persistent spring shortwave cloud radiative effect and the ... · persistent spring shortwave cloud...

19
Persistent Spring Shortwave Cloud Radiative Effect and the Associated Circulations over Southeastern China JIANDONG LI, a,b WEI-CHYUNG WANG, c JIANGYU MAO, a ZIQIAN WANG, d GANG ZENG, b AND GUOXING CHEN c a State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, China b Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science and Technology, Nanjing, China c Atmospheric Sciences Research Center, University at Albany, State University of New York, Albany, New York d School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou, China (Manuscript received 19 June 2018, in final form 26 February 2019) ABSTRACT Clouds strongly modulate regional radiation balance and their evolution is profoundly influenced by cir- culations. This study uses 2001–16 satellite and reanalysis data together with regional model simulations to investigate the spring shortwave cloud radiative effect (SWCRE) and the associated circulations over southeastern China (SEC). Strong SWCRE, up to 2110 W m 22 , persists throughout springtime in this region and its spring mean is the largest among the same latitudes of the Northern Hemisphere. SWCRE exhibits pronounced subseasonal variation and is closely associated with persistent regional ascending motion and moisture convergence, which favor large amounts of cloud liquid water and resultant strong SWCRE. Around pentad 12 (late February), SWCRE abruptly increases and afterward remains stable between 228 and 328N. The thermal and dynamic effects of Tibetan Plateau and westerly jet provide appropriate settings for the maintenance of ascending motion, while water vapor, as cloud water supply, stably comes from the southern flank of the Tibetan Plateau and South China Sea. During pentads 25–36 (early May to late June), SWCRE is further enhanced by the increased water vapor transport caused by the march of East Asian monsoon systems, particularly after the onset of the South China Sea monsoon. After pentad 36, these circulations quickly weaken and the SWCRE decreases accordingly. Individual years with spring strong and weak rainfall are chosen to highlight the importance of the strength of the ascending motion. The simulation broadly repro- duced the observed results, although biases exist. Finally, the model biases in SWCRE–circulation associa- tions are discussed. 1. Introduction Clouds play significant roles in climate systems by modulating the radiative energy budget and hydrologi- cal cycle. The importance and complexity of cloud physics processes cause them to contribute the biggest uncertainty in current climate simulations and predictions (Stephens 2005; Boucher et al. 2013). Cloud formation, type, and spatial distribution are deeply affected by general circulations and then exhibit distinctive regional features (Rogers and Yau 1989; Bony et al. 2015). Southeastern China (SEC) is located in the East Asian monsoon re- gion, which is characterized by prevailing monsoon cir- culations and complicated topography. Meanwhile, large amounts of continental stratus clouds and strong cloud albedo effect are distributed in SEC (Klein and Hartmann 1993; Yu et al. 2001, 2004). Due to monsoon circulations, cloud radiative characteristics over SEC also have remarkable seasonal variation, with strong winter–summer contrast (Luo et al. 2009; Li et al. 2017a). At present, climate models show evident biases of shortwave cloud radiative effect (CRE) at the top of the atmosphere (TOA) and cloud cover over SEC, Denotes content that is immediately available upon publica- tion as open access. Corresponding author: Dr. Jiandong Li, [email protected] This article is licensed under a Creative Commons Attribution 4.0 license (http://creativecommons.org/ licenses/by/4.0/). VOLUME 32 JOURNAL OF CLIMATE 1JUNE 2019 DOI: 10.1175/JCLI-D-18-0385.1 Ó 2019 American Meteorological Society 3069

Upload: others

Post on 03-Apr-2020

38 views

Category:

Documents


0 download

TRANSCRIPT

Persistent Spring Shortwave Cloud Radiative Effect and the AssociatedCirculations over Southeastern China

JIANDONG LI,a,b WEI-CHYUNG WANG,c JIANGYU MAO,a ZIQIAN WANG,d GANG ZENG,b

AND GUOXING CHENc

a State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics,

Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing, ChinabCollaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of

Information Science and Technology, Nanjing, ChinacAtmospheric Sciences Research Center, University at Albany, State University of New York, Albany, New York

d School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou, China

(Manuscript received 19 June 2018, in final form 26 February 2019)

ABSTRACT

Clouds strongly modulate regional radiation balance and their evolution is profoundly influenced by cir-

culations. This study uses 2001–16 satellite and reanalysis data together with regional model simulations to

investigate the spring shortwave cloud radiative effect (SWCRE) and the associated circulations over

southeastern China (SEC). Strong SWCRE, up to2110Wm22, persists throughout springtime in this region

and its spring mean is the largest among the same latitudes of the Northern Hemisphere. SWCRE exhibits

pronounced subseasonal variation and is closely associated with persistent regional ascending motion and

moisture convergence, which favor large amounts of cloud liquidwater and resultant strong SWCRE.Around

pentad 12 (late February), SWCRE abruptly increases and afterward remains stable between 228 and 328N.

The thermal and dynamic effects of Tibetan Plateau and westerly jet provide appropriate settings for the

maintenance of ascending motion, while water vapor, as cloud water supply, stably comes from the southern

flank of the Tibetan Plateau and South China Sea. During pentads 25–36 (early May to late June), SWCRE is

further enhanced by the increasedwater vapor transport caused by themarch of EastAsianmonsoon systems,

particularly after the onset of the South China Sea monsoon. After pentad 36, these circulations quickly

weaken and the SWCRE decreases accordingly. Individual years with spring strong and weak rainfall are

chosen to highlight the importance of the strength of the ascending motion. The simulation broadly repro-

duced the observed results, although biases exist. Finally, the model biases in SWCRE–circulation associa-

tions are discussed.

1. Introduction

Clouds play significant roles in climate systems by

modulating the radiative energy budget and hydrologi-

cal cycle. The importance and complexity of cloud

physics processes cause them to contribute the biggest

uncertainty in current climate simulations and predictions

(Stephens 2005; Boucher et al. 2013). Cloud formation,

type, and spatial distribution are deeply affected by general

circulations and then exhibit distinctive regional features

(Rogers and Yau 1989; Bony et al. 2015). Southeastern

China (SEC) is located in the East Asian monsoon re-

gion, which is characterized by prevailing monsoon cir-

culations and complicated topography. Meanwhile,

large amounts of continental stratus clouds and strong

cloud albedo effect are distributed in SEC (Klein and

Hartmann 1993; Yu et al. 2001, 2004). Due to monsoon

circulations, cloud radiative characteristics over SEC

also have remarkable seasonal variation, with strong

winter–summer contrast (Luo et al. 2009; Li et al.

2017a). At present, climate models show evident biases

of shortwave cloud radiative effect (CRE) at the top of

the atmosphere (TOA) and cloud cover over SEC,

Denotes content that is immediately available upon publica-

tion as open access.

Corresponding author: Dr. Jiandong Li, [email protected]

This article is licensed under a Creative Commons

Attribution 4.0 license (http://creativecommons.org/

licenses/by/4.0/).

VOLUME 32 J OURNAL OF CL IMATE 1 JUNE 2019

DOI: 10.1175/JCLI-D-18-0385.1

� 2019 American Meteorological Society 3069

where a large regional cloud albedo effect is particularly

underestimated (Flato et al. 2013; F. Wang et al. 2014).

These model biases significantly contribute to the un-

certainties of regional cloud feedbacks and climate

projections. Thus, accurate simulation of climate pro-

cesses over East Asia requires a more in-depth un-

derstanding of cloud radiation characteristics over SEC.

In addition to the winter–summer monsoon period,

spring is another distinctive period over SEC, where

persistent spring rainfall is a unique regional climatic

phenomenon (Tian and Yasunari 1998; Wan and Wu

2007). In this region, the spring rainfall contributes more

than 30% of the total annual precipitation (LinHo 2008;

Wu and Mao 2016) and also has obvious subseasonal

and interannual variation (Pan et al. 2013; Huang et al.

2015). The anomalies of spring rainfall and relevant

circulation exert great impacts on the weather, climate

processes, agriculture, and socioeconomics of this region.

It is noted that distinct spring shortwave CRE (SWCRE)

also appears over SEC. Work by Wang et al. (2004) and

Li et al. (2017a) withmonthly datasets showed that strong

SWCREover SEC occurs in winter and persists until full

spring. As shown in Fig. 1, spring mean SWCRE over

SEC (around 1048–1228E and 228–328N) is the strongest

at the same latitudes of the Northern Hemisphere with a

maximum value of up to 2120Wm22, the intensity of

which is much larger than over the eastern coasts of the

Pacific (Fig. 1a). Unlike the usual amount of spring

precipitation in SEC, sparse precipitation occurs over

the eastern coastal regions of the Pacific where de-

scending motion prevails (Fig. 1b). Meanwhile, the

spring SWCRE averaged over SEC is the largest of the

four seasons (Fig. 2). So robust spring SWCRE certainly

exerts a nonnegligible force on the regional climate

system. Moreover, spring is the winter–summer transi-

tion season, and cloud roles in radiation budget and at-

mospheric heating potentially affect the following

migration of East Asian summer monsoon circulation

(Guo et al. 2015; Li et al. 2015). Figure 1 also shows that

the spatial pattern of SWCRE is somewhat similar to

that of rainfall over the midlatitudes of the Northern

Hemisphere. Hence, it is very important to investigate

spring CRE over SEC for understanding the radiation

budget and hydrological cycle, and their seasonal tran-

sition in East Asia.

