On the summertime planetary boundary layer with

7 downloads 0 Views 6MB Size Report
Amenu, G. G., P. Kumar, and X. Z. Liang, 2005: Interannual variability of deep-layer ..... Analysis of GPS Radiosonde RS92 Performance. ... 702. 2172-2176, doi:10.1007/s11431-014-5639-5. 703. Zhou, X., and B. Geerts, 2013: The influence of ...
1

On the summertime planetary boundary layer with

2

different thermodynamic stability in China: A

3

radiosonde perspective

4

Wanchun Zhang1#,Jianping Guo1#*,Yucong Miao1,Huan Liu1,Yu Song2,

5

Zhang Fang3, Jing He1, Mengyun Lou1, Yan Yan1, Yuan Li1, Panmao Zhai1* 1

6

State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing 100081, China

7 2

8

Department of Environmental Science, Peking University, Beijing, 100871, China

9 10

State Key Joint Laboratory of Environmental Simulation and Pollution Control,

3

Public Meteorological Service Center, China Meteorological Administration, Beijing 100081, China

11 12 13 14 15

# These co-authors contribute equally to this work.

16

Correspondence to:

17

Dr./Prof. Jianping Guo, Email: [email protected]

18

Prof. Panmao Zhai, Email: [email protected]

19

Key Points:

20

l

neutral PBL types across China for the first time

21 22

l

The PBL height is positively (negatively) associated with near-surface temperature (humidity) for both convective and neutral PBLs

23 24

PBL height is estimated from afternoon soundings for convective, stable and

l

Clouds tend to cause more stable atmosphere, resulting in lower PBL height 1

25

Abstract

26

Strongly influenced by thermodynamic stability, the planetary boundary layer (PBL) is

27

key to the exchange of heat, momentum, and moisture between the ground surface and

28

free troposphere. The PBL with different thermodynamic stability across the whole

29

China, however, is not yet well understood. In this study, the occurrence frequency and

30

spatial distribution of convective boundary layer (CBL), neutral boundary layer (NBL),

31

and stable boundary layer (SBL) were systematically investigated, based on intensive

32

summertime soundings launched at 1400 Beijing time (BJT) throughout the China’s

33

Radiosonde Network (CRN) for the period 2012 to 2016. Overall, the occurrences of

34

CBL, NBL, and SBL account for 70%, 26%, and 4%, respectively, suggesting CBL

35

dominates in summer throughout China. In terms of the spatial pattern of PBL height,

36

a prominent north-south gradient can be found with higher PBL height in Northwest

37

China. Besides, the PBL heights of CBL and NBL were found to be positively

38

(negatively) associated with near-surface air temperature (humidity), whereas no

39

apparent relationship was found for SBL. Furthermore, clouds tend to reduce the

40

occurrence frequency, irrespective of PBL types. Roughly 70% of SBL cases occur

41

under overcast conditions, much higher than those for NBL and CBL, indicating that

42

clouds govern to some extent the occurrence of SBL. In contrast, except for the

43

discernible changes in PBL height under overcast condition relative to that under clear-

44

sky conditions, the changes in PBL height under partly cloudy condition are no more

45

than 170 m for both NBL and CBL types.

46

Keywords: Radiosonde, convective PBL, PBL height, thermodynamic stability, China 2

47

1. Introduction

48

The planetary boundary layer (PBL), the lowest layer of troposphere close to the

49

surface, plays an important role in air pollution, weather and climate (Garratt 1992;

50

1994; Stull 1988) . The turbulent PBL is also the main place where the vertical exchange

51

of momentum, heat, moisture, and atmospheric pollutants occurs between ground

52

surface and free troposphere (e.g., Garratt 1992; Oke 2002; Hu et al. 2010; Zhang et

53

al. 2015; Guo et al. 2016a). The turbulent motions in the PBL are responsible for the

54

mixing processes in the atmosphere, which affects the vertical redistribution of moisture

55

and momentum. This in turn impacts on the formation and evolution of boundary-layer

56

clouds such as stratocumulus and cumulus (e.g., Paluch and Lenschow 1991; Eltahir

57

1996; Amenu and Kumar 2005; Sherwood et al. 2014; Eitan et al. 2017; Zhang et al.

58

2017).

59

The PBL is increasingly recognized as one of important parameters that strongly

60

influence the air quality (Guo et al. 2009; Sandip and Martial 2015; Li et al. 2017),

61

partly account for the varying regional warming rate (Davy and Esau 2016), and

62

significantly affect global weather and climate system (Medeiros et al. 2005; Esau et

63

al. 2010; Fletcher et al. 2016; Solomon et al. 2017). Previous studies show that the

64

heating and cooling of ground surface, cloudiness, and the variability in surface

65

characteristics could affect the evolution of the PBL (e.g., Garratt 1994; Miao et al.

66

2015a; 2017). The PBL height during the daytime is found to be significantly negatively

67

correlated with surface relative humidity, but positively correlated with surface

68

temperature (Zhang et al. 2013). Likewise, on the seasonal and annual timescales, the 3

69

PBL height is positively associated with surface temperature and 10-m wind speed, but

70

negatively associated with surface pressure (Seidel et al. 2012; Guo et al. 2016b).

71

The PBL height is traditionally determined from the vertical profiles of temperature,

72

humidity, and wind from atmospheric soundings. In addition, several other methods

73

based on new data sources (e.g., lidar, doppler radar, and sodar) have been developed to

74

elucidate the PBL structure/processes (e.g., Holzworth 1964; Coulter 1979; Beyrich

75

1997; Steyn et al. 1999; Seibert 2000; Asimakopoulos et al. 2004; Hennemuth and

76

Lammert 2006; Lammert and Bösenberg 2006; Dandou et al. 2009; Mao et al. 2009;

77

McGrath-Spangler and Denning 2012; Chan and Wood 2013; Sawyer and Li 2013;

78

Zhang et al. 2014; Guo et al. 2016b; Zhang et al. 2016).

79

Based on the atmospheric soundings collected in several major field campaigns, the

80

diurnal variation of three dominant PBL types, namely convective boundary layer

81

(CBL), stable boundary layer (SBL), and neutral boundary layer (NBL) in the

82

continental U.S., were documented (Liu and Liang 2010): the CBL mainly occurs in

83

the daytime with a peak at 1500 local time, in contrast to the SBL dominating at

84

nighttime. Seidel et al. (2010) calculated the PBL height by applying seven commonly

85

methods to the profiles of temperature, potential temperature, virtual potential

86

temperature, relative humidity, specific humidity, and refractivity, and found there

87

existed large uncertainties with respect to various methods to estimate PBL height .

88

Then the climatology of the PBL height over Europe and the continental U.S. were

89

compiled using the bulk Richardson number method (Seidel et al. 2012). The same

90

method has also been utilized by Guo et al. (2016b) to develop a radiosonde-based PBL 4

91

height climatology in China, which exhibits large diurnal, seasonal, and spatial

92

variations. This could be at least in part in association with the strong variations in

93

aerosol pollution in China (Guo et al. 2011; Guo et al. 2016c) and atmospheric

94

thermodynamic stability (Tokinaga et al. 2006; Tang et al. 2016). Although the PBL

95

height over China has been well characterized from space-borne lidars (e.g., Liu et al.

96

2015; Zhang et al. 2016) and radiosondes (Guo et al., 2016b), most of other PBL

97

characteristics remain poorly understood, including the thermodynamic stability and its

98

associated meteorological factors.

