SCIENCE CHINA Earth Sciences • RESEARCH PAPER •

November 2010 Vol.53 No.11: 1734–1746 doi: 10.1007/s11430-010-4024-x

An East Asian land-sea atmospheric heat source difference index and its relation to general circulation and summer rainfall over China ZHANG Bo1*, ZHOU XiuJi1, CHEN LongXun1, ZHU YanFeng2 & ZHAO Bin3 1

Chinese Academy of Meteorological Sciences, Beijing 100081, China; 2 National Climate Center, Beijing 100081, China; 3 National Meteorological Center, Beijing 100081, China

Received June 15, 2009; accepted February 5, 2010; published online September 29, 2010

Using a monthly precipitation dataset of 160 stations over China and a daily and monthly National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis dataset from 1961 to 2006, we here define an East Asian land-sea atmospheric heat source difference index ILSQD and investigate its relationship to summer rainfall in China and East Asian general circulation. The results show that ILSQD more closely reflects the anomalous variations in summer monsoon phenomena; in the high-index (HI) cases, the strong low-level southerlies over East China and the strong high-level westerlies over middle latitudes indicate an active summer monsoon, and vice versa in the low-index (LI) cases. This index also reflects summer rainfall anomalies over East China; in the HI (LI) cases rainfall increases (decreases) over North China and at the same time decreases (increases) over the mid-lower Yangtze River valley and the southern Yangtze River. Hence, ILSQD can be utilized as a summer monsoon index. There is also remarkable correlation between ILSQD in March and the following summer rainfall over the mid-lower Yangtze River valley. Finally, the Community Atmospheric Model Version 3.1 (CAM3.1) of NCAR is used to run numerical experiments, which verify that the anomalous summer precipitation in simulations is similar to that of diagnosis analysis based on the anomalous summer atmospheric heating forcing. Similarly, the atmospheric heating rate in March can force summer rainfall anomalies in the simulations just as observed in the data. land-sea atmospheric heat source difference index, East Asian summer monsoon, summer rainfall, forecasting meaning Citation:

Zhang B, Zhou X J, Chen L X, et al. An East Asian land-sea atmospheric heat source difference index and its relation to general circulation and summer rainfall over China. Sci China Earth Sci, 2010, 53: 1734–1746, doi: 10.1007/s11430-010-4024-x

Since the 1980s, Chinese meteorologists [1–4] have defined the East Asian monsoon systems, which include the Australian winter monsoon in the Southern Hemisphere, the South China Sea (SCS) and western Pacific tropical monsoon, and the mainland China and western Pacific subtropical monsoon. To depict the interannual and interdecadal variations of the East Asian summer monsoon (EASM), investigators have sought various EASM indices. Many have investigated *Corresponding author (email: [email protected])

© Science China Press and Springer-Verlag Berlin Heidelberg 2010

the impact of the East Asian (EA) subtropical monsoon intensity on rainfall and the rain belt over eastern China. For the Indian summer monsoon (ISM), the Indian rainfall index and the Indian wind index have generally been accepted. However, quantifying the EASM variability is much more difficult, and its indices need to be especially defined. The different EASM indices have been defined according to sea level pressure (SLP), 850-hPa vorticity, and high-level and low-level wind fields, among other factors. The 25 existing circulation indices have been compared and analyzed in the literature [5]. Guo [6] purported earth.scichina.com

www.springerlink.com

ZHANG Bo, et al.

Sci China Earth Sci

that the east-west thermal contrast plays an important role in the formation of the EA subtropical monsoon, and that the intensity of the summer monsoon is determined by summing pressure difference between 110°E (land) and 160°E (sea). Her index was subsequently modified by Shi et al. [7]. Zhao and Zhou [8] defined a simple index of the EA subtropical summer monsoon according to the SLP difference between a land region over Mongolia and an oceanic region over the western Pacific. Many of the EASM indices are constructed according to the 850-hPa wind fields over the EA subtropics [9–15]. According to the temperature difference between a land region over the EA monsoon region and an oceanic region over the subtropical northwestern Pacific, Sun et al. [16] defined a complicated land-sea thermal difference index that includes not only zonal differences between surface temperature over East Asia and sea surface temperature (SST) over the western Pacific but also the meridional difference between surface temperature over South China and SST over SCS. The resulting index [16] is consistent with the anomalous variations in precipitation over the Yangtze River. Tang et al. [17] found that the correlation between rainfall and the index defined by land surface temperature and SST is the most significant in contrast to the relationships between the summer rainfall and the EASM indices defined by Huang and Yan [18], Zhang et al. [13], Sun et al. [16] and Guo [6]. However, there are data issues because surface temperature datasets update relatively slowly, and the time scale of the timely SST datasets is short. Furthermore, the land-sea thermal difference includes not only the thermal difference at the surface but also the thermal contrast in the atmosphere. Based on these analyses, the following question has emerged: can we define a thermal difference index based on atmospheric heat source (AHS)? The impact of AHS anomalies over the eastern Tibetan Plateau (TP) on precipitation and general circulation has been studied extensively [19–24]. Huang and Yan [25] noted that diabatic heating over the TP plays an important role in the formation and perpetuation of the Asian summer monsoon. Wu and Zhang [26] showed that thermal forcing and mechanical forcing are important factors in causing the periodical and regional occurrence of the Asian monsoon. Gong et al. [27] indicated that the difference in heat source LFO characteristics over the Asian monsoon area can lead to the abnormal summer drought/flood in the YangtzeHuaihe River basin. He et al. [28] concluded that in late March, the thermal conditions of the eastern Asian landmass change from cold to heat sources for the atmosphere, emphasizing the significant impact of this zonal thermal reversal on the EA subtropical monsoon. Using the T42L9 atmospheric circulation model, Ma and Sun [29] studied the characteristics of the persistent anomaly of the SCS summer monsoon, along with its influence on global atmospheric circulation; the results of the numerical simulation are nearly identical to those in their diagnostic analyses [30].