Some previous work noted that spring cloud radiative

characteristics over SEC are connected with regional

circulations to some extent. Yu et al. (2001) suggested

that the convergence below 700hPa in winter and spring

helps to the middle to low cloud formation over eastern

FIG. 1. Geographical distributions of spring mean [March–May (MAM)] (a) top-of-the-at-

mosphere shortwave cloud radiative effect (SWCRE; Wm22) derived from CERES-MODIS

(Clouds and the Earth’s Radiant Energy System–Moderate Resolution Imaging Spectroradi-

ometer) and (b) precipitation (mmday21) from the Global Precipitation Climatology Project

during the period of 2001–16. The black solid line is the Tibetan Plateau (over 3000m).

3070 JOURNAL OF CL IMATE VOLUME 32

China, where total cloud fraction peaks in spring. Work

by Yu et al. (2004), Li and Gu (2006), and Zhang et al.

(2013) showed that Tibetan Plateau (TP) dynamic ef-

fects contribute to the low-level convergence, ascending

motion, and midlevel divergence over eastern China,

favoring regional stratus clouds and high SWCRE in the

cold season. Besides the TP, other regional circulations,

such as the low-level southerly wind from the South

China Sea, the northwest Pacific anticyclone, and the

East Asian subtropical jet, also have pronounced roles

in spring vertical motion and moisture supply over SEC

(Liang and Wang 1998; Zhao et al. 2007; Zhu et al.

2007), which are essential for regional clouds and rain-

fall. Moreover, distinct subseasonal features exist in the

East Asian monsoon region including SEC, where the

intensity, spatial distribution, and movement timing of

convection, moisture, and prevailing low-level air currents

experience remarkable subseasonal variations especially in

the winter–spring and spring–summer transition periods

(Lau and Yang 1997; Tian and Yasunari 1998; He et al.

2008). Consequently, spring cloud radiative characteristics

potentially show corresponding subseasonal variation and

interannual differences as influencing circulations. How-

ever, previous studies focused on seasonal or monthly

mean states of cloud radiative characteristics and

mostly did not reveal subseasonal features and associa-

tions with regional circulations. Particularly for spring

(March–May), the following issues remain unclear over

SEC: 1) How does subseasonal SWCRE vary? 2) What

types of relationships exist between SWCRE and re-

gional circulations in spring mean and subseasonal

scales? The purpose of this study is therefore to address

these questions. This study investigates spring SWCRE,

including the seasonal mean, subseasonal features and

some interannual differences, and identifiesmajor regional

circulation conditions contributing to the occurrence and

maintenance of spring SWCRE over SEC. Our analyses

are conducted using satellite and reanalysis data, as well

as a regional model. The regional simulation is used to

verify the climatological and case-year results and to

further enhance our confidence in the robustness of as-

sociated climatic features.

The remainder of this paper is organized as follows. In

section 2, the data and method used are introduced.

Section 3 presents the multiyear spring mean features of

SWCRE and circulations over SEC, and analyzes their

possible associations. Section 4 shows corresponding

subseasonal variations during late winter to early sum-

mer and investigates their connections in temporal

evolutions. Section 5 investigates the interannual dif-

ferences of SWCRE and regional circulations with the

analysis and simulation in two case years. Finally, the

conclusions and discussion are presented in section 6.

2. Data and method

a. Observational and reanalysis datasets

The cloud-radiation datasets used in this study consist

of two parts. The first part of the data is from satellite

retrievals. Daily and monthly mean TOA radiative

fluxes are fromNASA’s Clouds and the Earth’s Radiant

Energy System (CERES), which utilizes the most recent

CERES Energy Balanced and Filled at the TOA

(EBAF-TOA) Ed2.8 dataset (Doelling et al. 2013). The

CERES-EBAF includes incident shortwave (SW) flux

and outgoing SW and longwave (LW) radiative fluxes at

the TOA under clear-sky and all-sky conditions. This

dataset has been widely used to study the role of clouds

and the energy cycle in the Earth climate system (Loeb

et al. 2009; Wild et al. 2013). Thus, CERES-EBAF is the

observational radiation dataset used herein. The afore-

mentioned datasets have a spatial resolution of 1.08latitude by 1.08 longitude and are available from March

2000 to February 2017.

The second part of the data is from ERA-Interim

(Dee et al. 2011); these data include daily mean cloud

amount, liquid, and ice water. The aforementioned

cloud-radiation variables in ERA-Interim are calculated

in the host assimilation model and are physically con-

sistent with the reanalyzed meteorological variables to a

large degree. ERA-Interim provides reasonable de-

scription of spatiotemporal features of cloud-radiation

variables (e.g., total cloud amount, cloud vertical struc-

tures, and TOA CREs) over East Asia although some

intensity biases exist (Yin et al. 2015; Li and Mao 2015;

Li et al. 2017b). Another great advantage of ERA-

Interim reanalysis data is their continuous and long-term

states. The satellite data are limited by their operational

period, instrument performance and retrieval algorithm,

FIG. 2. Seasonal mean cloud radiative effects (CREs) averaged

over southeastern China (228–328N, 1048–1228E) during the periodof 2001–16. Here LWCRE, SWCRE, and NCRE are the abbrevi-

ations of longwave, shortwave, and net CREs, respectively. ANN,

DJF, MAM, JJA, and SON indicate the annual, winter, spring,

summer, and autumn mean, respectively.

1 JUNE 2019 L I E T AL . 3071

and cannot fully cover the spatiotemporal resolution in

this study. Thus, daily cloud cover and cloud water from

ERA-Interim are used to explain relevant cloud-radiation

issues, particularly in the subseasonal variation and

vertical distribution. The meteorological variables are

also taken from ERA-Interim and show good perfor-

mance in reproducing wind fields, atmospheric mois-

ture, and precipitation (Simmons et al. 2014; Huang

et al. 2016). The spatial resolution of ERA-Interim is

1.58 latitude by 1.58 longitude, with 37 vertical pressure

layers in this study.

The daily mean precipitation data are from theGlobal

Precipitation Climatology Project (GPCP) with a spatial

resolution of 1.08 latitude by 1.08 longitude, which rangesfromOctober 1996 to February 2017 (Adler et al. 2003).

The monthly GPCP has a spatial resolution of 2.58 lati-tude by 2.58 longitude. The GPCP dataset represents the

climatological state of precipitation well (Yin et al. 2004;

Huffman et al. 2009). In this study, CERES satellite re-

trievals, ERA-Interim meteorological fields, and GPCP

precipitation are used as observational references.

b. Analysis methods

CREs are widely used variables for effectively de-

scribing the bulk cloud effects on air-surface systems.

Therefore, in this study, CREs are used as key study var-

iables to analyze seasonal features of the regional cloud-

radiation process. Following previous work (Ramanathan

et al. 1989; Boucher et al. 2013), TOA CREs are defined

as differences in TOA radiative fluxes between clear-sky

and all-sky conditions:

LWCRE5OLRCS2OLR, (1)

SWCRE5RSUTCS2RSUT, (2)

NCRE5LWCRE1 SWCRE, (3)

where OLRCS and OLR are outgoing LW radiative

fluxes at the TOA under clear-sky and all-sky condi-

tions, respectively; RSUTCS and RSUT are the corre-

sponding outgoing SW radiative fluxes; and net CRE

(NCRE) is the arithmetic sum of LWCRE and SWCRE.

The radiative fluxes and CREs in this study are for TOA

without a special description.

The best CREs data are from CERES-EBAF, which

begins in March 2000. Thus, in this study, only rean-

alyzed and simulated datasets since the beginning of the

CERES satellite era are used. The spring climatological

state and interannual variabilities are examined in this

study. The 16-yr climatological means of the meteoro-

logical fields and cloud-radiation variables are obtained

for the period of 2001–16. The spring seasonal average is

calculated using the results for March, April, and May.

The climatological pentad (5 day) average is used to

analyze the subseasonal variation to identify the general

circulation influence on regional cloud radiative effects.

This procedure of pentad mean is effective for suppress-

ing synoptic disturbances and yet retains the prominent

feature of interest (Zhao et al. 2007; LinHo et al. 2008).