99

Since 2011, intensive radiosonde observational campaigns at 1400 Beijing time (BJT)

100

in summer (May-June-July-August) have been performed at most sites of the newly-

101

updated L-band China Radiosonde Network (CRN) as operated by the China

102

Meteorological Administration (CMA) (Guo et al. 2016b; Zhang et al. 2016). The

103

afternoon soundings provide us a unique opportunity to look into the thermodynamic

104

stability of PBL. Thus, one of the main objectives of this study is to unravel the

105

occurrence frequency and spatial distribution of CBL, SBL, and NBL across China

106

using long-term fine-resolution atmospheric soundings. Another objective is to

107

understand the relationships between meteorological variables and PBL height that

108

differs by three PBL types.

109

The rest of the present paper is organized as follows. Section 2 describes the data and

110

method used in this study, as well as the uncertainty analysis for the PBL height

111

retrievals. Section 3 presents the climatology of different PBL types in China, and its

112

association with other atmospheric variables. In section 4, detailed discussions with 5

113

regard to the PBL height climatology and the potential influential factors on the

114

development of PBL will be provided. Finally, the key findings are summarized in

115

section 5.

116

2. Data and Methods

117

2.1 Radiosonde and meteorological data

118

The L-band radiosondes of CRN are conventionally launched twice a day at 0800

119

and 2000 BJT, and additional soundings are made at 1400 BJT in summer (May-June-

120

July-August) at most radiosonde sites mainly for improving the capability of predicting

121

high-impact weather in China. The L-band radiosondes are produced by the Nanjing

122

Daqiao Co. LTD., and the Shanghai Changwang Meotech Co. LTD. (Xu et al. 2007),

123

which provide fine resolution (1s) profiles of temperature, humidity, wind speed and

124

wind direction (Guo et al. 2016b; Zhang et al. 2016). As the previous studies (Tao et al.

125

2006; Xing et al. 2009; Bian et al. 2011; Ma et al. 2011) pointed out, the accuracy of

126

the temperature profiles within lower portion of troposphere is similar to that of GPS

127

RS92 radiosonde (produced by Vaisala), which is less than 0.1 K. Such good data

128

accuracy provides us enough confidence to investigate PBL climatology in China using

129

these L-band soundings of CRN.

130

In this study, the characteristics of afternoon PBL during summer were investigated

131

using the 1400 BJT soundings from 2012 to 2016 collected from CRN, due to most of

132

the exchanges of energy, moisture and air pollutants occurring in the PBL in the

133

afternoon (Garratt 1994; Guo et al. 2016b). Considering the time consistency, the near6

134

surface meteorological observations from the collocated radiosonde sites are used to

135

understand the factors governing the PBL development, including near-surface

136

temperature (Tsfc), near-surface relative humidity (RHsfc), near-surface pressure (Psfc)

137

and 10-m wind speed (WS10). The former three meteorological factors were calculated

138

by taking average over the five lowest consecutive soundings starting from the ground

139

surface, whereas WS10 calculated by taking average over the wind profiles from the

140

second lowest to third lowest soundings.

141

Given the substantial difficulties and uncertainties in determining the PBL types

142

caused by precipitation, the 1400 BJT sounding data have been excluded for the days

143

when precipitation occurred from 1200 BJT to 1400 BJT. Additionally, the sites that

144

have less than 30 valid soundings at 1400 BJT from 2012 to 2016 are not considered.

145

As a result, there are 5399 valid summertime soundings collected from the 92

146

radiosonde sites of CRN at 1400 BJT. The total cloud cover (CLD) data measured at

147

1400 BJT at the same radiosonde sites have been used, in addition to the hourly rain

148

gauge measurements recorded at the hours from 1200 BJT to 1400 BJT.

149

To understand the impacts of land surface processes on the development of PBL in

150

the afternoon, the dataset from Global Land Data Assimilation System (GLDAS, Rodell

151

et al., 2004) were analyzed. The GLDAS contains a series of geophysical parameters

152

reflecting the state of land surface (e.g., soil moisture and surface temperature) and flux

153

(e.g., evaporation and sensible heat flux), based on four land surface models (CLM,

154

Mosaic, Noah and VIC). The surface sensible heat fluxes (SHFsfc) and soil moisture

155

fields in China derived from the Noah model (Chen and Dudhia 2001) were used as 7

156

157

well.

2.2 Determination of PBL type

158

During the daytime, due to the sufficient solar radiation reaching the land surface,

159

the CBL typically dominates over land, which can reach a few kilometers in the

160

afternoon (Chen and Houze 1997). By contrast, when the surface cools by the nocturnal

161

radiation processes, the SBL takes over the bottom portion of the troposphere ( Zhang

162

et al. 2011; Miao et al. 2015b). The residual layer is neutrally stratified, resulting in

163

turbulence that is nearly of equal intensity in all directions ( Sivaraman et al. 2013;

164

Blay-Carreras et al. 2014), referred to as the NBL.

165

Despite the dominance of CBL in the afternoon, the SBL and NBL may form under

166

certain meteorological conditions (Medeiros et al. 2005; Poulos et al. 2002; Stull 1988).

167

The complicated connections between the PBL structure and meteorology make it

168

imperative that the PBL types be determined upfront.

169

Here the determination of PBL types is largely based on the methods developed by

170

Liu and Liang (2010), where the soundings they used are at a vertical resolution of ~5

171

hPa, which is in sharp contrast to the vertical resolution of L-band radiosonde (~1 hPa)

172

to be used in the present study. Therefore, subsample operation in the vertical has been

173

performed by resampling 5 consecutive original 1-hPa measurements to one 5-hPa

174

measurement. As such, a series of coarser-resolution soundings (5-hPa) are derived,

175

and Figure S1 is a case in point. Then, the PBL types were determined by calculating

176

the near-surface potential temperature difference (PTD) between the fifth lowest 8

177

measurement and second lowest measurement. The threshold value of PTD is set to be

178

0.1 K, and the thermodynamic stability of PBL has to be further determined using the

179

bulk Richardson number based on the original lowest 100 m radiosonde measurements.

180

Specifically, if the PTD lies between -0.1 K and 0.1 K, the PBL is identified as NBL; if

181

the PTD is higher than 0.1 K and the bulk Richardson number is positive, the PBL types

182

is identified as SBL; other PBL cases can be considered as CBL (Vogelezang and

183

Holtslag 1996; Eresmaa et al. 2006; Tang et al. 2015).

184

2.3 PBL height estimation and uncertainty analysis

185

The estimation method of PBL height differs by varying PBL types. Similar to Liu

186

and Liang (2010), the PBL height for both CBL and NBL is calculated as the height at

187

which an air parcel rising adiabatically from the surface becomes neutrally buoyant

188

(Stull 1988). Starting from the surface, the height where the gradient of potential

189

temperature firstly becomes greater than a certain gradient threshold (GT) of potential

190

temperature is considered as the estimated PBL height.

191

For the SBL, the determination of the PBL height is much more uncertain.