1735

November (2010) Vol.53 No.11

Based on the important role of the AHS in the Asian monsoon, it is important to define the EASM index using the AHS. We will, therefore, define an East Asian land-sea heat source difference index and investigate its correlation with the East Asian general circulation and summer rainfall in China.

1 1.1

Data and methodology Data

The data used in this study include the following sources: (1) the 1961–2006 daily and monthly mean reanalysis data provided by NCEP/NCAR (version 1), including wind, temperature, and geopotential height fields at 17 levels; besides, the velocity field at 12 levels and specific humidity at 8 levels at 2.5°×2.5° resolution; (2) the 1961–2006 monthly precipitation data of 160 stations over China. 1.2

Methodology

From the thermodynamic equation, AHS can be denoted as k

⎛ P ⎞ ⎛ ∂θ ∂θ ⎞ + V ⋅ ∇θ + ω Q1 = C p ⎜ ⎟ ⎜ , ∂P ⎟⎠ ⎝ P0 ⎠ ⎝ ∂t

(1)

where P0=1000 hPa, θ is the potential temperature, and ω is vertical velocity [31]. Q1 =

1 Ps Q1dp, g ∫ PT

(2)

where Ps and PT are pressures at surface and at top (100 hPa) respectively, and Q1 is the vertical integration of Q1 within a unit of air column. Here, the summer refers to the period from June to August.

2 Definition of the East Asian land-sea heat source difference index To consider the climatological features of AHS, Figure 1 is included to show the distribution of AHS differences between July and January. Positive differences are found over Eurasia, the northern Indian Ocean, the Bay of Bengal (BOB), and the intertropical convergence zone (ITCZ) over the Northern Hemisphere; this means that the AHS intensity over these regions in July is stronger than in January. The positive difference centers are over the northern coast of the BOB, the central BOB, the western coast of the Indian Peninsula (IP), and eastern China. Incidentally, the major AHS

1736

ZHANG Bo, et al.

Sci China Earth Sci

Figure 1 The distribution of AHS differences between July and January during 1961–2006. Dashed lines denote the topographic contour of 3000 m, shaded areas denote the positive difference regions, unit: W·m−2.

in summer season over the Northern Hemisphere appears over the above areas. There is a negative AHS difference over the western North Pacific and in the tropics and subtropics over the Southern Hemisphere, which suggests that the AHS intensity over these regions in July is weaker than in January. The negative AHS centers are located over the western Pacific to the east of Japan, northern Australia, and the western part of the southern Indian Ocean. It is clear that the AHS differences between summer and winter over the EA continent are nearly opposite to those over the adjacent western North Pacific, which means that the zonal thermal difference reverses from winter to summer. The climatological pentad-varying zonal AHS difference averaged over 25°–45°N (defined as the discrepancy between the AHS at a given longitude and the mean AHS averaged over 90°–160°E) is shown in Figure 2(a). The seasonal reversal of the zonal AHS difference is obvious, with the negative difference clearly located over the western Pacific from May to August and with the negative center in the western Pacific to the east of 140°E, and with positive differences over these regions during the other months and a positive maxima appearing in winter. Over EA there is a negative difference from October to March, and a positive difference centered over 110°–120°E (eastern China) and the regions to the west of 100°E during the other months. In other words, the prominent feature of this data is a remarkable seasonal reversal of the AHS over East Asia and the western Pacific, with the AHS intensity over the East Asian continent much stronger than that over the western Pacific in the summer, and vice versa in winter. Hence, in this study, a land-sea AHS difference index between the area over East Asia and the area over the western Pacific (ILSQD, Figure 2(b) shows the key region) is defined as I LSQD = Q1(25D− 45D N,90D − 120D E) − (a × Q1(27.5D − 40D N,122.5D − 130D E) + b × Q1(27.5D − 45D N,130D − 160D E) ) / c,

c = a + b,

(3) (4)

November (2010) Vol.53 No.11

where a and b are the number of grid points in the region (27.5°–40°N, 122.5°–130°E) and the region (27.5°–45°N, 130°–160°E), a=24, and b=104. The first and second term on the right side of eq. (3) represent the averaged AHS over EA (25°–45°N, 90°–120°E) and the western Pacific (27.5°–40°N, 122.5°–130°E; 27.5°–45°N, 130°–160°E), respectively. Figure 2(c) shows the series of zonal AHS deviations over East Asia (25°–45°N, 90°–120°E) and its adjacent western Pacific (27.5°–40°N, 122.5°–130°E) and (27.5°– 45°N, 130°–160°E). It is clear in the figure that the pattern of negative deviation over land and positive deviation over sea reverses around pentad 22, and is restored around pentad 50. This seasonal reversal of the East Asian land-sea AHS contrast has been confirmed. He et al. [28] reported that the subtropical zonal thermal contrast reverses in late March. Because the key regions of this study are different from those of He et al. [28], the reversal time of this study is not identical. The land-sea thermal contrast will affect the monsoon intensity. To illustrate this, Figure 3(a) shows the normalized ILSQD from 1961 to 2006. Positive values of the normalized ILSQD correspond to the strong AHS contrast between East Asia and its adjacent western Pacific; negative values correspond to the weak AHS contrast. In Figure 3(a), the normalized ILSQD is characterized by positive values from the 1960s to the middle and late 1970s, and again in the late 1980s, which correspond to the strong EASM. Negative values of the normalized ILSQD indicate a weak EASM in the early 1980s and from the 1990s through to the present. The summer AHS series over East Asia and the western Pacific as well as the AHS difference series between these two regions indicate that the land AHS experiences a decreasing trend with corresponding equation as Y=−0.27X+100.05. The sea AHS experiences an increasing trend with corresponding equation as Y=0.39X+22.45, and the AHS difference between land and sea has a clearly decreasing trend represented with corresponding equation as Y=−0.66X+77.59. Therefore, ILSQD shows an obvious decreasing trend under the common influence of the AHS over land and over sea. Figure 3(b) shows the series of the summer normalized meridional and zonal wind velocity over the region (25°–45°N, 110°–120°E), and clearly meridional wind velocity exhibits a strongly decreasing trend. The positive value prior to 1976 indicates heavy southerlies over the eastern EA; after 1976, the value of meridional wind velocity becomes negative, meaning the intensity of the southerlies weakened. The correlation coefficient between normalized ILSQD and normalized meridional wind has a value of 0.41, which passes test at 95% confidence level. With regard to the zonal wind velocity, there is a remarkable decreasing trend from the 1960s to the mid-1970s and during the early part of the 21st century, and a stable characteristic from the mid-1970s to the end of the 20th century; therefore, the zonal wind does not significantly decrease. The correlation coefficient between normalized ILSQD and