The subseasonal period of interest is pentads 1–36, which

ranges from late winter to early summer. Besides cli-

matological states, two real case years (2010 and 2011)

are selected to compare the differences in cloud-

radiation and circulation over SEC. The domain of

SEC is defined as 228–328N, 1048–1228E based on pre-

vious studies (Liang andWang 1998; Yu et al. 2001;Wan

andWu 2007; Li et al. 2017a; Li et al. 2018), which covers

the continental areas of the middle and lower Yangtze

River and South China.

c. Model and experimental design

In this study, the Weather Research and Forecasting

(WRF) Model (version 3.7.1) is used for the regional

simulation. The WRF Model shows considerable re-

producibility for the East Asian climate (e.g., Kim and

Hong 2010; Z. Wang et al. 2014). In our simulation,

physical packages include theWSM6 cloudmicrophysics

scheme (Hong and Lim 2006), Kain–Fritsch convective

scheme (Kain 2004), Noah land surface scheme (Chen

and Dudhia 2001), Yonsei University (YSU) planetary

boundary layer (PBL) scheme (Hong et al. 2006), and

RapidRadiativeTransferModel forGCMs (RRTMG) for

longwave and shortwave radiation (Iacono et al. 2008).

The simulation domain covers most parts of East Asia

and adjacent oceans with 260 grid points along the east–

west direction and 170 along the north–south direction

(the buffer zone has 10 grid points). A Lambert pro-

jection is adopted, and the domain is centered over

308N, 112.58E. The model has a 30-km horizontal reso-

lution and 30 vertical layers with a terrain-following

sigma coordinate and prescribed model top at 50 hPa.

The initial state of the atmosphere and lateral boundary

conditions (updated every 6 h) are from ERA-Interim,

which have a spatial resolution of 0.758 latitude by

0.758 longitude, and the daily updated sea surface tem-

perature (SST) forcing dataset is also fromERA-Interim.

For the climatological simulation of 2001–16, one run

is conducted and its initial run time is 0000 UTC

16December. The 16-yr simulation is averaged to obtain

multiyear mean spring and subseasonal results. As for

each case year (2010 and 2011), an ensemble experiment

with four runs is performed under different initial

conditions at 0000 UTC 16 December, 0600 UTC

17 December, 1200 UTC 18 December, and 1800 UTC

19December, and each run ends at 1800UTC 31August

of the following year. The simulation period before

3072 JOURNAL OF CL IMATE VOLUME 32

0000 UTC 1 January is considered to be the spinup time.

The simulation ranging from 1 January to 31 August is

used in this study. First, an arithmetic mean of four runs

is taken, and then the pentad mean is obtained to ana-

lyze subseasonal variations.

3. Results

a. Climatological states of spring mean

Figure 3 presents geographical distributions of spring

mean circulations and cloud water path. Here, the cir-

culations include 200-hPa westerly jet, 500-hPa vertical

velocity, 850-hPa wind, column-integrated water vapor

flux, and divergence. In spring, a strong westerly jet at

200hPa lies between the East China Sea and southern

Japan, and the southern side of the entrance region of

the jet is just located over SEC (Fig. 3a), which aids in

regional updraft motion (Liang and Wang 1998; Holton

and Hakim 2012). Low-level southwesterly wind to the

southern flank of the TP and southerly wind from the

west of South China Sea bring abundant water vapor

into SEC, where is an obvious moisture sink region

(Figs. 3a,b). Because of the above circulation distribu-

tions, considerable water vapor readily condenses as

cloud water with the ascending motion. As shown in

Fig. 3c, substantial cloud liquid water exists over SEC

(Fig. 3c). Large SWCRE, ascendingmotion, water vapor

sink, and cloud water are all distributed over SEC and

their high value centers are around 1108E (Figs. 1a and

3). This indicates that spring circulations over SEC favor

not only the occurrence of rain but also the formation of

cloud water and resultant SWCRE.

Figure 4 presents simulated SWCRE, circulations,

and cloud water path in spring. The simulation can re-

produces spring ascending motion at 500 hPa over SEC,

but the ascending motion over the Indochina Peninsula

is much larger than the observation. In the simulation,

the low-level southerly wind from South China Sea is

captured but the southwesterly wind from east to the TP

is weaker (Fig. 4b), inducing a weaker transfer of water

vapor over SEC. The spatial pattern of cloud water path

is well reproduced by themodel and its high value center

is very close to the observation (Fig. 4c). Note that

the simulated cloud water path is weaker than the

FIG. 3. Spring mean (a) circulations, (b) column water vapor flux (kgm21 s21) and divergence (1024 kgm22 s21),

and (c) cloud water path (CWP; gm22) from ERA-Interim during the period of 2001–16. In (a), the green line

indicates the 200-hPawesterly jet and the vector is 850-hPawind (m s21). In (c), the shading is the total CWP (liquid

and ice) and the blue contour line is liquid CWP with an interval of 20 (gm22). The black solid line is the TP (over

3000m). Here, column water vapor divergence is integrated from surface to 100 hPa. The vector data are masked

below 850 hPa.

1 JUNE 2019 L I E T AL . 3073

observational reference, but the simulated cloud liquid

radius is also small (not shown) and helps to intensify

cloud optical depth and SWCRE over SEC and its east-

ern sea. The weak CWP and small cloud radius in WRF

simulation show compensatory roles in determining

SWCRE, resulting in a comparable intensity of simulated

SWCRE relative to the observation (Figs. 1a and 4a). The

simulation basically reproduces strong spring SWCRE

and corresponding circulations, although biases exist.

Note that the high-value center of spring SWCRE is

mainly distributed inwestern and central SEC, but stronger

rainfall appears eastern SEC. This high-value distribution

difference between spring SWCREand rainfall is related to

orography and land–sea effects. Li and Yu (2014) showed

that compared with eastern SEC, weaker rainfall with

higher frequency is dominant over western SEC because of

terrain effects. Higher rainfall frequency means longer

cloud duration time and larger SWCRE. Relatively, east-

ern SEC is more adjacent to oceans, and the abundant

moisture and land–sea effects enable larger amounts of

rainfall (Yuan et al. 2012). Additionally, heavy aerosol

loading over western and central SEC also contributes to

small cloud radius and strong SWCRE (Li et al. 2017a).

SWCRE at subtropical andmidlatitudes highly depends

on low- to midlevel cloud water content (Cess et al. 1990;

Zhang et al. 2005). We further analyzed vertical distri-

butions of spring meridional mean cloud water and

winds to examine the relationship among vertical cir-

culations, cloudwater, and SWCRE.As clearly shown in

Fig. 5a, the westerly jet core is around 308N, and de-

scending motion and northerly wind occur north to

308N. The obvious ascending motion and southerly wind

appear over 208–308N (the SEC latitudinal zone) and

just correspond to large cloud water content between

850 and 500 hPa, the magnitude of which is much

larger than at same tropical longitudes (Fig. 5a). From

the pressure–longitude cross section averaged over

228–328N (Fig. 5b), the strongest ascending motion oc-

curs at low to midlevels immediately adjacent to the TP

(1048–1108E). The large cloud water appears between

the area east of the TP and the middle reaches of the

Yanzi River, and extends easterly to ocean regions

around 1228E. Moreover, cloud water is mainly distrib-

uted below 500hPa and the isotherm of 08C (Fig. 5b),

showing that cloud water over SEC mainly consists of

liquid water. At the same time, the westerly wind over

the TP and eastern China is conducive to the transfer of

cloud water downstream.

The simulation can capture major vertical distributions

of cloud water and wind as shown in Figs. 5c and 5d.

FIG. 4. Simulated spring mean (a) SWCRE, (b) circulations, and (c) cloud water path (gm22) during the period

2001–16. In (c), the shading is total CWP (liquid and ice) and the blue contour line is liquid CWPwith an interval of

20 (gm22). The black solid line is the TP (over 3000m). The vector data are masked below 850 hPa.

3074 JOURNAL OF CL IMATE VOLUME 32

Compared to the observation, the simulated central in-

tensity and location of westerly jet are stronger and more

southerly, and the ascending motion between 108 and

158N is also too large (Fig. 5c). Besides the low-level at-

mosphere, large cloud water content is also distributed

between 300 and 400hPa and extends eastward to the

East China Sea (Fig. 5d), where this considerable high-

level cloudwater complicates cloud overlap (Fig. 4a). The

TP is a heat source in spring relative to the same latitudes

over eastern China (Yeh et al. 1957; Wu et al. 2007). As

shown in Figs. 5b and 5d, a temperature gradient from

west to east exists in themiddle atmosphere around 1048–1108E, which can increase the stability at low to midlevels

over SEC (Wu and Zhang 1998; Zhang et al. 2013).

Previous work (Yu et al. 2001, 2004; Li and Gu 2006)

suggested that the TP can reduce the westerly wind and

then low-level convergence and midlevel divergence ap-

pear on the lee side of theTP. These thermal anddynamic

roles of the TP limit updraft motion in low to midlevels

but aid the accumulation of cloud water. Because of the

above-mentioned circulation distributions, spring cloud

water over SEC is mainly located in the low to midlevels.