192

Turbulence in the SBL can result from either buoyancy forcing or wind shear (Bonner

193

1968; Garreaud and Muñoz 2005). If both the stability-derived and wind-shear-derived

194

PBL heights are derived, the lowest of which is estimated to be the PBL heights for

195

SBL (Liu and Liang 2010; Sivaraman 2013). Figure S2 illustrates the typical profiles

196

of CBL, NBL, and SBL, and their corresponding PBL heights are 1650, 721, and 249

197

m above ground level (AGL), respectively. 9

198

To quantify the uncertainties in the estimation of PBL heights, sensitivity analysis

199

by selecting various values of GT has been performed (Argentini et al. 2005). Figures

200

1a-b compared the PBL heights calculated using 3.5 K/km, 4 K/km 4.5 K/km as GT.

201

As expected, the PBL heights of GT = 4.5 K/km (3.5 K/km) are generally higher (lower)

202

than those of GT= 4 K/km. The heights of NBL and CBL based on different GT are

203

significantly correlated with each other (R = 0.98). In contrast, the heights of SBL based

204

on various GT have relatively weaker correlations with each other.

205

We further checked the changes in occurrence frequency of SBL, NBL and CBL

206

caused by the different critical GT values applied. The changes in their corresponding

207

PBL heights were examined as well, which is summarized in Table S1. Overall, the

208

deviation of frequency for SBL, NBL and CBL is less than 13%, and the changes in

209

PBL heights is not more than 177 m, which could make sense for SBL cases. However,

210

the uncertainties can be negligible for NBL and CBL, since most of their PBL heights

211

are much higher.

212

3. Results

213

This section presents the basic climatology of three PBL types (i.e., SBL, NBL, and

214

CBL). Comparisons of radiosonde-based PBL height climatology under different CLD

215

conditions are made to elucidate the impacts of cloud on the development of PBL in the

216

afternoon during the summer. Then we present comprehensive analysis results with

217

regard to the correlations between PBL heights and four typical meteorological

218

variables, including Tsfc, Psfc, RHsfc, and WS10. 10

219

3.1 Climatology of PBL height for various PBL types

220

Figure 2 shows the frequency distribution and cumulative frequency distribution of

221

PBL heights for SBL, NBL, and CBL. At 1400 BJT, 96 % of SBL are lower than 500

222

m, while 83 % (76 %) of CBL (NBL) are higher than 1000 m. On the whole, all PBL

223

heights estimated from soundings are lower than 5 km at 1400 BJT across China.

224

In terms of the numbers of valid soundings at each site at 1400 BJT during summer,

225

the numbers of soundings at 69 sites (about 75%) range from 30 to 60 (Figure 3). In

226

particular, the valid soundings at Beijing (39.80°N, 116.47°E) and Shanghai (31.40°N,

227

121.48°E) are larger than 150. Figures 3b-d presents the spatial distributions of

228

occurrence frequency with respect to CBL, NBL and SBL at 1400 BJT across China,

229

which on average are 70 %, 26 %, and 4 %, respectively. The occurrence frequency of

230

NBL is generally less than 40 % at most sites of CRN. By comparison, the CBL

231

dominates over most sites of CRN.

232

Irrespective of PBL regimes, Figure 4a displays the spatial distribution of mean PBL

233

height at 1400 BJT in China. The averaged PBL height of all sites is 1694 m AGL,

234

indicative of high tendency of well-developed PBL in the afternoon. A prominent north-

235

south PBL height gradient is found, with higher PBL height in northwestern China,

236

which is consistent with spatial pattern revealed by previous PBL studies derived from

237

the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) onboard CALIPSO

238

(Huang et al. 2009; Liu et al. 2015; Zhang et al. 2016). The climatological PBL height

239

in northwest China is generally greater than 1800 m AGL, whereas the values are 11

240

usually less than 1400 m AGL over southeast and coastal areas. The high soil moisture

241

may be one of the causes accounting for the relatively shallower daytime PBL over the

242

east coast (McGrath-Spangler and Denning 2012; Wang and Wang 2014; 2016). In

243

contrast, the low soil moisture in the north and west regions tends to favor the PBL

244

development by partitioning more solar radiation to sensible heat flux (Wang and Wang

245

2014).

246

Figures 4b-d illustrate the spatial distributions of PBL height of SBL, NBL, and CBL.

247

Specifically, Figure 4b shows that the PBL height for SBL is merely 152 ± 100 m AGL

248

(mean value ± standard deviation). However, the values increase sharply to 1656 ± 562

249

m AGL for NBL, and 1803 ± 550 m AGL for CBL (Figures 4c-d). In other words, the

250

mean PBL heights tend to be on average reduced by ~150 m in the presence of NBL as

251

compared with those for CBL. Although the SBL develop more frequently after sunset,

252

which can also be established during the daytime under certain synoptic conditions.

253

To reveal the spatial discrepancy in PBL, we further selected six typical regions of

254

interest (ROIs, defined and shown in Figure S3), including (1) the North China Plain

255

(NCP), (2) the Yangtze River Delta (YRD), (3) the Pearl River Delta (PRD), (4) the

256

Taklamakan Desert (TKD), (5) the Tibetan Plateau (TBP), and (6) the Sichuan Basin

257

(SCB). The occurrence frequency and averaged PBL height of CBL, NBL and SBL at

258

these six ROIs are illustrated in Figure 5 and Table 1. Except for PRD, in which the

259

1400 BJT soundings were only launched in May and June, the 1400 BJT soundings in

260

other ROIs are all launched throughout the whole summer season (June-July-August),.

12

261

Overall, the CBL occurs most frequently in summer (60%), about 2 times higher than

262

the frequency of NBL (less than 30%), let alone the frequency of SBL (less than 10%).

263

Among these six ROIs, the TBP has the highest occurrence frequency of CBL (Figure

264

5), characterized by an averaged PBL height of 2222 m AGL (Table 1). Although both

265

PRD and TKD have similar occurrence frequencies for all the three PBL types, the

266

mean PBL heights are quite different, which are 1067 and 2248 m AGL for PRD and

267

TKD, respectively (Table 1). This large difference in PBL height over PRD and TKD

268

could be relevant to the large discrepancy of RHsfc and SHFsfc. The mean RHsfc over

269

PRD and TKD is 50 % and 22 %, respectively, whereas the mean SHFsfc 113 W/m2 and

270

290 W/m2 at these two ROIs. In the afternoon, less moisture in the PBL of arid TKD is

271

associated often with lower RH and higher SHF, which generally facilitates the

272

development of PBL (Stull 1988; Liu et al. 2004; Dirmeyer et al. 2014). This could be

273

likely the main reason why the high PBL height occurred in TKD rather than in PRD.

274

3.2 The potential impact of cloud on the development of various PBL regimes

275

It is well known that clouds alter the solar radiation reaching the land surface, which

276

further modulates the PBL in the afternoon. Figure 6 shows the mean PBL height and

277

frequency for SBL, NBL, and CBL under different CLD conditions. It is intriguing to

278

note that the PBL height does not change much when CLD ranges from 0 % to 80 %,

279

no matter what the PBL regime is. Only when the CLD becomes higher than 80%,

280

which is referred to as the overcast condition, the PBL height for SBL, NBL and CBL

281

decreases by a certain magnitude. As compared with that under clear-sky condition 13

282

(CLD < 20%), the averaged PBL height at 1400 BJT under overcast condition decreases

283

by ~400 m and ~300 m for NBL and CBL, respectively. In contrast, the depth of SBL

284

keeps almost constant. Relative to PBL under clear-sky condition, the changes in PBL

285

height under partly cloudy condition are less than 170 m for both NBL and CBL regimes

286

(Table S2).