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

1737

Figure 2 The climatological pentad-varying zonal AHS difference averaged over 25°–45°N (unit: W·m−2) (a), the key region (b) and the series of zonal AHS deviations over East Asia (25°–45°N, 90°–120°E) and its adjacent western Pacific (c). (c) Solid line: AHS over land; dashed line: AHS over ocean; unit: W·m−2.

3 Relationship of ILSQD with summer general circulation over East Asia To investigate the relationship between the summer ILSQD and the summer wind field, a correlation vector is defined as G G G R = RI LSQD & u i + RI LSQD & v j , (5) where RI LSQD & u and RI LSQD & v are correlation coefficients

Figure 3 The series of the summer normalized ILSQD during 1961–2006 (a) and the series of the summer normalized meridional and zonal wind velocity over the region (25°–45°N, 110°–120°E) (b). (b) Solid line: zonal wind velocity; dashed line: meridional wind velocity.

normalized zonal wind velocity, at a value of 0.26, exceeds test at 90% confidence level. The following questions have subsequently been raised: what is the relationship of ILSQD to EA general circulation and summer rainfall over China? Can the value of ILSQD reflect the intensity of EASM?

between ILSQD and the zonal and meridional wind fields, respectively. Figure 4 shows the correlation vector distribution between ILSQD and the 850-hPa and 200-hPa wind fields in the summer season. Figure 4(a) shows clearly that a strong low-level cyclone exists around the TP over East Asia, and that the westerlies from the Indian Ocean combine with the southerlies from the equator over the eastern BOB and turn into southwesterly winds, passing through the Indo-China Peninsula and flowing into western China. A low-level anticyclone covers the western Pacific to east of EA; the southwesterly winds over the western anticyclone control East China and combine with the southwesterlies from the Indo-China Peninsula over North China. These data indicate that during the period of high ILSQD values, the southwesterly flows cover most of East China, extending northward to the latitude around 50°N. In addition, a cyclone exists over China-Mongolia, where the westerly winds to the south of this area combine with the southwesterly flows from East China over the eastern part of Northwest China. At 200 hPa (Figure 4(b)), two high-level anticyclone circulations appear over the region from western TP to Iran and Afghanistan, and the region from Northwest

1738

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

Figure 4 The correlation vector distribution between ILSQD and the 850-hPa wind field (a), (b) same as (a) but for 200 hPa, and (c) same as (a) but for the whole column moisture flux. Long dashed line: the topographic contour of 3000 m, in shaded areas correlation coefficients are statistically significant at 90 % confidence level.

China to Japan, respectively. The center of the latter highlevel anticyclone is located westward about 10 degrees of the low-level anticyclone; high-level southwesterly winds control North and Northwest China. The above analysis shows a remarkable correlation between ILSQD and the highlevel and low-level wind fields in the summer over EA. When the value of ILSQD is high (low), the intensity of southwesterly winds over East China is strong (weak), the summer monsoon extends northward more easily (with more difficulty). Figure 4(c) shows that the southerlies from the intense (weak) anticyclone over the western Pacific will bring more (less) moisture into China, and that summer rainfall increases (decreases) over North China whereas drought (flood) may occur over the Yangtze-Huaihe River basin and mid-lower reaches of the Yangtze River. Further studies on these phenomena will be carried out in the future using composite analysis. According to the EA land-sea AHS difference index ILSQD, the eight strongest-index years from 1961 to 2006 (called the high-index (HI) cases) occurred in 1963, 1964, 1971, 1973, 1976, 1984, 1986, and 1990; the eight weakest-index years (called the low-index (LI) cases) occurred in 1972, 1993, 1998, 1999, 2000, 2001, 2002, and 2006. Figure 5 shows the distributions of the high-level, mid-level, and low-level composite wind anomalies in the HI and LI cases. In Figure 5(a), anticyclone anomalies are shown over the Korean Peninsula (KP), Japan, and the western North Pacific, meaning that the western Pacific subtropical high (WPSH) is located much further north than its normal position. At the same time, the southwesterly anomalies from the Indian Ocean pass through IP, the BOB, and the Indo-China Peninsula, then combine with the southeasterly

anomalies from the western Pacific over Southeast China and extend northward to around latitude 55°N. Most of Northeast China and KP are controlled by the southwesterly anomalies; furthermore, there are cyclone anomalies over China-Mongolia that strengthen the southerly flows over the monsoon region, bringing additional moisture to the midhigh latitudes, and increasing rainfall over North China. Figure 5(b) shows that anticyclone anomalies are located over China-Mongolia; cyclone anomalies are found over Northeast China, Japan, and its adjacent western Pacific, which indicates that the WPSH is anomalously in the south. The northerly anomalies pass through North, East, and South China; a small portion of them then turns to westerly anomalies and flows into the western Pacific through the Taiwan strait, while most of northerly anomalies continue to move southward along the Indo-China Peninsula and ultimately arrive in the BOB. The direction of northerly anomalies over East China is nearly opposite to that of southerly winds in the summer over China. These circulation anomalies are unfavorable for northward movement of southerly flows in the East Asian monsoon region, weakening the EASM intensity, and increasing rainfall in the valley of the Yangtze River while decreasing rainfall in North China. Figure 5(c) shows that at 500 hPa in the HI cases, anticyclone anomalies are located in East China, KP, Japan, and the eastern area of EA, meaning that the WPSH and westerlies move northward anomalously. These circulation anomalies are favorable for the northward movement of the summer monsoon over EA and rainfall belt in China. Meanwhile, the single summer monsoon over the mid-lower reaches of the Yangtze River causes drought over this re-