The work by Luo et al. (2009) also shows that low to

middle clouds during March to May are the dominant

cloud types over eastern China with the almost same

domain as this study. As a result, large amounts of liquid

cloud reflect shortwave radiation and cause strong

SWCRE over SEC.

b. Subseasonal variation

Given that some subseasonal features cannot be re-

flected in spring mean (March to May) results, we used

the pentad mean variation to further investigate the

SWCRE and relevant circulations over SEC. This study

focuses on the spring and therefore pentads 1–48 (early

January to late September) are selected for analysis in

this study. The vertical velocity between 850 and 500 hPa

is averaged to represent the low-to-midlevel vertical

motion. Ascending motion and water vapor supply are

the major circulation conditions for the formation of

large-scale clouds and rainfall. Figure 6 presents the

observed subseasonal variation of rainfall, SWCRE, and

the two general circulations mentioned above. Some

researchers (He et al. 2007, 2008; Zhao et al. 2007; Zhu

et al. 2011) have pointed out that the subtropical spring

rain (around pentads 12–24), the establishment of the

South China Sea monsoon (around pentads 25–28), and

the following northwardmovement of summermonsoon

FIG. 5. Spring mean (a) pressure–latitude cross section of cloud water content (1026 kg kg21) and wind circu-

lations averaged over 1048–1228E and (b) corresponding pressure–longitude cross section averaged over 228–328N.

In (a), the red contour is westerly wind (m s21) and the vector is the meridional (20m s21)/vertical wind (5 31023 Pa s21) component. In (b), the red contour is air temperature (8C) with intervals of 38C and vector is zonal

(m s21)/vertical wind (2 3 1023 Pa s21) component. (c),(d) As in (a) and (b), but from the WRF simulation. The

gray shading is the TP altitude.

1 JUNE 2019 L I E T AL . 3075

rain (after pentad 28) consecutively occur over Asian

regions, and the wind field and moisture then show

dramatic variations during late winter to early summer

over SEC. These features are also very clearly seen in

pentad mean variation of rainfall (Fig. 6a). Note that the

intensity of rainfall, SWCRE, ascending motion, and

water vapor convergence abruptly increases around

pentad 12 (late February) in Fig. 6, and the corre-

sponding low-level southerly wind and westerly jet

synchronously reduce (not shown). This abrupt change

almost simultaneously appears with the reversal of

temperature gradient between the Indochina Peninsula

and the western Pacific to the east of the Philippines in

late February (Tian and Yasunari 1998). After pentad

12, the low-to-midlevel ascending motion increases up

to250hPaday21 and persists until pentad 36 (late June)

in Fig. 6c. During the same period, SWCRE and water

vapor convergence gradually increase and then further

intensify starting at pentad 16 (mid-March), when

SWCRE intensity reaches up to 2110Wm22. After

pentad 24 (late April), SWCRE and water vapor con-

tinue to increase until pentad 36, and afterward reduce

quickly (Figs. 6b and 6d). The cloud water content shows

similar variations to those described above (not shown).

As listed in Table 1, the temporal correlations between

area-averaged SWCRE and updraft velocity (water va-

por convergence) during pentads 1–36 are over 0.84 in

the observation and simulation, showing that the two

circulation conditions are essential to maintain strong

SWCRE over SEC.

Figure 7 is simulated pentad mean variations. The

simulation can well reproduce subseasonal variation of

SWCRE to the north of 208N. The simulation accurately

captures the abrupt increase in SWCRE, ascending

motion, and water vapor convergence at pentad 12, and

their further intensification at pentads 16 and 24

(Figs. 7b–d). In contrast, simulated SWCRE around the

SEC latitudinal zone is weaker during pentads 1–20, and

rainfall, ascending motion, and water vapor conver-

gence to the south of 208N are stronger after pentad 24.

Moreover, large rainfall remains after pentad 36, which

is associated with strong convections from WRF cumu-

lus parameterization used in this study. In the simula-

tion, ascending motion and water vapor supply over

tropical regions to south of SEC (the South China Sea)

are also greatly enhanced since pentad 24, but the cor-

responding SWCRE is much weaker than that over

SEC. Note that large SWCRE stably exists at SEC lat-

itudinal bands (228–328N) during pentads 25–36, and

does not retreat southward and further intensify north-

ward as rainfall (Figs. 6b and 7b). This difference of

subseasonal variations between SWCRE and rainfall

is relevant to respective cloud-precipitation features

over subtropical (SEC) and tropical (South China Sea)

monsoon regions. Stratiform precipitation with large

area and long cloud duration time accounts for a

FIG. 6. Pentadmean variations of observed (a) precipitation (mmday21), (b) SWCRE (Wm22), (c) 850–500-hPa

mean ascending motion (hPa day21), and (d) column water vapor flux divergence (1024 kgm22 s21) averaged over

1048–1228E during the period of 2001–16. The x axis denotes the pentad number.

3076 JOURNAL OF CL IMATE VOLUME 32

considerable proportion over SEC and markedly in-

creases after the onset of South China Sea monsoon; by

comparison, although the convective precipitation

amount over South China Sea is dramatically enhanced

after pentad 24, its precipitation area quickly decreases,

indicative of short duration time (Hu et al. 2011). As

clearly shown in the simulation (Figs. 7a,b), the con-

vective rainfall ratio during pentads 25–36 is up to 0.9

over South China Sea, and then cloud lifetime and re-

sultant SWCRE are not as large as over SEC.

The above subseasonal variations of SWCRE are also

reflected in their regional mean time series. As shown in

Fig. 8, the observed and simulated SWCREdramatically

intensify around pentad 12 and then gradually increase.

The maximum of observed and simulated SWCRE oc-

curs in pentads 33 (mid-June) and 28 (late May), with

values of 2128 and 2120Wm22, respectively. The

earlier peak time of simulated SWCRE arises from the

high-value position bias of SWCRE, which is east in

comparison with the observation. During pentads 12–36,

LWCRE also increases and offsets with SWCRE, and

then NCRE intensity stably remains over 260Wm22.

As illustrated in Fig. 2, Fig. 8 further shows that SWCRE

over SEC is much larger than LWCRE and dominates

regional CREs. The intensity ratio of LWCRE to

SWCRE is only 0.2 before pentad 12 (Fig. 8). Afterward,

with the increase in ascending motion and high cloud,

the ratio of LWCRE gradually increases and exceeds 0.4

at around pentad 36. The simulated interannual vari-

abilities of SWCRE and NCRE are larger than the ob-

servations. Another interesting feature is that the

interannual variability of SWCRE is very close to

NCRE during pentads 1–30, indicating that SWCRE

dominates interannual variability of CREs over SEC.

Based on the pentad mean variations mentioned

above, Fig. 9 presents several important pentads to show

geographical distributions of water vapor and ascending

motion affecting SWCRE. In pentads 12, 18, and 24, the

westerly jet stays between eastern China and Japan, and

obvious moisture convergence and ascending motion

occur over SEC. Thewater vapormainly comes from the

westerly wind south to the TP and the southerly wind

TABLE 1. Temporal correlations between 850–500-hPa mean vertical

velocity (hPa day21)/column moisture divergence (1024 kgm22 s21)

and SWCRE (Wm22)/precipitation (mmday21) during pentads

1–36 for the observed and simulated data. The data are averaged

over SEC (228–328N, 1048–1228E) during the period of 2001–16.

‘‘Obs.’’ and ‘‘Sim.’’ are the abbreviations for observation and

simulation.

MAM

850–500-hPa mean

vertical velocity

Column moisture

divergence

SWCRE Obs. 0.85 0.90

Sim. 0.93 0.84

Precipitation Obs. 20.67 20.91

Sim. 20.94 20.97

FIG. 7. As in Fig. 6, but for theWRF simulation. In (a), the blue line is the ratio of convective rainfall amount to total

rainfall amount.

1 JUNE 2019 L I E T AL . 3077

from the South China Sea. At pentad 28 (mid-May)

when the South China Sea monsoon has been estab-

lished, the water vapor transport by the southwesterly

wind from the Bay of Bengal is greatly enhanced.

SWCRE over SEC then further increases since pentad

28. Afterward, the west Pacific subtropical high extends

westward and advances northward. In pentad 38, the

subtropical ridge is close to the middle and lower rea-

ches of the Yanzi River and suppresses strong updraft

motion; consequently, moisture convergence, updraft

motion, and SWCRE weaken. The simulation basically

reproduces the aforementioned circulation distributions,

but some evident biases of moisture transport exist.

Particularly, moisture transfer by the southwest air

current is weak over SEC in pentad 18 but too strong in

pentads 28, 33, and 38 over the Indochina Peninsula, the

South China Sea, and south of Japan (Fig. A1). The

overestimated moisture leads to large rainfall over

tropical regions south of SEC.