287

Figure 6b shows the occurrence frequency of SBL, NBL, and CBL under different

288

CLD conditions in summer. In general, the occurrence frequency of CBL is higher than

289

40 % under overcast conditions, almost twice the frequency under clear-sky conditions.

290

Similarly, the occurrence frequencies of NBL are ~20 % and ~50 % for the clear-sky

291

and overcast conditions, respectively. For SBL, the occurrence frequency is merely ~10 %

292

under clear-sky conditions, as compared to ~70 % under overcast conditions. Such huge

293

differences existing in the occurrence frequency of CBL, NBL, and SBL suggested that

294

the CLD could play a significant role in modulating the development of PBL. The

295

presence of clouds tends to reduce the solar radiation reaching surface in the afternoon,

296

and thus weaken the vertical turbulent mixing, which in turn suppress the PBL

297

development (Li et al. 2017). Therefore, a reduced PBL height typically ensues (Wetzel

298

et al. 1996; Freedman et al. 2001; Zhou and Geerts 2013).

299

To figure out the spatial distribution of potential cloud influences on CBL, 29

300

radiosonde sites evenly scattered across the China were selected, in which the number

301

of valid CBL sounding is more than 30 for both overcast and clear-sky conditions. As

302

shown in Figure 7, one noticeable feature is that the PBL height under clear-sky

303

condition is systematically lower than that under overcast condition throughout all these 14

304

29 sites, even though the PBL height under both conditions ranges from 800 m to 2600

305

m. On average, the difference of mean PBL height is 336 m, as indicated in Figure 7c.

306

3.3 Relationships between PBL height and meteorological variables

307

In this section, we will analyze the relationships between PBL height and several

308

meteorological parameters, including Tsfc, RHsfc, Psfc, and WS10. To better understand

309

the development of SBL in the afternoon, the meteorological factors for the days with

310

SBL were compared with those with CBL, including CLD, Tsfc, RHsfc, WS10 and SHFsfc.

311

Comparing with the CBL regime, the SBL tends to be formed under the meteorological

312

conditions characterized by relatively high CLD and RHsfc, and low Tsfc, WS10 and

313

SHFsfc (Figure 8).

314

Figure 9 shows the correlation coefficient (R) of PBL height and various

315

meteorological factors in the six ROIs for different PBL regimes, and the gray shaded

316

areas indicate the R values are statistically significant at 95% confidence level.

317

Typically, the PBL height is positively correlated with Tsfc in six ROIs for both NBL

318

and CBL, including the NCP (R = 0.37, 0.25 for NBL and CBL, respectively), YRD (R

319

= 0.55, 0.62), PRD (R = 0.73, 0.66), SCB (R = 0.68 under CBL), and TBP (R = 0.19,

320

0.24), and the lowest correlation (R = 0.02, 0.13) is found in TKD. For SBL, only the

321

correlation of YRD region show here (R = -0.30). More details on the correlations can

322

be found in Table S3.

323

In contrast, negative relationships are observed between the PBL height and RHsfc in

324

summer for NBL and CBL (Figure 9), and the R values are mostly lower than -0.6 15

325

except for those in TKD (R = -0.39 for NBL, -0.51 for CBL) regions. By comparison,

326

the positive correlations (R = 0.36) in TKD between PBL height and RHsfc are found in

327

summer for SBL. For a specific location, the high RHsfc may be associated with high

328

soil moisture and high latent heat flux, which will inhibit the occurrence of convection

329

and turbulence within SBL, and thus lead to a relatively deeper SBL. For CBL and NBL,

330

higher Tsfc and lower RHsfc have been found to be linked to more sensible heat fluxes,

331

favoring the development of PBL during the daytime (Zhang et al. 2013).

332

Besides, the relationships are ubiquitously negative between the PBL height and Psfc

333

in summer for CBL, NBL and SBL. Similar correlation analyses using WS10 show that

334

these parameters are not significantly correlated with PBL height at 1400 BJT in

335

summer.

336

4. Discussions

337

Regarding the large discrepancy in PBL height in PRD and TKD (Figure 5), it is

338

known that PRD is characterized by richer vegetation cover compared with TKD, the

339

relatively strong surface evaporation in PRD can substantially increase the near-surface

340

humidity, and lead to less surface sensible heat flux, which suppress the PBL height

341

over PRD (Zhang et al. 2013; Dirmeyer et al. 2014).

342

In addition to the local-scale atmospheric processes, the large-scale synoptic

343

conditions have been well recognized to be able to govern the development of the PBL

344

(Whiteman and Doran 1993; Garratt 1994; Hoover et al. 2015; Miao et al. 2017). For

345

example, the subsidence associated with the high-pressure systems (e.g., anti-cyclone) 16

346

may suppress the development of PBL, while the large-scale upward motions induced

347

by the low-pressure systems (e.g., fronts, cyclone) would favor the growth of PBL (Stull

348

1988). Specifically, the cold front passage generally comes with CBL, since the strong

349

winds within PBL would favor the development the PBL dynamically. In contrast,

350

sometimes the presence of warm front could facilitate the formation of SBL, since warm

351

air mass behind the front would climb over the local colder air, leading to an increase

352

in atmospheric stability and a suppressed growth in PBL to some extent (Keyser and

353

Anthes 1982; Stull 1988; Sun et al. 2002). Conversely, the stable PBL is also favorable

354

for the low-level stratus cloud formation and maintenance (Lin et al. 2016).

355

Furthermore, clouds (especially under overcast condition) are generally associated

356

with lower PBL height (Figure 6). As shown in Figure 10, compared with the negligible

357

association between CLD and WS10, high CLD was found to be associated with low

358

Tsfc and higher RHsfc (Betts et al. 2017; Garcia-Carreras et al. 2017), which also

359

accompanied with the lower PBL height in the afternoon (Zhang et al. 2013). Although

360

high wind speed is recognized to be positively associated with annual averaged PBL

361

height in the morning or afternoon (Guo et al. 2016b), this positive relation cannot be

362

apparently seen from the afternoon PBL during summer. There may exist complicated

363

coupling effects between the near-surface meteorological factors, CLD, and PBL in the

364

afternoon. For instance, high CLD is often associated with high RHsfc and low PBL

365

height, due to the modulation of surface solar radiation. Meanwhile, the low PBL height

366

may limit the vertical exchange of water vapor between the moister surface layer and

367

upper dryer free troposphere, leading to the relatively high RHsfc (Hohenegger et al. 17

368

2009;Zhou and Geerts 2013).

369

In addition, the soil moisture may be another important factor dictating the

370

development of PBL, since it affects the land surface energy budget (e.g., surface

371

sensible heat flux and latent heat flux). For a specific location, the lower soil moisture

372

generally comes with high surface sensible heat flux, which facilitates the PBL

373

development in the afternoon (McCumber and Pielke 1981; Sanchez-Mejia and Papuga

374

2014; Rihani et al. 2015). As illustrated in Figure S4, the northwest of China is mainly

375

covered by bare land, characterized by low soil moisture, high SHFsfc, high Bowen ratio,

376

and low latent heat flux, all of which favors the development of PBL. On the contrary,

377

the high soil moisture, high latent heat flux, low SHFsfc and low Bowen ratio in the

378

southern China are unfavorable to the development of PBL. Besides, the high latent

379

heat flux tends to favor cloud formation, and the increased clouds will in turn results in

380

reduced surface temperature, increased soil moisture and decreased PBL height (Betts

381

et al. 1996; Eltahir 1998; Pal and Eltahir 2001; Zhou and Geerts 2013). Additionally,

382

the different land covers in the northwest and southeast of China may also play a role

383

in modulating the land surface processes and PBL development. The complex

384

relationships/connections between the land surface properties and PBL in China need

385

more studies in the future.