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

1739

Figure 5 The distributions of summer composite wind anomalies at 850 hPa in the HI cases (a); at 850 hPa in the LI cases (b); at 500 hPa in the HI cases (c); at 500 hPa in the LI cased (d); at 200 hPa in the HI cases (e) and at 200 hPa in the LI cases (f). Long dashed line: the topographic contour of 3000 m, in shaded areas correlation coefficients are statistically significant at 90% confidence level, unit: m/s.

gion (Figure 5(a)). In addition, cyclone anomalies cover the region from south of Lake Baikal to northeastern TP, corresponding to a trough over this region. The distribution of the 500-hPa general circulation anomalies in the HI cases is similar to that of the first abnormal circulation shown by Zhao and Song [32]; again, summer rainfall increases (decreases) over North China, while draught (flood) may occur over the Yangtze-Huaihe River basin. In the LI cases (Figure 5(d)), the opposite general circulation distribution indicates that cyclone anomalies exist over East China, KP and the western Pacific and that two anomalous anticyclone centers exist over the western IP and China-Mongolia. The circulation anomalies are unfavorable for northward movement of summer monsoons (Figure 5(b)), and they cause less summer rainfall over North China and more rainfall over the Yangtze-Huaihe River basin. Figure 5(e) shows that at 200 hPa in the HI cases, two anomalous strong anticyclone centers are located over the region from the western TP to Iran and Afghanistan and the region around northeastern China and Japan, respectively. The relevant phenomena are the westerly anomalies over most of the East Asian middle latitudes, the easterly anomalies over the subtropics, and the anticyclone anomalies in the mid-high lati-

tudes, all of which indicate that this distribution of general circulation is consistent with that of the 200-hPa circulation anomalies that cause summer rainfall increases over North China [33]. Figure 5(f) shows that in the LI cases, cyclone anomalies are located over the western TP, northeastern China, and Japan. Whereas easterly anomalies cover most of the middle latitudes in Asia, and the westerly anomalies occur to the west of 80°E and east of 110°E in the subtropics. Figure 6 shows the composite geopotential height fields in the HI and LI cases. In the HI cases, the contour 5880 gpm at 500 hPa does not appear, the west end of the contour 5860 gpm is located around 120°E (Figure 6(a)), and the east end of the contour 12520 gpm at 200 hPa is located around 100°E (Figure 6(c)). In the LI cases, the west end of the contour 5880 (5860) gpm at 500 hPa is located around 135°E (112°) (Figure 6(b)), and the east end of the contour 12520 gpm at 200 hPa is located around 110°E (Figure 6(d)). Furthermore, the areas encircled by the contours 5880 gpm and 12520 gpm in the LI cases are larger than those in the HI cases. The composite analysis indicates that the land-sea AHS contrast between East Asia and the western Pacific plays an important role in the east-west movement

1740

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

Figure 6 The summer composite geopotential height fields at 500 hPa in the HI cases (a); at 500 hPa in the LI cases (b); at 200 hPa in the HI cases (c) and at 200 hPa in the LI cases (d). (d) shaded area: the topographic contour of 3000 m, unit: m/s.

of the WPSH and the South Asia High. In the HI cases, the South Asia High moves to the west and the WPSH moves to the east, inducing summer rainfall to increase over North China and decrease over the Yangtze valley. The general circulation is the opposite in the LI cases; the South Asia High moves to the east and the WPSH moves to the west, inducing summer rainfall to decrease over North China and increase over the Yangtze valley [32, 34]. In summary, according to the analyses above, ILSQD captures the variability of summer monsoon circulation. In the HI cases, the WPSH is anomalously in the north, and the southerly anomalies appear not only at 850 hPa but also at 200 hPa over East China and its adjacent sea. This southerly anomaly permeates the entire troposphere (Figure omitted). The active southerly anomalies extend into North China, bringing increasing moisture from the western Pacific and creating favorable conditions for heavy rainfall in North China. This result is consistent with that of Lu [35]. In the LI cases, the WPSH is anomalously in the south and northerly anomalies are found in the troposphere over East China; this circulation is unfavorable for the northward movement of moisture.