According to the results shown in Figs. 6, 7, and 9,

large SWCRE over SEC is closely associated with re-

gional water vapor supply and low-to-midlevel ascend-

ing motion, which exhibit pronounced stage features.

Around pentad 12, ascending motion, moisture supply,

and SWCRE over SEC abruptly increases, which is

caused by the reversal of thermal contrast between the

Indochina Peninsula and the western Pacific to east of

the Philippines in late February. The subsequent pentad

mean variations of SWCRE over SEC are roughly di-

vided into two stages during late winter to early summer.

In stage 1 (pentads 12–24), regional ascending motion is

mainly caused by the TP thermal and dynamic roles and

westerly jet while water vapor as cloud water supply

stably comes from the southern flank of the TP and

South China Sea. In stage 2 (pentads 25–36), SWCRE

intensity is strongly affected by the march of East Asian

monsoon systems (the tropical South China Sea mon-

soon and subtropical East Asian monsoon). Particularly,

FIG. 8. Pentad mean variations of CREs (Wm22) in the (a) observation and (b) simulation

averaged over SEC (228–328N, 1048–1228E) during the period of 2001–16. Here, the orange line

denotes the ratio of the LWCRE magnitude to that of negative SWCRE. The vertical bars

denote the standard deviation at each pentad, reflecting the interannual variability. The x axis

denotes the pentad number.

3078 JOURNAL OF CL IMATE VOLUME 32

after the onset of the South China Sea monsoon (around

pentads 24–28), the greatly intensified moisture transport

from the Bay of Bengal increases cloud water supply and

then enhances SWCRE over SEC.

c. Case studies in 2010 and 2011

Spring circulation conditions have remarkable in-

terannual variation over SEC, and regional cloud radi-

ative characteristics including SWCRE readily exhibit

interannual differences. Here, spring SWCRE and rel-

evant circulations are further examined in two real case

years using satellite and reanalyzed datasets as well as

WRF regional simulation. The most severe drought in

the last 30 years occurred persistently over SEC from

January to May 2011, but the precipitation amount was

relatively larger in 2010; the corresponding general cir-

culations showed a sharp contrast between the two years

(Sun and Yang 2012). Furthermore, these two years are

the only pair of continuous years with the occurrence of

contrasting precipitation in spring since the CERES-

MODIS satellite era. Thus, 2010 and 2011 are selected

as the strong and weak case years. During the period of

2001–16, the normalized standard deviations of spring

mean precipitation averaged over SEC are 0.92

and 22.46 in 2010 and 2011, respectively, and the cor-

responding values of SWCRE are 1.6 and 20.46, re-

spectively, showing large interannual differences in the

two case years. In this section, pentad 1–36 is selected for

the analysis of subseasonal variations.

Figure 10 presents pentad mean precipitation and

SWCRE averaged over SEC in 2010 and 2011. Com-

pared to the observation, the simulation well reproduces

subseasonal variations of rainfall and SWCREover SEC

in 2010, with temporal correlation of 0.78 and 0.86, re-

spectively (Figs. 10a and 10b). In contrast, the temporal

correlation of SWCRE in 2011 is only 0.55 because the

simulated SWCRE is much weaker before pentad 22.

During the strong year (2010), strong ascending motion

appears early, with the development height reaching up

to 400 hPa in pentads 9 and 13, and corresponds well to

large moisture convergence and cloud liquid water path

(Figs. 10b and 11a–d). SWCRE (less than2110Wm22)

and moisture convergence are relatively stable until

pentad 18; afterward, SWCRE markedly increases up

to 2130Wm22 while the moisture convergence also

gradually increases and the ascending motion becomes

continuous and deep (Figs. 10b and 11a–d). The simu-

lation also performs well in subseasonal variations of

ascending motion and moisture convergence in 2010.

During the weak year (2011), deep ascending motion

before pentad 24 is much weaker than that of 2010,

corresponding to weaker moisture convergence and

cloud liquid water path during the same period

(Figs. 11e–h). Before pentad 18 of 2011, considerable

ascending motion still appears in low to midlevels

(Fig. 11e) and helps to maintain SWCRE (Fig. 10d).

Compared to the observation in 2011, the simulated

moisture convergence before pentad 18 is close to the

FIG. 9. Geographical distributions of 200-hPa westerly jet (red dashed line; m s21), column water vapor flux (vector; kgm21 s21), and

divergence (shading; 1024 kgm22 s21) at selected pentads averaged during 2001–16. The data are from ERA-Interim. The black solid line

is the TP (over 3000m). The vector data are masked below 850 hPa.

1 JUNE 2019 L I E T AL . 3079

observation (Fig. 11g), but the simulated low-to-midlevel

ascending motion is obviously weaker (Figs. 11e,f), which

can partly explain the weaker SWCRE in the simulation.

As shown in Fig. 11, although the simulated intensity has

some deviation, the simulation basically captures the

timing of strong ascending motion in 2011. Note that the

contrast of cloud liquid water path between 2010 and

2011 is larger than cloud ice water path (Figs. 11d and

11h), indicating that cloud liquid water is more sensitive

to the ascending motion.

To examine the geographical distribution of the in-

terannual differences, the observed and simulated mean

of pentads 18–30 is selected as the spring mean period

when the simulation performs relatively better and the

interannual contrast is also large. Figure 12 presents the

SWCRE, circulations, and relevant differences aver-

aged during pentads 18–30 in 2010 and 2011. In 2010

(strong year), there is strong ascending motion and

SWCRE over SEC as well as a strong low-level south-

westerly from the Bay of Bengal and a southerly from

the South China Sea (Figs. 12a,c). In contrast, the

SWCRE and southwesterly over SEC are much weaker

in 2011 (weak year) (Figs. 12b,e). In the strong spring

rainfall case year (2010), the larger SWCRE (up

to 260Wm22) corresponds to a larger updraft velocity

(less than 230Paday21) at 500hPa over SEC, the

southwesterly (southerly) brings abundant water vapor

into SEC, and the northwesterly from Siberia and

northern China enters this region and causes a strong

low-level convergence (Figs. 12c,f). The distributions of

the regional circulation and SWCRE over SEC are

closely related to the pattern of large-scale circulation in

both case years. Relative to the pattern that occurs in

2010, the westerly jet (.40ms21) at 200 hPa and the

northwestern Pacific subtropical high are weak, and

their center positions shift eastward in 2011. This cir-

culation pattern in 2011 produces a weaker southwest-

erly, ascending motion and precipitation over SEC

(Fig. 12; see also Fig. A2 in the appendix). The strong

circulation differences between these two case years

were mainly caused by La Niña and the North Atlantic

Oscillation in 2010–11 (Sun and Yang 2012). The strong

differences between the two case years are also very

clear in the domain mean values, as listed in Table 2.

For example, the SWCRE differences are 231.6

and 235.2Wm22 for the observation and simulation,

respectively, and the vertical velocity differences at

500 hPa are 237.8 and 233.4 hPaday21, respectively

(Table 2). The observed and simulated differences of

cloud liquid water path are 45.4 and 38.0 gm22, re-

spectively and much larger than their counterparts of

cloud ice water path, with the values of 8.3 and 6.9 gm22,

respectively, showing that cloud water differences be-

tween 2010 and 2011 are mainly contributed by liquid

phase cloud.

The above results show that the ascending motion is

one of key general conditions relating the regional cir-

culation to the SWCRE. Water vapor is condensed to

cloud water through the lifting role of the ascending

motion, the maintenance of which provides stable con-

ditions for cloud formation. Thus, the domain 228–328N,

1108–1208E where large SWCRE contrast exists is

FIG. 10. Pentad mean variations of observed (black line) and simulated (blue line) (a) precipitation (mmday21)

and (b) SWCRE (Wm22) averaged over SEC (228–328N, 1048–1228E) in 2010. (c),(d) As in (a) and (b), but for the

corresponding results in 2011. The temporal correlation coefficients are marked at the top-right corner in each

subfigure. The x axis is the pentad number.

3080 JOURNAL OF CL IMATE VOLUME 32

selected to examine the scatter relationship between the

SWCRE and ascending motion at 500 hPa in pentads

18–30 of 2010 and 2011. In the observations, the larger

SWCRE (points with negative value) nearly corre-

sponds to the ascending motion (points with negative

value), and the linear correlation is up to 0.70 (Fig. 13a).

In the simulation, the value of most points is less than

zero, the linear correlation is 0.64, and the regressive

slope is also very close to the observation (Fig. 13b). In

addition to the updraft motion, water vapor convergence

FIG. 11. Pentad mean variations of (a) ERA-Interim reanalyzed vertical velocity (hPa day21), (b) simulated vertical velocity

(hPa day21), (c) column water vapor flux divergence (1024 kgm22 s21), and (d) simulated cloud water path (gm22) averaged over SEC

(228–328N, 1048–1228E) in 2010. (e)–(h) As in (a)–(d), but showing the corresponding results in 2011. Here, IWP denotes cloud ice water

path. The x axis is the pentad number.