386

5. Conclusions

387

Based on the 1400 BJT soundings in summer collected from the CMA radiosonde

388

network across China from 2012 to 2016, this study investigated the occurrence

18

389

frequency, spatial distribution of different PBL types (i.e., SBL, NBL, and CBL) in

390

China. Besides, the relationships between the PBL heights and other meteorological

391

variables were examined, including CLD, Tsfc, Psfc, RHsfc and WS10.

392

The occurrence frequency of CBL is about 70 %, followed by 26 % for NBL, and 4 %

393

for SBL. At 1400 BJT, the spatial distributions of PBL heights for NBL and CBL are

394

quite similar, demonstrating a prominent north-south gradient of PBL height. The depth

395

of SBL (152 ± 100m) is generally lower than those of NBL (1656 ± 562 m) and CBL

396

(1803± 550 m). At 1400 BJT, clouds tend to reduce the occurrence frequency of CBL

397

and NBL. In terms of cloud impact on the frequency of SBL, it is found to be merely

398

~10 % under clear-sky conditions, as compared to ~70 % under overcast conditions,

399

indicating that clouds are one of the key factors dictating the frequency of SBL. With

400

the exception of the discernible changes in PBL height under overcast condition relative

401

to the PBL height under clear-sky condition, the PBL height under partly cloudy

402

condition does not change much (less than 170 m) under CBL and NBL regimes. This

403

indicates that high cloud cover could be an important factor modulating PBL

404

development. Besides, the PBL heights of CBL and NBL were found to be positively

405

correlated with the near-surface meteorological factors such as Tsfc, but negatively

406

associated with RHsfc. For a specific location, the low RHsfc usually accompanies with

407

low CLD and low soil moisture, which could favor the development of PBL in the

408

afternoon with high surface sensible heat flux.

409

Although the CBL dominates over China in the summer afternoon, the NBL and SBL

410

could also be established and cannot be overlooked. This study merely investigated the 19

411

afternoon PBL in summer, more efforts should be made to expand the climatology of

412

different PBL types to other seasons in China when more sounding data are available

413

in the future.

414

Acknowledgements

415

This work was supported by the National Natural Science Foundation of China under

416

Grants 41771399, 41705002, 91544217 and 41471301, Ministry of Science and

417

Technology under Grant 2014BAC16B01, and Chinese Academy of Meteorological

418

Sciences under Grants 2017Z005 and 2017Y002. The authors would like to

419

acknowledge Chinese Meteorological Administration for providing the radiosonde

420

dataset. Last, but not least, special thanks go to four anonymous reviewers for their

421

valuable comments and suggestions that helped improve the quality of our manuscript.

20

422

Reference

423

Amenu, G. G., P. Kumar, and X. Z. Liang, 2005: Interannual variability of deep-layer

424

hydrologic memory and mechanisms of its influence on surface energy fluxes. J.

425

Climate, 18(23), 5024-5045, doi:10.1175/JCLI3590.1.

426

Argentini, S., A. Viola, A. Sempreviva, and I. Petenko, 2005: Summer boundary-layer

427

height at the plateau site of Dome’C, Antarctica. Bound. Lay. Meteorol., 115(3), 409-

428

422, doi:10.1007/s10546-004-5643-6.

429

Asimakopoulos, D., C. Helmis, and J. Michopoulos, 2004: Evaluation of SODAR

430

methods for the determination of the atmospheric boundary layer mixing height.

431

Meteorol. Atmos. Phys., 85(1-3), 85-92, doi:10.1007/s00703-003-0036-9.

432

Betts, A. K., A. B. Tawfik, and R. L. Desjardins, 2017: Revisiting hydrometeorology

433

using cloud and climate observations. J. Hydrometeorol., 18(4), 939-955,

434

doi:10.1175/JHM-D-16-0203.1.

435

Betts, A. K., J. H. Ball, A. C. M. Beljaars, M. J. Miller, and P. A. Viterbo, 1996: The

436

land surface-atmosphere interaction: A review based on observational and global

437

modeling perspectives. J. Geophys. Res. Atmos., 101 (D3), 7209–7225,

438

doi:10.1029/95JD02135.

439 440

Beyrich, F., 1997: Mixing height estimation from sodar data - a critical discussion, Atmos. Environ., 31(23), 3941-3953, doi:10.1016/S1352-2310(97)00231-8.

441

Bian, J., H. Chen, H. Vömel, Y. Duan, Y. Xuan, and D. LV, 2011: Intercomparison of

442

humidity and temperature sensors: GTS1, Vaisala RS80, and CFH. Adv. Atmos. Sci.,

443

28, 1, 139-146, doi:10.1007/s00376-010-9170-8.

444

Blay-Carreras, E., D. Pino, J. Vilà-Guerau de Arellano, A. van de Boer, O. De Coster,

445

C. Darbieu, O. Hartogensis, F. Lohou, M. Lothon, and H. Pietersen, 2014: Role of

446

the residual layer and large-scale subsidence on the development and evolution of

447

the convective boundary layer. Atmos. Chem. Phys., 14(9), 4515-4530,

448

doi:10.5194/acp-14-4515-2014.

449 450 451

Bonner, W. D., 1968: Climatology of the low level jet. Mon. Weather Rev, 96(12), 833850, doi:10.1175/1520-0493(1968)0962.0.CO;2. Chan, K. M., and R. Wood, 2013: The seasonal cycle of planetary boundary layer depth 21

452

determined using COSMIC radio occultation data. J. Geophys. Res. Atmos., 118(22),

453

doi:10.1002/2013JD020147.

454

Chen, F., and J. Dudhia, 2001: Coupling an Advanced Land Surface–Hydrology Model

455

with the Penn State–NCAR MM5 Modeling System. Part I: Model Implementation

456

and

457

0493(2001)1292.0.CO;2.

Sensitivity.

Mon.

Weather

Rev.

129,

569-585,

doi:10.1175/1520-

458

Chen, S. S., and R. A. Houze, 1997: Diurnal variation and life ‐ cycle of deep

459

convective systems over the tropical Pacific warm pool. Q. J. Roy. Meteorol. Soc.,

460

123(538), 357-388, doi: 10.1002/qj.49712353806.

461

Coulter, R. L., 1979: A comparison of three methods for measuring mixing-layer height.

462

J.

Appl.

Meteorol.,

18(11),

463

0450(1979)0182.0.CO;2.

1495-1499,

doi:10.1175/1520-

464

Dandou, A., M. Tombrou, K. Schäfer, S. Emeis, A. P. Protonotariou, E. Bossioli, N.

465

Soulakellis, and P. Suppan, 2009, A comparison between modelled and measured

466

mixing-layer height over Munich. Bound. Lay. Meteorol., 131(3), 425-440,

467

doi:10.1007/s10546-009-9373-7.

468

Davy, R., and I. Esau, 2016: Differences in the efficacy of climate forcings explained

469

by variations in atmospheric boundary layer depth. Nat. Commun., 7,

470

doi:10.1038/ncomms11690.