4 Relationship of ILSQD with summer rainfall over China 4.1 Relationship of ILSQD with rainfall in summer season over China According to the analyses above, rainfall increases over North China in the HI cases, while general circulation is favorable for more rainfall over the Yangtze valley in the LI cases. The following is a further investigation of this result. Figure 7(a) shows the correlations between the summer

ILSQD and summer rainfall from 160 stations over China. Significant positive correlations are located over most of North China, Liaoning Province, and the southwestern part of Jilin Province, all satisfying significance at the 90% confidence level, and the positive maxima value of 0.5 surpasses significance at the 99% confidence level. At the same time, negative correlations prevail over the south of the mid-lower valleys of the Yangtze River, with most of them surpassing significance at the 90% confidence level. The correlations exhibit a “+ −” pattern from north to south over East China; this distribution is nearly identical to the E1 rain pattern over the eastern monsoon region as provided by Zhao et al. [36]. This means that when the summer monsoon is intense, rainfall increases over most of North China, Liaoning province, and the southwestern part of Jilin province, while decreasing over the south of the mid-lower reaches of the Yangtze River. When the summer monsoon is especially active in the troposphere over East China, the strong southerly winds bring increased moisture into North China and the WPSH is anomalous in the north. Under these conditions, the summer rainfall belt stays over North China and rainfall decreases over the south of the mid-lower valleys of the Yangtze River due to the influence of the summer monsoon; the opposite occurs in the LI cases. Rainfall anomalies also exist in the composite distributions of summer rainfall anomalies of both the HI and LI cases (Figure omitted). In the HI cases and in the strong summer monsoon years, rainfall increases over North, Northeast, and Southwest China, while decreasing over the valley of the Yangtze River, the southern Yangtze River, and South China; thus the summer rainfall belt is anomalous in the north. In the LI cases as well as the weak summer monsoon years, the location of the summer rainfall belt is south of place where it is in the HI cases, with rainfall increasing over the valley of Huaihe River, south of Yangtze River,

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

1741

Figure 7 The correlations between summer rainfall and ILSQD in summer (a), in March (b). Unit: ×0.01, in the shadow and heavy shaded areas correlation coefficients are statistically significant at 90% and 95% confidence level respectively.

and South China, and decreasing over North and Northeast China. Tang et al. [17] compared the correlations between summer rainfall over China and summer monsoon indices defined by Huang and Yan [18], Zhang et al. [13], Sun et al. [16], and Guo [6]. Significant correlations are found over the valley of the Yangtze River, but there is no real correlation over North China. However, with regard to correlations between ILSQD defined in this paper and summer rainfall, there are both significant negative correlations over the south of the mid-lower reaches of the Yangtze River as well as significant positive correlations over North China. Based on the analyses described above, ILSQD can be said to reflect variability in general circulation and rainfall over East China, and thus used as an EA subtropical summer monsoon index. 4.2 Relationship of ILSQD in March with summer rainfall over China The studies above indicate significant correlations between summer ILSQD and summer rainfall over China. Next, we will discuss whether ILSQD has potential for application to forecasting summer rainfall over China. In this study, the relationship between ILSQD in March and summer rainfall is investigated. In March, the EA continent goes from being an atmospheric heat sink to a heat source, while atmospheric heat sources are also located over the western Pacific. Because the value of ILSQD in March is negative, a small (big) value of ILSQD in March (called the high-index (lowindex) cases) corresponds to a strong (weak) land-sea AHS contrast. Figure 7(b) shows the correlation distribution between ILSQD in March and the following summer rainfall, and indicates that there are significant negative correlations that surpass the 90% confidence level over the south of the valley of the Yangtze River. High-index (low-index) in March corresponds to more (less) summer rainfall over the southern Yangtze River valley. The relationships of ILSQD in March to rainfall in June, July, and August were investigated, beginning with June. Significant negative correlations appear east of 115°E and south of the Yangtze River, meaning that when ILSQD in March is high (low) rainfall increases (decreases) over these regions. There were no significant correlations between ILSQD in March and rainfall

in July, indicating that there is no potential for forecasting in July using ILSQD. The relationship of ILSQD in March with rainfall in August shows significant negative correlations just to locate south of the middle Yangtze River valley, so that a high (low) ILSQD corresponds to more (less) rainfall over this region in August. In summary, by using the AHS contrast between East Asia and the western Pacific, this new monsoon index can reflect not only the variability of summer monsoon in East Asia but also summer rainfall anomalies over eastern China. Furthermore, there is a significant correlation between ILSQD in March and summer rainfall over the southern Yangtze River valley.

5 Numerical simulation Diagnostic analysis has been performed to study the relationship of the land-sea AHS contrast index (ILSQD) in EA with general circulation and rainfall over East China, revealing remarkable correlations between ILSQD in March and summer rainfall over the southern Yangtze River valley. The following studies will verify the results of our diagnostic analysis. As the fifth-edition general circulation model for studying weather and climate, CAM3.1/NCAR is utilized in this investigation. This approach modifies the physical representation of key climate processes, including clouds, rainfall, aerosols, and radiation, and improves simulation of temperature over the tropopause in the tropics, winter surface temperature over the northern region, surface sunlight, and clear surface radiation over the polar region. A comprehensive description of the major components of CAM3.1/ NCAR was given by Collins et al. [37]. CAM3.1/NCAR is a global spectral atmospheric model using Sigma-P hybrid coordinates with the model top at 2.914 hPa, a horizontal T42 spectral resolution, and time step of 1200 s. To further verify the diagnostic analyses, we changed the last term on the right side of eq. (6) over EA (25°–45°N, 90°–120°E) and the western Pacific (27.5°–40°N, 122.5°–130°E; 27.5°–45°N, 130°–160°E) in the sensitivity experiments, such that the thermodynamics equation in CAM3.1/NCAR would be

1742

ZHANG Bo, et al.