1 JUNE 2019 L I E T AL . 3081

supplies the source of cloud water, and then the intensity

of the moisture convergence over SEC affects the re-

gional SWCRE to some extent. As for the scatter re-

lationship between the SWCRE and column moisture

convergence in the above region, the linear correlation

coefficients are 0.51 and 0.55 for the observation and

simulation, respectively (not shown). Water vapor con-

vergence still corresponds well to larger SWCRE in

the observation; however, the simulation poorly cap-

tures the good observed scatter distribution and un-

derestimates the difference of water vapor convergence

between the two case years. The above analysis shows

that the contribution of the updraft motion to cloud

water and SWCRE is likely larger than the air moisture

in the simulation. In Asian monsoon regions, water va-

por transportation is mainly driven by the regional cir-

culation (Wang et al. 2017) and the vertical motion is the

prominent influencing factor for the spring moisture

budget over SEC (Li et al. 2018). In this sense, theWRF

model has broadly reproducibility in interannual dif-

ferences of above cloud radiative characteristics over

SEC. According to the results in Figs. 10–13, favorable

spring general circulations cause stronger ascending

motion over SEC in the strong case year, and then

produce more cloud liquid water and larger SWCRE

relative to the weak case year. Consequently, the

FIG. 12. Geographical distributions of (a) CERES SWCRE (shading; Wm22), vertical velocity (Pa day21) at 500 hPa, and wind field at

850 hPa averaged during the period of pentads 18–30 in 2010, (b) the corresponding results for 2011, and (c) the differences between 2010

and 2011. (d)–(f) As in (a)–(c), but for the corresponding results from the WRF simulation. The black solid line is the TP (over 3000m).

The vector data are masked below 850 hPa.

TABLE 2. Observed and simulated cloud radiation and meteorological variables averaged over SEC (228–328N, 1048–1228E) during the

18th–30th pentad in 2010 and 2011. Here, the differences are between the corresponding values in 2010 and 2011.

18th–30th pentad mean 2010 2011 2010 2 2011

SWCRE (Wm22) Obs. 2127.9 296.3 231.6

Sim. 2126.2 291.0 235.2

Cloud liquid water path (gm22) Obs. 139.9 94.5 45.4

Sim. 73.1 35.1 38.0

Cloud ice water path (gm22) Obs. 37.6 29.3 8.3

Sim. 41.4 34.5 6.9

Total cloud water path (gm22) Obs. 177.5 123.8 53.7

Sim. 114.5 69.5 45.0

W500 (hPa day21) Obs. 231.6 6.2 237.8

Sim. 228.2 5.2 233.4

Water vapor divergence (1024 kgm22 s21) Obs. 20.641 20.185 20.456

Sim. 20.746 20.522 20.224

3082 JOURNAL OF CL IMATE VOLUME 32

interannual differences of SWCRE in other cases very

likely contribute to the anomalous ascending motion

over SEC.

4. Conclusions and discussion

This study shows that SWCRE over SEC (228–328N,

1048–1228E) exhibits strong and persistent features in

spring. As for the spring (March–May) mean, the in-

tensity of SWCRE over SEC is up to2110Wm22, being

the largest at the same latitude bands of the Northern

Hemisphere. In this region, strong spring SWCRE

is mainly caused by large and stable cloud liquid

water, which is closely associated with persistent water

vapor transport and low-to-midlevel ascending motion.

SWCRE over SEC also has distinctive subseasonal

variation and its high value stays over SEC during

pentads 12–36 (late February to late June). Around

pentad 12, ascent motion, moisture supply, and SWCRE

over SEC abruptly increases, which is accompanied by

the reversal of thermal contrast between the Indochina

Peninsula and western Pacific to east of the Philippines

on late February. During pentads 12–24 (late February

to late April), the TP thermal and dynamic roles and

high-level westerly jet provide appropriate settings for

the stable occurrence of continuous ascending motion

over SEC, while water vapor is transferred from south of

the TP and the South China Sea. During pentads 25–36

(earlyMay to late June), SWCRE is strongly affected by

the march of East Asian monsoon systems (the tropical

South China Sea monsoon and subtropical East Asian

monsoon) and its intensity further increases. Particu-

larly, after the onset of South China Sea monsoon

(around pentads 24–28), the greatly intensified moisture

transport from the Bay of Bengal increases cloud water

supply and then enhances SWCRE over SEC. After

pentad 36, the above circulations favoring the mainte-

nance of cloud water and SWCRE quickly weaken and

then large SWCRE cannot remain over SEC. These

subseasonal results indicate that the spring SWCRE

over SEC is not just affected by specific circulations

(e.g., the dynamical and thermal roles of the TP, west-

erly jet) but is also deeply influenced by the seasonal

march of the East Asian monsoon system. In the observed

case analysis, stronger ascending motion, larger amounts

of cloud liquid water, and SWCRE occur during a strong

spring rainfall year, and the differences in SWCRE be-

tween strong and weak spring rainfall years are well

related to the strengths of the anomalous ascending

motion over SEC.

The regional model (WRF) can basically reproduce

large spring SWCRE over SEC and its persistence

during late winter to early summer. In contrast, the

simulated association of SWCRE with ascending mo-

tion is closer than water vapor convergence. As shown

in Figs. 9 and A1, considerable simulation biases exist

over the East Asian monsoon regions regarding the

geographical distribution and intensity of moisture

transport and convergence. Similar deficiencies of

moisture simulation are also common in current state-

of-the-art climate models (Sperber et al. 2013; Yang

et al. 2015) and closely relevant to the uncertainties of

FIG. 13. Scatterplot of SWCRE (Wm22) and vertical velocity (Pa day21) at 500 hPa over SEC (228–328N, 1108–1208E). Here, the SWCRE (vertical velocity) is the difference between the 18th–30th pentad mean results in 2010

and 2011 (as shown in Figs. 11c and 11f). The grid over SEC is selected to plot the scatter figure. The regression

equation and linear correlation coefficient are shown in the lower right of each panel.

1 JUNE 2019 L I E T AL . 3083

physics parameterizations (e.g., convection, cloud

precipitation, and planet boundary layer) and numer-

ical algorithms. The complicated topography in the TP

and surrounding areas further increases above simu-

lation difficulties. To well capture the roles of water

vapor in spring SWCRE over SEC, more model work is

needed to carefully select and tune model parameter-

ization schemes regarding moisture and cloud water.

Both observation and simulation show that sub-

seasonal evolution of SWCRE is not exactly same with

rainfall over SEC and South China Sea, reflecting that

ambient conditions influencing SWCRE and rainfall are

different over these regions. Besides cloud water and

associated regional circulations, SWCRE is also sensitive

to the microcloud radius, which is highly modulated by

aerosols as cloud condensation nuclei (Gettelman and

Sherwood 2016). A large number of aerosols are dis-

tributed over SEC and readily produce remarkable con-

trasts of cloud microphysical characteristics (e.g., cloud

number concentration and radius) and radiative proper-

ties relative to the adjacent oceans (Li et al. 2016). Thus,

more cloud microphysics analysis and well-designed

simulations should be conducted to explore climatic

mechanisms resulting in the maintenance of spring

SWCRE over SEC.

The case studies clearly show that circulations (e.g., the

East Asian westerly jet, the northwestern Pacific sub-

tropical high, and the low-latitude tropical circulation)

FIG. A1. As in Fig. 9, but from the WRF simulation.

FIG. A2. Observational distribution of the 850-hPa wind (vector), zonal wind speed (red dashed line; m s21) at

200 hPa, and precipitation (shading; mmday21) averaged during the 18th–30th pentads of (a) 2010 and (b) 2011.

The pentad number is marked at the top-right corner in each subfigure. The black solid line is the TP (over 3000m).

The vector data are masked below 850 hPa.

3084 JOURNAL OF CL IMATE VOLUME 32

around East Asia have obvious interannual variations,

which affect the distribution, strength, and variability of

CREs over SEC. To identify individual roles of these

circulations in regional SWCRE, it is very necessary to

quantitatively examine the associations between these

circulations and SWCRE in the interannual scale. Addi-

tionally, CERES satellite and ERA-Interim reanalysis

data used in this study still have some uncertainties in

describing regional cloud–radiation variables, especially

in their quantitative aspects. As for further work, more

comparative analysis with model and observational data

should be conducted to obtain reasonable physical con-

sistency results for East Asian cloud–radiation processes.

Acknowledgments. This research was jointly sup-

ported by the National Key R&D Program of China

(2017YFA0603503), the National Science Foundation

of China (91737306, 41730963, and 41876020), the Pri-

ority Research Program of the Chinese Academy of

Sciences (QYZDY-SSW-DQC018), the Strategic Pri-

ority Research Program of Chinese Academy of Sci-

ences (Grant XDA17010105), and a grant (toUniversity

at Albany) from the Office of Science (BER), U.S.

