressure; temperature of high, middle, and low clouds; cloud optical thickness; cloud liquid water path; and mean ozone amount.

In addition, this study used the dataset of ERA-40, which is derived from integration of ground observations, radiosonde, satellites, aircraft, and sea buoys. The technique of data assimilation is applied to form a global gridded dataset ( Uppala et al. , 2005), the resolution of which from September 1957 to August 2002 is 2.5°×2.5°. Moreover, monthly total cloud cover data of ERA-40 are used.

The CRU3 dataset, which is compiled by the Climate Research Unit of East England University in Britain, interposes observations from thousands of meteorology stations from January 1901 to June 2006 into a latitude-longitude grid with are solution of 0.5°×0.5° ( New et al. , 1999, 2000).

Ground stations data from 1954 to 2005 are provided by the Chinese meteorological data sharing center. Homogeneous testing was applied by using the RHtestV2 (Relative Homogeneity Test Version 2) algorithm ( Wang, 2007; Xia, 2010). For comparison with the CRU3, ERA- 40, and ISCCP datasets, these ground observations are gridded into the same 2.5°×2.5° resolution.

3 Comparison among ISCCP, ERA-40, CRU3, and ground station datasets
3.1 Different values of seasonal and annual total cloud cover in China

The average values of total seasonal and annual cloud cover and coefficient of variation (CV) from the four datasets are listed in Table 1. The CV is defined as the standard deviation divided by the mean in percentage. For ISCCP, the average is computed on the basis of data recorded during 1984-2002. The averages for ERA-40 and CRU3 are calculated on the basis of data recorded during 1984-2002 and during 1958-2002, respectively. Results show that the average total cloud cover of ISCCP is the largest and that the other three datasets are were all in close proximity. The difference between ISCCP and the other three datasets is 8.1% in annual, 10.4% in spring, 6% in summer, 8.8% in autumn, and 9.4% in winter; however, the seasonal variations in the four datasets are consistent such that they were larger in spring and summer and smaller in autumn and winter. Differences between 1984-2002 and 1958-2002 values are small. The largest differences, occurring in autumn, are 2.1% for ERA-40, 0.6% for CRU3, and 1.6% for the station dataset. The annual variation in coefficients from the four datasets is less than 3%, which shows that the annual variation of the total cloud cover is small. The CV values in spring and summer range from 2.2% to 4.1%, which are somewhat smaller than those in autumn and winter at 4.1%-7.5%.

The linear correlative coefficients of total annual and seasonal cloud cover from the four datasets are listed in Table 2. The asterisk indicates significance at 0.01 level. The correlative coefficients of total annual and seasonal cloud cover among ERA-40, ISCCP, and ground station datasets generally exceed 0.60. However, those between CRU3 and other three datasets are generally small; for example, the correlation between CRU3 and ISCCP is 0.25 for the annual time series of total cloud cover. It is interesting to note that although the correlation coefficient between CRU3 and the station dataset derived from 1958 to 2002 is better than that from 1984 to 2002, a contrary result is derived for ERA-40 and the station datasets.

Table1 Seasonal and annual total average cloud cover and variation coefficient from ISCCP, ERA-40, CRU3, and station datasets.
Table 2 Seasonal and annual linear correlative coefficient from ISCCP, ERA-40, CRU3, and station datasets.
3.2 Distribution of total annual mean cloud covers in China

The total annual mean cloud covers of the four datasets calculated from March 1984 to February 2002 in China, presented in Fig. 1, show a remarkable clear spatial pattern over China; the values are smaller in the northern China than those in the southern China. In the figure, contour lines represent the CRU3, ERA-40, and station datasets. The minimum difference between North China and South China is derived from ISCCP because that type indicates that the largest total cloud cover appeared in the northern China.

Overall, the total cloud cover of the four datasets can be used in climate studies. CRU3 and ISCCP datasets are more suited to usage with large-scale regional climates; ERA-40 and ground station datasets are suitable for small- and medium-scale regional climates because they provide more detailed information of total cloud cover, particularly in the northwestern China.

3.3 Monthly mean total cloud cover in China

The monthly mean total cloud covers of the four data- sets are plotted in Fig. 2. The purple line of CV indicates annual variation. The figure shows that seasonal fluctuation is remarkably high, ranging from 40% to 67% in ERA-40, CRU3, and station datasets and from 50% to 72% in ISCCP. The CV ranges from 2% to 8% in ISCCP and from 2% to 12% in the other three datasets. The monthly mean total cloud covers of all four datasets are the highest in June and July and are the lowest in November and December.

Figure 1 Distributions of annual total cloud cover in China determined by datasets from (a) ISCCP, (b) ERA-40, (c) CRU3, and (d) ground stations. AVE and CV indicate average value and coefficient of variation, respectively.
Figure 2 Monthly mean total cloud covers inChina from (a) ISCCP, (b) ERA-40, (c) CRU3, and (d) stations datasets, accompanied by relevant coefficient of variation (%).