471

Dirmeyer, P. A., Z .Wang, M. J. Mbuh, and H. E. Norton, 2014: Intensified land surface

472

control on boundary layer growth in a changing climate. Geophys. Res. Lett., 41(4),

473

1290-1294, doi:10.1002/2013GL058826.

474

Eresmaa, N., A. Karppinen, S. Joffre, J. Räsänen, and H. Talvitie, 2006: Mixing height

475

determination by ceilometer. Atmos. Chem. Phys., 6(6), 1485-1493, doi:10.5194/acp-

476

6-1485-2006.

477

Eitan, H., K. Ilan, A. Orit, L. Zev and A. Eyal, 2017: Enhanced humidity pockets

478

originating in the mid boundary layer as a mechanism of cloud formation below the

479

lifting condensation level. Environ. Res. Lett., 12(2): 024020. doi:10.1088/1748-

480

9326/aa5ba4.

481 482

Eltahir, E. A., and C. Gong, 1996: Dynamics of wet and dry years in West Africa. J. Climate,

9(5), 22

1030-1042,

483 484 485 486 487

doi:10.1175/15200442(1996)0092.0.CO;2. Eltahir, E. A. B., 1998: A soil moisture–rainfall feedback mechanism. Part I: Theory and observations. Water Resour. Res., 34, 765–776, doi: 10.1029/97WR03499. Esau, I., and S. Zilitinkevich, 2010: On the role of the planetary boundary layer depth in the climate system. Adv. Sci. Res., 4, 63, doi:10.5194/asr-4-63-2010.

488

Freedman, J. M., D. R. Fitzjarrald, K. E. Moore, and R. K. Sakai, 2001: Boundary layer

489

clouds and vegetation-atmosphere feedbacks. J. Climate, 14, 180–197, doi:

490

10.1175/1520-0442(2001)0132.0.CO;2.

491

Fletcher, J., S. Mason, and C. Jakob, 2016: The climatology, meteorology, and boundary

492

layer structure of marine cold air outbreaks in both hemispheres. J. Climate, 29(6),

493

1999-2014, doi: 10.1175/JCLI-D-15-0268.1.

494

Garcia-Carreras, L., J. H. Marsham, and D. V. Spracklen, 2017: Observations of

495

increased cloud cover over irrigated agriculture in an arid environment. J.

496

Hydrometeorol., 18(4), 2161-2172, doi:10.1175/JHM-D-16-0208.1.

497 498 499 500 501

Garratt, J. R., 1992: The Atmospheric Boundary Layer. Cambridge University Press, 316 pp. Garratt, J. R., 1994: Review: the atmospheric boundary layer, Earth Sci. Rev., 37(1-2), 89-134, doi: 10.1016/0012-8252(94)90026-4. Garreaud, R., and R. C. Muñoz, 2005: The low-level jet off the west coast of subtropical

502

South America: Structure and variability. Mon. Weather Rev., 133(8), 2246-2261, doi:

503

10.1175/MWR2972.1.

504

Guo, J., X. Zhang, H. Che, S. Gong, X. An, C. Cao, J. Guang, H. Zhang, Y. Wang, and

505

X. Zhang, 2009: Correlation between PM concentrations and aerosol optical depth

506

in

507

doi:10.1016/j.atmosenv.2009.08.026.

eastern

China.

Atmos.

Environ.

,

43(37),

5876-5886,

508

Guo, J., X. Zhang, Y. Wu, Y. Zhaxi, H. Che, B. La, W. Wang, and X. Li, 2011: Spatio-

509

temporal variation trends of satellite-based aerosol optical depth in China during

510

1980–2008.

511

doi:10.1016/j.atmosenv.2011.03.068.

512

Atmos.

Environ.,

45(37),

6802-6811,

Guo, J., J. He, H. Liu, Y. Miao, H. Liu, and P. Zhai, 2016a: Impact of various emission 23

513

control schemes on air quality using WRF-Chem during APEC China 2014. Atmos.

514

Environ., 140, 311-319, doi:10.1016/j.atmosenv.2016.05.046.

515

Guo, J., Y. Miao, Y. Zhang, H. Liu, Z. Li, W. Zhang, J. He, M. Lou, Y. Yan, L. Bian,

516

and P. Zhai, 2016b: The climatology of planetary boundary layer height in China

517

derived from radiosonde and reanalysis data. Atmos. Chem. Phys., 16(20), 13309-

518

13319, doi:10.5194/acp-16-13309-2016.

519

Guo, J., H. Liu, F. Wang, J. Huang, F. Xia, M. Lou, Y. Wu, H. Jiang, T. Xie, Y. Zhaxi,

520

and L. Yung, 2016c: Three-dimensional structure of aerosol in China: A perspective

521

from

522

doi:10.1016/j.atmosres.2016.05.010.

multi-satellite

observations.

Atmos.

Res.,

178–179,

580–589,

523

Hennemuth, B., and A. Lammert, 2006: Determination of the atmospheric boundary

524

layer height from radiosonde and lidar backscatter. Bound. Lay. Meteorol., 120(1),

525

181-200, doi:10.1007/s10546-005-9035-3.

526

Holzworth, G. C., 1964: Estimates of mean maximum mixing depths in the contiguous

527

United

States.

Mon.

Weather

Rev.,

92(5),

528

0493(1964)0922.3.CO;2.

235-242,

doi:10.1175/1520-

529

Hoover, J. D., D. R. Stauffer, S. J. Richardson, L. Mahrt, B. J. Gaudet, and A. Suarez

530

2015: Submeso motions within the stable boundary layer and their relationships to

531

local indicators and synoptic regime in moderately complex terrain. J. Appl.

532

Meteorol. Clim., 54(2), 352-369, doi:10.1175/JAMC-D-14-0128.1.

533

Hohenegger, C., P. Brockhaus, C. S. Bretherton, and C. Schar, 2009: The soil moisture–

534

precipitation feedback in simulations with explicit and parameterized convection. J.

535

Climate, 22, 5003–5020, doi:10.1175/2009JCLI2604.1.

536

Hu, X., G. Nielsen, John W, and F. Zhang, 2010: Evaluation of three planetary boundary

537

layer schemes in the WRF model. J. Appl. Meteorol. Clim., 49(9), 1831-1844,

538

doi:10.1175/2010JAMC2432.1.

539

Huang, J., Q. Fu, J. Su, Q. Tang, P. Minnis, Y. Hu, Y. Yi, and Q. Zhao, 2009: Taklimakan

540

dust aerosol radiative heating derived from CALIPSO observations using the Fu-

541

Liou radiation model with CERES constraints. Atmos. Chem. Phys., 9, 4011-4021,

542

doi:10.5194/acp-9-4011-2009.

543

Keyser, D., and R. A. Anthes, 1982: The influence of planetary boundary layer physics 24

544

on frontal structure in the Hoskins-Bretherton horizontal shear model. J. Atmos. Sci.,

545

39(8), 1783-1802.

546 547

Lammert, A., and J. Bösenberg, 2006: Determination of the convective boundary-layer height with laser remote sensing. Bound. Lay. Meteorol., 119(1), 159-170.

548

Li Z., J. Guo, A. Ding, H. Liao, J. Liu, Y. Sun, T. Wang, H. Xue, H. Zhang, B. Zhu,

549

2017: Aerosol and boundary-layer interactions and impact on air quality. Natl. Sci.

550

Rev., doi: 10.1093/nsr/nwx117.