Department of Energy. CERES-MODIS cloud prop-

erties and CERESEBAF products used in this study are

produced by the NASA CERES Team, available at

http://ceres.larc.nasa.gov.

APPENDIX

Geographical Distribution of Circulations in theSimulation and Case Years

FigureA1 shows 200-hPa zonalwind speed, columnwater

vapor flux, and divergence simulated from WRF Model

at selected pentads averaged during 2001–16. Figure A1

clearly shows the simulated biases of air moisture regarding

its spatial location and intensity. The possible reasons are

discussed in the main text. To show the differences of large-

scale circulations in two case years, Fig.A2presents 850-hPa

wind, 200-hPa zonal wind speed, and precipitation aver-

aged from the 18th–36th pentads of 2010 and 2011.

REFERENCES

Adler, R. F., and Coauthors, 2003: The version-2 Global Precipitation

Climatology Project (GPCP) Monthly Precipitation Analysis

(1979–present). J. Hydrometeor., 4, 1147–1167, https://doi.org/10.1175/1525-7541(2003)004,1147:TVGPCP.2.0.CO;2.

Bony, S., and Coauthors, 2015: Clouds, circulation and climate sensi-

tivity. Nat. Geosci., 8, 261–268, https://doi.org/10.1038/ngeo2398.Boucher, O., and Coauthors, 2013: Clouds and aerosols. Climate

Change 2013: The Physical Science Basis. T. F. Stocker et al.,

Eds., Cambridge University Press, 571–657.

Cess, R. D., and Coauthors, 1990: Intercomparison and in-

terpretation of climate feedback processes in 19 atmospheric

general circulation models. J. Geophys. Res., 95, 16 601–16 615,

https://doi.org/10.1029/JD095iD10p16601.

Chen, F., and J. Dudhia, 2001: Coupling an advanced land surface–

hydrology model with the Penn State–NCAR MM5 model-

ing system. Part I: Model implementation and sensitivity.

Mon. Wea. Rev., 129, 569–585, https://doi.org/10.1175/

1520-0493(2001)129,0569:CAALSH.2.0.CO;2.

Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis:

Configuration and performance of the data assimilation system.

Quart. J. Roy. Meteor. Soc., 137, 553–597, https://doi.org/

10.1002/qj.828.

Doelling, D. R., and Coauthors, 2013: Geostationary enhanced

temporal interpolation for CERES flux products. J. Atmos.

Oceanic Technol., 30, 1072–1090, https://doi.org/10.1175/

JTECH-D-12-00136.1.

Flato, G., and Coauthors, 2013: Evaluation of climate models.

Climate Change 2013: The Physical Science Basis, T. F. Stocker

et al., Eds., Cambridge University Press, 571–658.

Gettelman, A., and S. C. Sherwood, 2016: Processes responsible for

cloud feedback. Curr. Climate Change Rep., 2, 179–189,

https://doi.org/10.1007/s40641-016-0052-8.

Guo, Z., T. Zhou, M. Wang, and Y. Qian, 2015: Impact of cloud

radiative heating on East Asian summer monsoon circulation.

Environ. Res. Lett., 10, 074014, https://doi.org/10.1088/1748-

9326/10/7/074014.

He, J. H., and Coauthors, 2007: Reinvestigations on the East Asian

subtropical monsoon and tropical monsoon (in Chinese).

Chin. J. Atmos. Sci., 31, 1257–1265.

——, P. Zhao, C. W. Zhu, R. H. Zhang, X. Tang, L. X. Chen, and

X. J. Zhou, 2008: Discussions on the East Asian subtropical

monsoon. Acta Meteor. Sin., 66, 683–696.

Holton, J. R., and G. J. Hakim, 2012: An Introduction to Dynamic

Meteorology. Elsevier, 532 pp.

Hong, S. Y., and J. Lim, 2006: The WRF Single-Moment 6-Class

Microphysics Scheme (WSM6). J. Korean Meteor. Soc., 42,

129–151.

——, Y. Noh, and J. Dudhia, 2006: A new vertical diffusion package

with an explicit treatment of entrainment processes.Mon. Wea.

Rev., 134, 2318–2341, https://doi.org/10.1175/MWR3199.1.

Hu, L., Y.D. Li, Y. Song, andD. F.Deng, 2011: Seasonal variability

in tropical and subtropical convective and stratiform pre-

cipitation of theEastAsianmonsoon. Sci. ChinaEarth Sci., 54,

1595, https://doi.org/10.1007/s11430-011-4225-y.

Huang, D.-Q., J. Zhu, Y.-C. Zhang, J. Wang, and X.-Y. Kuang,

2015: The impact of the East Asian subtropical jet and polar

front jet on the frequency of spring persistent rainfall over

southern China in 1997–2011. J. Climate, 28, 6054–6066,

https://doi.org/10.1175/JCLI-D-14-00641.1.

——,——,——, Y. Huang, and X.-Y. Kuang, 2016: Assessment of

summer monsoon precipitation derived from five reanalysis

datasets over East Asia. Quart. J. Roy. Meteor. Soc., 142,

108–119, https://doi.org/10.1002/qj.2634.

Huffman, G. J., R. F. Adler, D. T. Bolvin, and G. Gu, 2009: Im-

proving the global precipitation record: GPCP version 2.1.

Geophys. Res. Lett., 36, L17808, https://doi.org/10.1029/

2009GL040000.

Iacono, M. J., J. S. Delamere, E. J. Mlawer, M. W. Shephard, S. A.

Clough, and W. D. Collins, 2008: Radiative forcing by long-

lived greenhouse gases: Calculations with the AER radiative

transfer models. J. Geophys. Res., 113, D13103, https://doi.org/

10.1029/2008JD009944.

1 JUNE 2019 L I E T AL . 3085

Kain, J. S., 2004: The Kain–Fritsch convective parameterization:

An update. J. Appl. Meteor., 43, 170–181, https://doi.org/

10.1175/1520-0450(2004)043,0170:TKCPAU.2.0.CO;2.

Kim, E. J., and S. Y. Hong, 2010: Impact of air–sea interaction on

East Asian summer monsoon climate in WRF. J. Geophys.

Res., 115, D19118, https://doi.org/10.1029/2009JD013253.

Klein, S. A., and D. L. Hartmann, 1993: The seasonal cycle of low

stratiform clouds. J. Climate, 6, 1587–1606, https://doi.org/

10.1175/1520-0442(1993)006,1587:TSCOLS.2.0.CO;2.

Lau, K.-M., and S. Yang, 1997: Climatology and interannual vari-

ability of the Southeast Asian summer monsoon.Adv. Atmos.

Sci., 14, 141–162, https://doi.org/10.1007/s00376-997-0016-y.

Li, J., and R. Yu, 2014: Characteristics of cold season rainfall over

the Yungui Plateau. J. Climate Appl. Meteor., 53, 1750–1759,

https://doi.org/10.1175/JAMC-D-13-0285.1.

Li, J. D., and J. Y. Mao, 2015: The preliminary evaluation of

global and East Asian cloud radiative effects in reanalyses.

Atmos. Oceanic Sci. Lett., 8, 100–106, https://doi.org/10.1080/

16742834.2015.11447245.

——, W.-C. Wang, X. Dong, and J. Mao, 2017a: Cloud–radiation–

precipitation associations over the Asian monsoon region: An

observational analysis. Climate Dyn., 49, 3237–3255, https://

doi.org/10.1007/s00382-016-3509-5.

——, J. Y. Mao, and F. Wang, 2017b: Comparative study of five

current reanalyses in characterizing total cloud fraction and

top-of-the-atmosphere cloud radiative effects over the Asian

monsoon region. Int. J. Climatol., 37, 5047–5067, https://

doi.org/10.1002/joc.5143.

Li, P. X., T. J. Zhou, and X. Chen, 2018: Water vapor transport for

spring persistent rains over southeastern China based on five

reanalysis datasets. Climate Dyn., 51, 4243–4257, https://

doi.org/10.1007/s00382-017-3680-3.

Li, Y., D. W. J. Thompson, and S. Bony, 2015: The influence of

atmospheric cloud radiative effects on the large-scale atmo-

spheric circulation. J. Climate, 28, 7263–7278, https://doi.org/

10.1175/JCLI-D-14-00825.1.

Li, Y. Y., andH.Gu, 2006: Relationship betweenmiddle stratiform

clouds and large scale circulation over eastern China.Geophys.

Res. Lett., 33, L09706, https://doi.org/10.1029/2005GL025615.

Li, Z. Q., and Coauthors, 2016: Aerosol and monsoon climate in-

teractions over Asia. Rev. Geophys., 54, 866–929, https://

doi.org/10.1002/2015RG000500.

Liang, X.-Z., and W.-C. Wang, 1998: Associations between China

monsoon rainfall and tropospheric jets.Quart. J. Roy. Meteor.