3.4 Trend differences of total annual average cloud cover in China

The time series of total annual average cloud cover of four datasets are plotted in Fig. 3, in which dotted and dashed lines indicate trends for the period 1958-2002 and 1984-2002, respectively. The CRU3, station, and ERA-40 datasets show the same negative trend during 1958-2002 with the maximum negative trend of -0.9% per decade derived from station datasets. However, analysis of data for the period 1984-2002 differs significantly. The datasets from ISCCP, stations, and ERA-40 show a decreasing trend of similar magnitude; however, an increasing trend is derived from CRU3 data due to sudden increase in values after 1995.

Figure 3 Time series and varying trends of total annual average cloud cover determined from the four datasets. Dotted lines show trends for the period 1958-2002; dashed lines represent the period 1984-2002.

4 Conclusions

The distribution patterns of total annual mean cloud cover reported during acommon 19-year period from March 1984 to February 2002 from four cloud climate datasets in China are analyzed in this study. The relative deviations of ISCCP, ERA-40, and CRU3 from the dataset of 600 ground stations are calculated to derive the following conclusions:1) The spatial distributions of the four datasets appear in a similar pattern that increase gradually from north to south. The value of ISCCP is larger than that of the other three datasets, which is mainly attributed to the various observation methods.

2) A negative trend is derived from CRU3, station, and ERA-40 datasets reported during 1958-2002, and the largest decreasing trend is derived from the station dataset. Similar decreasing trends are derived from station, ERA-40, and ISCCP datasets; however, an increasing trend is derived from CRU3 during the same period.

3) The correlative coefficients among ISCCP, ERA-40, and station datasets are remarkably high, while those of total annual cloud cover between CRU3 and the other three datasets are lower. These results are in agreement with the changes in trends for time series.

References
1 Ding, S. G. , C. S. Zha, G. Y. Shi, 2005: Analysis of global total cloud amount variation over the past 20 years, J. Appl. Meteor. Sci. (in Chinese), 16(5), 670-677.
2 Harrison, E. F. , P. Minnis, B. R. Barkstrom, 1990: Seasonal variation of cloud radiative forcing derived from the Earth radiation budget experiment, J. Geophys. Res. , 95(D11), 18687-18703.
3 IPCC, 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge and New York, 996pp
4 Jung, T. , S. , K. Gulev, I. Rudeva, et al. , 2006: Sensitivity of extra tropical cyclone characteristics to horizontal resolution in the ECMWF model, Quart. J. Roy. Meteor. Soc. , 132, 1839-1857.
5 Kaiser, D. P. , 1998: Analysis of total cloud amount over China, Geophys. Res. Lett. , 25, 3599-3602.
6 Kaiser, D. P. , 2000: Decreasing cloudiness over China: An updated analysis examining additional variables, Geophys. Res. Lett. , 27, 2193-2196.
7 Kalnay, E. , M. Kanamitsu, R. Kistler, 1996: The NCEP/NCAR 40-year reanalysis project, Bull. Amer. Meteor. Soc. , 77, 437-471.
8 Liu, R. X. , H. B. Chen, Z. J. Zheng, 2009: Analysis and validation of total cloud amount data in China, J. Appl. Meteor. Sci. (in Chinese), 20(5), 571-578.
9 New, M. , M. Hulme, P. Jones, 1999: Representing twentieth-century space-time climate variability. Part I: Development of a 1961-1990 mean monthly terrestrial climatology, J. Climate, 12, 829-856.
10 New, M. , M. Hulme, P. Jones, 2000: Representing twentieth-century space-time climate variability. Part II: Development of a 1901-1996 monthly grids of terrestrial surface climate, J. Climate, 13, 2217-2238.
11 Ronald, M. E. , G. Ohring, F. Weng, 2007: A simulation of satellite cloud and precipitation observations in numerical weather prediction models: Introduction to the JAS special collection, J. Atmos. Sci. , 64(11), 3737-3741.
12 Rossow, W. B. , L. C. Garder, 1993: Cloud detection using satellite measurements of infrared and visible radiances for ISCCP, J. Climate, 6, 2341-2369.
13 Rossow, W. B. , R. A. Schiffer, 1999: Advances in understanding clouds from ISCCP, Bull. Amer. Meteor. Soc. , 80, 2261-2287.
14 Wang, M. Y. , B. M. Wang, 2009: Total cloud amount difference between ISCCP product and ground observation over China, J. Appl. Meteor. (in Chinese), 20(4), 411-418.
15 Wang, X. , 2007: Penalized maximal t test for detecting undocumented mean shift without trend change, J. Appl. Meteor. Climatol. , 46, 916-931, doi: 10. 1175/JAM2504. 1.
16 Wang, S. H, Z. G. Han, Z. G. Yao, 2010: Comparison of cloud amounts from ISCCP and CloudSat over China and its neighborhood, Chinese J. Atmos. Sci. (in Chinese), 34(4), 767-779.
17 Weng, D. M. , A. M. Han, 1998: Comparison between total cloudiness from satellite cloud pictures and ground observations over China, Quart. J. Appl. Meteor. , 9(1), 32-37.
18 Uppala, S. M. , P. W. K?llberg, A. J. Simmons, 2005: The ERA- 40 re-analysis, Quart. J. Roy. Meteor. Soc. , 131, 2961-3012.
19 Xia, X. , 2010: Spatiotemporal changes in sunshine duration and cloud amount as well as their relationship in China during 1954- 2005, J. Geophys. Res. , 115, D00K06, doi: 10. 1029/2009JD012879.