Sci China Earth Sci

⎤ ∂T −1 ⎡ ∂ ∂ = (UT ) + cos φ (VT )⎥ ⎢ 2 ∂t a cos φ ⎣ ∂λ ∂φ ⎦ ∂T R ω + T δ − η + Tv + Q, ∂η C *p p

can be seen over China, with positive anomalies over the western Pacific in the LI cases (Figure 8(b)). Three experiments are listed in Table 1. (6)

where a is the radius of the Earth, λ is the longitude, Φ is the latitude, u and v are the meridional and zonal wind velocities respectively, p is pressure, δ is the horizontal divergence of wind, T is air temperature, Tv is virtual temperature, and Q is the atmospheric heat sources (sinks). 5.1

November (2010) Vol.53 No.11

Impacts of AHS on rainfall in summer season

There are three sensitivity experiments (i.e., control experiment IA, high-index experiment IB, and low-index experiment IC). The three experiments integrate data from May 1 to August 31, and the atmospheric heating rate over EA and the western Pacific in experiments IB and IC are added to the anomalous heating forcing during the period from June 1 to August 31. The anomalous atmospheric heating forcing fields are determined according to the heating rate anomalies of diagnostic analyses in the HI and LI cases (Figure 8). Figure 8(a) shows that in the HI cases the anomalous positive heating rate is over China mainland around 25°–45°N, and the anomalous negative heating rate is over the western Pacific. Negative heating rate anomalies

5.2

Impacts of AHS in March on summer rainfall

In these numerical simulations, there are still three experiments (i.e., control experiment IIA, low March index experiment IIB, and high March index experiment IIC). Figure 2(a) shows that heat sinks are located over land and heat sources are over sea in March, that when heat sources intensity on land are increased (decreased) while at sea they are decreased (increased), and that the land-sea AHS contrast between EA and the western Pacific decreased (increased). The three experiments integrate data from February 1 to August 31, and the atmospheric heating rate over EA and the western Pacific in experiments IIB and IIC incorporate the anomalous atmospheric heating rate seen during the period from March 1 to March 31. Based on our diagnostic analyses, Figure 9(a) shows the anomalous heating rate along 25°–45°N when ILSQD in March is lower, positive anomalies are found over China, and negative anomalies are found over the western Pacific, meaning that the land-sea AHS contrast is diminished. The opposite occurs when the ILSQD value in March is higher (Figure 9(b)). Three experiments are listed in Table 2.

Figure 8 Longitude-height cross section of anomalous atmospheric heating rate in summer along the latitudes 25°–45°N in the HI cases (a) and in the LI cases (b) according to the diagnostic analyses (unit: 10−6 K/s).

Table 1 Three sensitivity experiments simulated by the CAM3.1/NCAR model Experiment appellation IA

Experiment design None

IB

Add the abnormal heating rate shown in Figure 8(a) in order to increase the AHS contrast

IC

Add the abnormal heating rate shown in Figure 8(b) in order to decrease the AHS contrast

Forceable scope None (25°–45°N, 90°–120°E) (27.5°–40°N, 122.5°–130°E) (27.5°–45°N, 130°–160°E) Same as IB

Forceable time None From 1 June to 31 August Same as IB

ZHANG Bo, et al.

Sci China Earth Sci

1743

November (2010) Vol.53 No.11

Figure 9 Longitude-height cross section of anomalous atmospheric heating rate in March along the latitudes 25°–45°N in the HI cases (a) and in the LI cases (b) according to the diagnostic analyses (unit: 10−6 K/s).

Table 2

Three sensitivity experiments simulated by the CAM3.1/NCAR model

Experiment appellation IIA

Experiment design None

IIB

Add the abnormal heating rate shown in Figure 9(a) in order to decrease the AHS contrast

IIC

Add the abnormal heating rate shown in Figure 9(b) in order to increase the AHS contrast

5.3

Results of numerical simulation

The differences in summer rainfall (IB minus IA, IC minus IA) are shown in Figure 10(a) and (b). When the land-sea AHS contrast between EA and the western Pacific is increased, a positive difference is seen over North China centered over 116°E, 38°N, while a negative difference is seen over most of South China, and over the region from the Yangtze-Huaihe River basin valley to the southern Yangtze River valley through the mid-lower Yangtze River valley. Conversely, when the land-sea AHS contrast is decreased, a negative difference is found over part of North China centered in the northern Hebei Province, while a positive difference is found over south of the Yellow River and in most of the region to east of 110°E. There are two positive dif-

Figure 10 The differences in summer rainfall l between IB and IA (a), IC and IA (b). Unit: mm/d.

Forceable scope None (25°–45°N, 90°–120°E) (27.5°–40°N, 122.5°–130°E) (27.5°–45°N, 130°–160°E) Same as IIB

Forceable time None From 1 March to 31 March Same as IIB

ference centers over the mid-lower Yangtze River valley and the Pearl River delta. In summary, an enhanced (weakened) summer land-sea AHS contrast between East Asia and the western Pacific corresponding to high (low) summer ILSQD induces summer rainfall increases (decreases) over North China and decreases (increases) over mid-lower Yangtze River valley and south of the Yangtze River. The differential wind vector field at 850 hPa of IB minus IC is shown in Figure 11(a). The northeasterly difference flows into inland China through the southern Yangtze River, then turns to the southerly direction, and arrives in North China and the eastern part of Northwest China. An anticyclone difference is located over China-Mongolia, to the south of which northerly winds combine with the southerly winds from the western Pacific over North China and the east of Northwest China. This indicates that an enhanced land-sea AHS contrast strengthens the southerly winds over central China, which combines with the northerly winds from middle and high latitudes over North China, creating a circulation favoring heavy rainfall over North China. A weak land-sea AHS contrast causes the opposite results. The difference in wind vector fields at 200 hPa (IB minus IC) is shown in Figure 11(b). A strong anticyclone difference exists over the western TP and a cyclone difference over southern Lake Baikal. The southwesterly difference over southeast of Lake Baikal combines with the southwesterly difference from the western Pacific, which then

1744

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

Figure 11 The difference of wind fields between IB and IC in summer at 850 hPa (a) and at 200 hPa (b) (unit: m/s), the difference of 500-hPa geopotential height. Shaded areas: the topographic contour of 3000 m; unit: geopotential meter.