Soc., 124, 2597–2623, https://doi.org/10.1002/qj.49712455204.

LinHo, X. Huang, and N.-C. Lau, 2008:Winter-to-spring transition

in East Asia: A planetary-scale perspective of the south China

spring rain onset. J. Climate, 21, 3081–3096, https://doi.org/

10.1175/2007JCLI1611.1.

Loeb, N. G., and Coauthors, 2009: Toward optimal closure of the

Earth’s top-of-atmosphere radiation budget. J. Climate, 22,

748–766, https://doi.org/10.1175/2008JCLI2637.1.

Luo, Y., R. Zhang, and H. Wang, 2009: Comparing occurrences

and vertical structures of hydrometeors between eastern

China and the Indian monsoon region using CloudSat/

CALIPSO data. J. Climate, 22, 1052–1064, https://doi.org/

10.1175/2008JCLI2606.1.

Pan, W., J. Mao, and G. Wu, 2013: Characteristics and mechanism

of the 10–20-day oscillation of spring rainfall over southern

China. J. Climate, 26, 5072–5087, https://doi.org/10.1175/

JCLI-D-12-00618.1.

Ramanathan,V.,R.D.Cess,E.F.Harrison,P.Minnis,B.R.Barkstrom,

E. Ahmad, and D. Hartmann, 1989: Cloud-radiative forcing and

climate: Results from the Earth Radiation Budget Experiment.

Science, 243, 57–63, https://doi.org/10.1126/science.243.4887.57.

Rogers, R. R., and M. K. Yau, 1989: A Short Couse in Cloud

Physics. Butterworth-Heinemann, 304 pp.

Simmons, A. J., P. Poli, D. P. Dee, P. Berrisford, H. Hersbach,

S. Kobayashib, and C. Peubey, 2014: Estimating low-

frequency variability and trends in atmospheric temperature

using ERA-Interim.Quart. J. Roy. Meteor. Soc., 140, 329–353,

https://doi.org/10.1002/qj.2317.

Sperber, K. R., H. Annamalai, I. S. Kang, A. Kitoh, A. Moise,

A. Turner, B. Wang, and T. Zhou, 2013: The Asian summer

monsoon: An intercomparison of CMIP5 vs. CMIP3 simula-

tions of the late 20th century. Climate Dyn., 41, 2711–2744,

https://doi.org/10.1007/s00382-012-1607-6.

Stephens, G. L., 2005: Cloud feedbacks in the climate system: A

critical review. J. Climate, 18, 237–273, https://doi.org/10.1175/

JCLI-3243.1.

Sun, C., and S. Yang, 2012: Persistent severe drought in southern

China during winter–spring 2011: Large-scale circulation

patterns and possible impacting factors. J. Geophys. Res., 117,

D10112, https://doi.org/10.1029/2012JD017500.

Tian, S. F., and T. Yasunari, 1998: Climatological aspects and

mechanism of spring persistent rains over central China.

J. Meteor. Soc. Japan, 76, 57–71, https://doi.org/10.2151/

jmsj1965.76.1_57.

Wan, R. J., andG. X.Wu, 2007:Mechanism of the spring persistent

rains over southeastern China. Sci. China, 50D, 130–144,

https://doi.org/10.1007/s11430-007-2069-2.

Wang, F., S. Yang, and T. Wu, 2014: Radiation budget biases

in AMIP5 models over the East Asian monsoon region.

J. Geophys. Res. Atmos., 119, 13 400–13 426, https://doi.org/

10.1002/2014JD022243.

Wang, W.-C., W. Gong, W.-S. Kau, C.-T. Chen, H.-H. Hsu, and

C.-H. Tu, 2004: Characteristics of cloud radiation forcing over

east China. J. Climate, 17, 845–853, https://doi.org/10.1175/

1520-0442(2004)017,0845:COCRFO.2.0.CO;2.

Wang, Z., A. Duan, andG.Wu, 2014: Time-lagged impact of spring

sensible heat over the Tibetan Plateau on the summer rainfall

anomaly in East China: Case studies using the WRF model.

Climate Dyn., 42, 2885–2898, https://doi.org/10.1007/s00382-

013-1800-2.

——, ——, S. Yang, and K. Ullah, 2017: Atmospheric moisture

budget and its regulation on the variability of summer pre-

cipitation over the Tibetan Plateau. J. Geophys. Res. Atmos.,

122, 614–630, https://doi.org/10.1002/2016JD025515.

Wild, M., D. Folini, C. Schar, N. Loeb, E. G. Dutton, and G. K.

Langlo, 2013: The global energy balance from a surface per-

spective. Climate Dyn., 40, 3107–3134, https://doi.org/10.1007/

s00382-012-1569-8.

Wu, G. X., and Y. S. Zhang, 1998: Tibetan Plateau forcing and

timing of the monsoon onset in South Asia and South China

Sea. Mon. Wea. Rev., 126, 913–927, https://doi.org/10.1175/

1520-0493(1998)126,0913:TPFATT.2.0.CO;2.

——, and Coauthors, 2007: The influence of mechanical and ther-

mal forcing by the Tibetan Plateau on Asian climate.

J. Hydrometeor., 8, 770–789, https://doi.org/10.1175/JHM609.1.

Wu, X. F., and J. Y.Mao, 2016: Interdecadal modulation of ENSO-

related spring rainfall over South China by the Pacific decadal

oscillation. Climate Dyn., 47, 3203–3220, https://doi.org/

10.1007/s00382-016-3021-y.

Yang, B., Y. Zhang, Y. Qian, A. Huang, and H. Yan, 2015: Cali-

bration of a convective parameterization scheme in the WRF

model and its impact on the simulation of East Asian summer

3086 JOURNAL OF CL IMATE VOLUME 32

monsoon precipitation. Climate Dyn., 44, 1661–1684, https://

doi.org/10.1007/s00382-014-2118-4.

Yeh, T. C., S. W. Luo, and P. C. Chu, 1957: The wind structure

and heat balance in the lower troposphere over Tibetan

Plateau and its surroundings. Acta Meteor. Sin., 28, 108–

121.

Yin, J. F., D. H. Wang, H. B. Xu, and G. Q. Zhai, 2015: An in-

vestigation into the three-dimensional cloud structure over

East Asia from the CALIPSO-GOCCP data. Sci. China

Earth Sci., 58, 2236–2248, https://doi.org/10.1007/s11430-015-

5205-4.

Yin, X., A. Gruber, and P. Arkin, 2004: Comparison of the GPCP

and decadeAP merged gauge–satellite monthly precipitation

products for the period 1979–2001. J. Hydrometeor., 5, 1207–

1222, https://doi.org/10.1175/JHM-392.1.

Yu, R. C., Y. Q. Yu, and M. H. Zhang, 2001: Comparing cloud

radiative properties between the eastern China and the Indian

monsoon region. Adv. Atmos. Sci., 18, 1090–1102, https://

doi.org/10.1007/s00376-001-0025-1.

——, B. Wang, and T. J. Zhou, 2004: Climate effects of the deep

continental stratus clouds generated by the Tibetan Plateau.

J. Climate, 17, 2702–2713, https://doi.org/10.1175/1520-

0442(2004)017,2702:CEOTDC.2.0.CO;2.

Yuan, W., R. Yu, M. Zhang, W. Lin, H. Chen, and J. Li, 2012:

Regimes of diurnal variation of summer rainfall over

subtropical East Asia. J. Climate, 25, 3307–3320, https://

doi.org/10.1175/JCLI-D-11-00288.1.

Zhang, M., and Coauthors, 2005: Comparing clouds and their

seasonal variations in 10 atmospheric general circulation

models with satellite measurements. J. Geophys. Res., 110,

D15S02, https://doi.org/10.1029/2004JD005021.

Zhang, Y., R. Yu, J. Li, W. Yuan, and M. Zhang, 2013: Dynamic

and thermodynamic relations of distinctive stratus clouds on the

lee side of the Tibetan Plateau in the cold season. J. Climate, 26,

8378–8391, https://doi.org/10.1175/JCLI-D-13-00009.1.

Zhao, P., R. Zhang, J. Liu, X. Zhou, and J. He, 2007: Onset of

southwesterly wind over eastern China and associated atmo-

spheric circulation and rainfall. Climate Dyn., 28, 797–811,https://doi.org/10.1007/s00382-006-0212-y.

Zhu, C. W., X. J. Zhou, P. Zhao, L. X. Chen, and J. H. He, 2011:

Onset of East Asian subtropical summer monsoon and rainy

season in China. Sci. China Earth Sci., 54, 1845–1853, https://doi.org/10.1007/s11430-011-4284-0.

Zhu, Q. G., J. R. Lin, S. W. Shou, and D. S. Tang, 2007: Weather

Principles and Methods (in Chinese). Meteorological Press,

649 pp.

1 JUNE 2019 L I E T AL . 3087