551

Lin, Y., Y. Wang, R. Zhang, Y. Liu, 2016: Response of boundary layer clouds to

552

continental pollution during the RACORO campaign. J. Atmo. Sci. 73(9), 3681-3700.

553

Liu, J., J. Huang, B. Chen, T. Zhou, H. Yan, H. Jin, Z. Huang, and B. Zhang, 2015:

554

Comparisons of PBL heights derived from CALIPSO and ECMWF reanalysis data

555

over

556

doi:10.1016/j.jqsrt.2014.10.011.

557

China.

J.

Quant.

Spectrosc.

Radiat.

Transfer,

153,

102-112,

Liu, S., and X.-Z. Liang, 2010: Observed Diurnal Cycle Climatology of Planetary

558

Boundary

Layer

Height.

559

doi:10.1175/2010JCLI3552.1.

J.

Climate,

23(21),

5790-5809,

560

Liu, S., X. Yue, F. Hu, and H. Liu, 2004: Using a Modified Soil-Plant-Atmosphere

561

Scheme (MSPAS) to simulate the interaction between land surface processes and

562

atmospheric boundary layer in semi-arid regions. Adv. Atm. Sci., 21(2), 245-259,

563

doi:10.1007/BF02915711.

564

Mao, M., W. Jiang, J. Gu, C. Xie, and J. Zhou, 2009: Study on the mixed layer,

565

entrainment zone, and cloud feedback based on lidar exploration of Nanjing city.

566

Geophys. Res. Lett., 36(4), doi:10.1029/2008GL036768.

567

Ma Y., W. Yao, and B. Huang, 2011: Comparison of Temperature and Geopotential

568

Height Records Between 59 Type and L-band Radiosonde Systems. J. App. Meteor.

569

Sci., 22(3), 336-345, doi:10.3969/j.issn.1001-7313.2010.02.011.

570

Mccumber, M. C., and R. A. Pielke, 1981: Simulation of the effects of surface fluxes

571

of heat and moisture in a mesoscale numerical model: 1. Soil layer. J. Geophys. Res.

572

Oceans, 86(C10), 9929-9938, doi:10.1029/JC086iC10p09929.

573

McGrath-Spangler, E. L., and A. S. Denning, 2012: Estimates of North American 25

574

summertime planetary boundary layer depths derived from space-borne lidar. J.

575

Geophys. Res. Atmos., 117(D15), doi:10.1029/2012JD017615.

576 577

Medeiros, B., A. Hall, and B. Stevens, 2005: What controls the mean depth of the PBL?. J. Climate, 18(16), 3157-3172, doi:10.1175/JCLI3417.1.

578

Miao, Y., X. Hu, S. Liu, T. Qian, M. Xue, Y. Zheng, and S. Wang, 2015a: Seasonal

579

variation of local atmospheric circulations and boundary layer structure in the

580

Beijing-Tianjin-Hebei region and implications for air quality. J. Adv. Model Earth

581

Sy., 7(4), 1602-1626, doi:10.1002/2015MS000522.

582

Miao, Y., S. Liu, Y. Zheng, S. Wang, Z. Liu, and B. Zhang, 2015b: Numerical study of

583

the effects of planetary boundary layer structure on the pollutant dispersion within

584

built-up areas. J. Environ. Sci., 32, 168-179, doi:10.1016/j.jes.2014.10.025.

585

Miao, Y., J. Guo, S. Liu, H. Liu, Z. Li, W. Zhang, and P. Zhai, 2017: Classification of

586

summertime synoptic patterns in Beijing and their association with boundary layer

587

structure affecting aerosol pollution. Atmos. Chem. Phys., 17, 3097-3110,

588

doi:10.5194/acp-17-3097-2017.

589

Oke, T. R., 2002: Boundary layer climates. Routledge, 435 pp.

590

Pal, J. S., and E. A. B. Eltahir, 2001: Pathways relating soil moisture conditions to future

591

summer rainfall within a model of the land–atmosphere system. J. Climate, 14,

592

1227–1242, doi:10.1175/1520-0442(2001)0142.0.CO;2.

593 594

Paluch, I. R., and D. H. Lenschow, 1991: Stratiform cloud formation in the marine boundary layer. J. Atmos. Sci., 48(19): 2141-2158.

595

Poulos, G. S., W. Blumen, D. C. Fritts, J. K. Lundquist, J. Sun, S. Burns, C. Nappo, R.

596

Banta, T. Newsom, J. Cuxart, E. Terradellas, B. Balsley, and M. Jensen, 2002:

597

CASES-99: A comprehensive investigation of the stable nocturnal boundary layer.

598

B.

599

0477(2002)0832.3.CO;2.

Am.

Meteorol.

Soc.,

83(4),

555-581,

doi:10.1175/1520-

600

Rihani, J. F., F. K. Chow, and R. M. Maxwell, 2015: Isolating effects of terrain and soil

601

moisture heterogeneity on the atmospheric boundary layer: Idealized simulations to

602

diagnose land-atmosphere feedbacks. J. Adv. Model. Earth Sy., 7(2), 915-937,

603

doi:10.1002/2014MS000371.

26

604

Rodell, M., P. R. Houser, U. Jambor, J. Gottschalck, K. Mitchell, C-J. Meng, K.

605

Arsenault, B. Cosgrove, J. Radakovich, M. Bosilovich, J. K. Entin, J. P. Walker, D.

606

Lohmann, and D. Toll, 2004: The Global Land Data Assimilation System. Bull. Am.

607

Meteorol. Soc., 85(3), 381–394, doi:10.1175/BAMS-85-3-381.

608

Sanchez-Mejia, Z. M., and S. A. Papuga, 2014: Observations of a two-layer soil

609

moisture influence on surface energy dynamics and planetary boundary layer

610

characteristics in a semiarid shrubland. Water Resour. Res., 50(1), 306-317,

611

doi:10.1002/2013WR014135.

612

Sherwood, S. C., S. Bony, and J. L. Dufresne, 2014: Spread in model climate sensitivity

613

traced

to

atmospheric

614

doi:10.1038/nature12829.

convective

mixing.

Nature,

505,

37-42,

615

Sun, J., S. P. Burns, D. H. Lenschow, R. Banta, R. Newsom, R. Coulter, S. Frasier,T.

616

Ince, C. Nappo, J. Cuxart, W. Blumen, X. Lee, and X, Z. Hu, 2002: Intermittent

617

turbulence associated with a density current passage in the stable boundary layer.

618

Bound. Lay. Meteorol., 105(2), 199-219, doi:10.1023/A:1019969131774.

619

Sandip, P., and H. Martial, 2015: Forcing mechanisms governing diurnal, seasonal, and

620

interannual variability in the boundary layer depths: Five years of continuous lidar

621

observations over a suburban site near Paris. J. Geophys. Res. Atmos., 120(23),

622

11,936-11,956, doi:10.1002/2015JD023268.

623

Sawyer, V., and Z. Li, 2013: Detection, variations and intercomparison of the planetary

624

boundary layer depth from radiosonde, lidar and infrared spectrometer. Atmos.

625

Environ., 79, 518-528, doi:10.1016/j.atmosenv.2013.07.019.

626

Seibert, P., 2000: Review and intercomparison of operational methods for the

627

determination of the mixing height. Atmos. Environ., 34(7), 1001-1027,

628

doi:10.1016/S1352-2310(99)00349-0.