flows together into North China. Hence, the southerly difference covers North China from the lower level to higher level of the troposphere and brings increasing moisture into North China, inducing more rainfall. Figure 11(c) shows the differential 500-hPa geopotential height field of IB minus IC. The notable features of this graph are the presence of a negative difference over the Olhotsk Sea and a positive one over the area south of the Yangtze River and South China. The weak high-pressure system over the Olhotsk Sea and weak meridional circulation in the mid-high latitudes are not favorable for southward movement of cold air into North China. Under these conditions, the rainfall belt is anomalous in the north, and heavy precipitation occurs in North China while rainfall decreases over the mid-lower Yangtze River valley due to the influence of the summer monsoon. As a result, when the land-sea AHS contrast is enhanced, the high-pressure intensity over the Olhotsk Sea weakens and the southerlies from the lower to higher levels of the troposphere over eastern China are strengthened and extended northward. They in turn bring more moisture and thus more rainfall into North China; when the land-sea AHS contrast is diminished, the converse occurs. Figure 12(a) and (b) show the differential precipitation of IIB minus IIA and IIC minus IIA in the summer season. When the land-sea AHS contrast in March is diminished, negative rainfall differences occur over the area between the lower Yellow River and lower Yangtze River, the YangtzeHuaihe River basin, the mid-lower Yangtze River valley, and the northern part of South China. Simultaneously, positive rainfall differences are found over most of South China, Northwest and North China, and most of Northeast China. When the land-sea AHS contrast is enhanced in March, negative rainfall differences are found over the east of

Figure 12 The differential precipitation of IIB minus IIA (a), IIC minus IIA (b). Unit: mm/d.

Northwest China, North China, and most of Northeast China, while positive rainfall differences are seen over the region from the mid-lower Yangtze River valley to the eastern fringe of the TP. The above analyses are consistent with the results of diagnostic analyses. Figure 13(a) shows the differential atmospheric diabatic heating rate averaged over 25°–45°N of IIB minus IIC in the summer. Positive anomalies exist over the continent while negative anomalies center over the regions 120–130°E and 145–155°E, indicating that the weak land-sea AHS contrast in March enhances the summer land-sea AHS contrast, strengthens the EASM intensity (Figure 13(b)) and decreases rainfall over the mid-lower Yangtze River valley due to the influence of the summer monsoon over East China.

6

Conclusions and discussion

Using atmospheric heat source datasets, we define a new land-sea AHS difference index and investigate the relationship of this index with general circulation and summer

ZHANG Bo, et al.

Sci China Earth Sci

November (2010) Vol.53 No.11

1745

Figure 13 The differential atmospheric diabatic heating rate averaged over 25°–45°N of IIB minus IIC in the summer (a), the differential 850 hPa wind fields of IIB minus IIC in summer (b). Unit: m/s.

rainfall over China. We then design a series of numerical experiments to test this relationship using CAM3.1/NCAR. The results show that the land-sea AHS difference index ILSQD not only has a clear physical meaning, but also reflects the abnormal variability of summer monsoons. In the HI cases, the active high-level and low-level southwesterly winds over the EASM region bring more moisture into North China; in the LI cases, the converse occurs. There are significant correlations between summer ILSQD and summer rainfall over East China; in the HI (LI) cases rainfall increases (decreases) over North China and decreases (increases) over the mid-lower Yangtze River valley. The global atmospheric model can simulate the influence of summer land-sea AHS contrast on summer China rainfall. When summer land-sea AHS contrast is enhanced (diminished), the southwesterly winds strengthen (weaken) at higher and lower tropospheric levels over East China, rainfall increases (decreases) over North China and decreases (increases) over south of the Yellow River; these features are consistent with the results of the diagnostic analyses. The results of diagnostic analyses and numerical experiments indicate that ILSQD can be used to make meaningful forecasts; the negative correlations of ILSQD in March with summer rainfall over the mid-lower Yangtze River valley are remarkable. This all leads to the question: how does variability of AHS over East China and the western Pacific in March affect the following summer AHS variability over these regions? This issue is of great importance to further understand these interrelated phenomena, and it will be the focus of future research. We thank all anonymous reviewers and Miss Deng Xueyuan for their valuable comments in revising the paper. This work was supported by National Natural Science Foundation of China (Grant Nos. 90711003, 40505014, 40805035 and 40633018).

1

2 3

4

5 6 7

8

9 10 11

12 13

14

15

16

Chen L X, Jin Z H. On the interaction of circulations between the two hemispheres in the East Asia Monsoon circulation system during summer. In: Proceedings of the Symposium on the Summer Monsoon in Southeast Asia, 1982. 218–231 Chen L X, Luo S H, Shen R G. The Asian summer monsoon and its relations to the rainfall in China. Adv Atmos Sci, 1984, 1: 263–276 Zhu Q G, He J H, Wang P X. A study of circulation differences between East-Asian and Indian summer monsoons with their interaction. Adv Atmos Sci, 1986, 3: 466–477 Tao S Y, Chen L X. A review of recent research on the East Asian summer monsoon in China. In: Chang C P, Krishnamurti T N, eds. Monsoon Meteorology. Oxford: Oxford University Press, 1987. 6–92 Wang B, Wu Z W, Li J P, et al. How to measure the strength of the East Asian summer monsoon. J Clim, 2008, 21: 4449–4463 Guo Q Y. The summer monsoon intensity index in East Asia and its variation (in Chinese). Acta Geograph Sin, 1983, 38: 207–217 Shi N, Zhu Q G, Wu B G. The East Asian summer monsoon in relation to summer large scale weather-climate anomaly in China for last 40 years (in Chinese). Sci Atmosph Sin, 1996, 20: 575–583 Zhao P, Zhou Z J. East Asian subtropical summer monsoon index and its relationships to rainfall (in Chinese). Acta Meteorol Sin, 2005, 63: 933–941 Li J P, Zeng Q C. A unified monsoon index. Geophys Res Lett, 2002, 29: 1151–1154 Li J P, Zeng Q C. A New monsoon index and the geographical distribution of the global monsoons. Adv Atmos Sci, 2003, 20: 299–302 Li J P, Zeng Q C. A new monsoon index, its interannual variability and relationship with monsoon precipitation (in Chinese). Clim Environ Res, 2005, 10: 351–365 Wang H J. Instability of the East Asian summer monsoon-ENSO relations. Adv Atmos Sci, 2002, 19: 1–11 Zhang Q Y, Tao S Y, Chen L T. The inter-annual variability of East Asian summer monsoon indices and its association with the pattern of general circulation over East Asia (in Chinese). Acta Meteorol Sin, 2003, 61: 558–568 Zhu C W, He J G, Wu G X. East Asian monsoon index and its inter-annual relationship with large scale thermal dynamic circulation (in Chinese). Acta Meteorol Sin, 2000, 58: 391–402 Qiao Y T, Chen L T, Zhang Q Y. The definition of East Asian monsoon indices and their relationship to climate in China (in Chinese). Chin J Atmosph Sci, 2002, 26: 69–82 Sun X R, Chen L X, He J H. Index of land-sea thermal differences and its relation to interannual variation of summer circulation and rainfall over East Asia (in Chinese). Acta Meteorol Sin, 2002, 60: 164–172

1746 17 18

19

20

21

22

23

24

25

26

27

ZHANG Bo, et al.

Sci China Earth Sci

Tang X, Sun G W, Qian W H. Research of the Northern Fringe of the Asian Summer Monsoon. Beijing: China Meteorological Press, 2007. 41 Huang G, Yan Z W. The East Asian summer monsoon circulation anomaly index and its interannual variations. Chin Sci Bull, 1999, 44: 1325–1329 Duan A M, Liu Y M, Wu G X. Heating status of the Tibetan Plateau from April to June and rainfall and atmospheric circulation anomaly over East Asia in midsummer. Sci China Ser D-Earth Sci, 2005, 48: 250–257 Luo H B, Chen R. The impact on the anomalous heat sources over the eastern Tibetan Plateau on the circulation over East Asia in summer half year (in Chinese). Sci Meteorol Sin, 1995, 15: 94–102 Zhao P, Chen L X. Climatic features of atmospheric heat source/sink over the Qinghai-Xizang Plateau in 35 years and its relation to rainfall in China. Sci China Ser D-Earth Sci, 2001, 44: 858–864 Jian M Q, Luo H B, Qiao Y T. On the relationships between the summer rainfall in China and the atmospheric heat sources over the eastern Tibetan Plateau and the western Pacific warm pool (in Chinese). J Trop Meteorol, 2004, 20: 355–364 Jian M Q, Qiao Y T, Luo H B, et al. The relationship of the atmospheric heat sources in early summer over East Asia-Pacific to the abnormality of precipitation in the Yangtze River basin and its vicinity in July and August (in Chinese). Chin J Atm Sci, 2007, 31: 735–746 Hua W, Fan G Z, Zhou D W, et al. Preliminary analysis on the relationships between Tibetan Plateau NDVI change and its surface heat source and precipitation of China. Sci China Ser D-Earth Sci, 2008, 51: 677–685 Huang R H, Yan B L. The effects of topographical and thermal forcing on formation and maintenance of the summer monsoon over Asia (in Chinese). Acta Meteorol Sin, 1987, 45: 394–407 Wu G X, Zhang Y S. Tibetan Plateau forcing and the timing of the monsoon onset over south Asia and the South China Sea. Mon Weather Rev, 1998, 126: 913–927 Gong Y F, Duan T Y, Zhang H. Characteristics of the atmospheric

November (2010) Vol.53 No.11

28 29

30

31

32

33

34

35

36

37

heating source/sink over Asia and its relationship with drought/flood in the Yangtze River-Huaihe River valley (in Chinese). Chin J Atmos Sci, 2007, 31: 89–93 He J H, Zhao P, Zhu C W, et al. Discussions on the East Asian subtropical monsoon (in Chinese). Acta Meteorol Sin, 2008, 66: 683–696 Ma S J, Sun S Q. Characteristics of persistent anomaly of the South China Sea summer monsoon and its influence on the global atmospheric circulation, part II. Numerical experiment (in Chinese). Clim Environ Res, 2003, 8: 356–369 Ma S J, Sun S Q. Characteristics of persistent anomalies of the South China Sea summer monsoon and its influences on the global atmospheric circulation (in Chinese). Clim Environ Res, 2001, 6: 435–447 Yanai M, Esbensen S, Chu J H. Determination of bulk properties of tropical cloud clusters from large-scale heat and moisture budgets. J Atmos Sci, 1973, 30: 611–627 Zhao S R, Song Z S. Floods and droughts in northern China and general circulation anomalies over middle and high latitudes (in Chinese). Plateau Meteorol, 1999, 18: 535–540 Zhao S R, Song Z S, Ji L R. Study on the relationship between the anomalies of rainfall over North China and the Asian monsoon anomalies (in Chinese). Acta Meteorol Sin, 2002, 60: 68–75 Tao S Y, Zhu F K. The 100-mb flow patterns in southern Asia in summer and its relation to the advance and retreat of the west-Pacific subtropical anticyclone over the far East (in Chinese). Acta Meteorol Sin, 1964, 34: 358–396 Lu R Y. Atmospheric circulation anomaly associated with the spring rainfall anomaly in North China (in Chinese). Clim Environ Res, 2001, 6: 400–408 Zhao Z G, Zhu Y F, Liu Y X, et al. Decadal variation of summer rain-pattern in China during 1880–2006 (in Chinese). Adv Clim Change Res, 2008, 4: 95–100 Collins W D, Rasch P J, Boville B A, et al. The formulation and atmospheric simulation of the Community Atmosphere Model Version 3 (CAM3). J Clim, 2006, 19: 2144–2161