629

Seidel, D. J., C. O. Ao, and K. Li, 2010: Estimating climatological planetary boundary

630

layer heights from radiosonde observations: Comparison of methods and uncertainty

631

analysis. J. Geophys. Res. Atmos., 115(D16), doi:10.1029/2009JD013680.

632

Seidel, D. J., Y. Zhang, A. Beljaars, J. C. Golaz, A. R. Jacobson, and B. Medeiros, 2012:

633

Climatology of the planetary boundary layer over the continental United States and

634

Europe. J. Geophys. Res. Atmos., 117(D17), doi:10.1029/2012JD018143. 27

635

Sivaraman, C., S. McFarlane, E. Chapman, M. Jensen, T. Toto, S. Liu, and M. Fischer,

636

2013: Planetary Boundary Layer (PBL) Height Value Added Product (VAP):

637

Radiosonde Retrievals. U.S. Department of Energy, Gaithersburg, MD, USA,

638

DOE/SC-ARM-TR-132.

639

Solomon, A., M. D. Shupe, and N. B. Miller, 2017: Cloud–Atmospheric Boundary

640

Layer–Surface Interactions on the Greenland Ice Sheet during the July 2012 Extreme

641

Melt Event. J. Climate, 30(9), 3237-3252, doi:10.1175/JCLI-D-16-0071.1.

642

Steyn, D. G., M. Baldi, and R. Hoff, 1999: The detection of mixed layer depth and

643

entrainment zone thickness from lidar backscatter profiles. J. Atmos. Ocean.

644

Technol.,

645

0426(1999)0162.0.CO;2.

646 647

16(7),

953-959,

doi:10.1175/1520-

Stull, R. B., 1988: An Introduction to Boundary Layer Meteorology, Atmospheric Sciences Library, Dordrecht: Kluwer, PP670.

648

Tao, S., X. Chen, and J. Gong, 2006: Error analyses for temperature of L band

649

radiosonde. Meteor. Monogr, 32, 46-51, doi:10.3969/j.issn.1000-0526.2006.10.007.

650

Tang, G., X. Zhu, B. Hu, J. Xin, L. Wang, C. Münkel, G. Mao, and Y. Wang, 2015:

651

Impact of emission controls on air quality in Beijing during APEC 2014: lidar

652

ceilometer

653

doi:10.5194/acp-15-12667-2015.

observations.

Atmos.

Chem.

Phys.,

15(21),

12667-12680,

654

Tang, G., J. Zhang, X. Zhu, T. Song, C. Münkel, B. Hu, K. Schäfer, Z. Liu, J. Zhang, L.

655

Wang, J. Xin, P. Suppan, and Y. Wang, 2016: Mixing layer height and its implications

656

for air pollution over Beijing. Atmos. Chem. Phys., 16(4), 2459-2475,

657

doi:10.5194/acp-16-2459-2016.

658

Tokinaga, H., Y. Tanimoto, M. Nonaka, B. Taguchi, T. Fukamachi, S. P. Xie, H.

659

Nakamura, T. Watanabe, and I. Yasuda, 2006:

Atmospheric sounding over the

660

winter Kuroshio Extension: Effect of surface stability on atmospheric boundary

661

layer structure. Geophys. Res. Lett., 33(4), doi:10.1029/2005GL025102.

662

Vogelezang, D., and A. Holtslag, 1996: Evaluation and model impacts of alternative

663

boundary-layer height formulations. Bound. Lay. Meteorol., 81(3-4), 245-269,

664

doi:10.1007/BF02430331.

665

Wang, X., and K. Wang, 2016: Homogenized variability of radiosonde-derived 28

666

atmospheric boundary layer height over the global land surface from 1973 to 2014.

667

J. Climate, 29(19), 6893-6908, doi:10.1175/JCLI-D-15-0766.1.

668

Wang, X., and K. Wang, 2014: Estimation of atmospheric mixing layer height from

669

radiosonde data. Atmos. Meas. Tech., 7(6), 1701-1709, doi:10.5194/amt-7-1701-

670

2014.

671

Wetzel, P. J., S. Argentini, and A. Boone, 1996: Role of land surface in controlling

672

daytime cloud amount: Two case studies in the GCIP-SW area. J. Geophys. Res.

673

Atmos., 101, 7359–7370, doi:10.1029/95JD02134.

674 675

Whiteman, C. D., and J. C. Doran (1993). The relationship between overlying synopticscale flows and winds within a valley. J. Appl. Meteorol., 32(11), 1669-1682.

676

Xing Y., Z. Zhang, Y. Cao, Y. Li, H. Wang, X. Li, and S. Ma, 2009: Experiment and

677

Analysis of GPS Radiosonde RS92 Performance. Meteor. Sci. Technol., 37(3), 336-

678

340, doi:10.3969/j.issn.1671-6345.2009.03.014.

679

Xu W., Y. Guo, B. Huang, and S. Gu, 2007: Analysis of GTS Radiosonde Humidity

680

Sensor Testing Data and Correction of Upper-Air Relative Humidity Radiosonde

681

Data.

682

6345.2007.03.025.

Meteorol.

Sci.

Technol.,

35(3),

423-428,

doi:10.3969/j.issn.1671-

683

Zhang, W., M. Augustin, Y. Zhang, Z. Li, H. Xu, D. Liu, Z. Wang, Y. Zhang, Y. Ma,

684

and F. Zhang, 2015: Spatial and temporal variability of aerosol vertical distribution

685

based on lidar observations: A haze case study over Jinhua basin. Adv. Meteorol.,

686

doi:10.1155/2015/349592.

687

Zhang, W., J. Guo, Y. Miao, H. Liu, Y. Zhang, Z. Li, and P. Zhai, 2016: Planetary

688

boundary layer height from CALIOP compared to radiosonde over China. Atmos.

689

Chem. Phys., 16(15), 9951-9963, doi:10.5194/acp-16-9951-2016.

690

Zhang, Y., D. J. Seidel, J. C. Golaz, C. Deser, and R. A. Tomas, 2011: Climatological

691

characteristics of Arctic and Antarctic surface-based inversions. J. Climate, 24(19),

692

5167-5186, doi:10.1175/2011JCLI4004.1.

693

Zhang, Y., D. J. Seidel, and S. Zhang, 2013: Trends in planetary boundary layer height

694

over Europe. J. Climate, 26(24), 10071-10076, doi:10.1175/JCLI-D-13-00108.1.

695

Zhang, Y., L.J. Zhang, J.P. Guo, J.M. Feng, Y. Wang, Q. Zhou, 29

L. X. Li, B. Li, J. He,

696

H.

Xu,

L.

Liu,

N.

An,

H.

Liu,

2017:

697

The climatology of cloud base height from long-

698

term radiosonde measurements in China, Adv. Atmos. Sci., doi: 10.1007/s00376-

699

017-7096-0.

700

Zhang, Y., S. Zhang, C. Huang, K. Huang, Y. Gong, and Q. Gan, 2014: Diurnal

701

variations of the planetary boundary layer height estimated from intensive

702

radiosonde observations over Yichang, China. Sci. China Technol. Sci., 57(11),

703

2172-2176, doi:10.1007/s11431-014-5639-5.

704

Zhou, X., and B. Geerts, 2013: The influence of soil moisture on the planetary boundary

705

layer and on cumulus convection over an isolated mountain. Part I: observations.

706

Mon. Weather Rev., 141(3), 1061-1078, doi: 10.1175/MWR-D-12-00150.1.

30

707

Figure captions:

708

Figure 1. Scatter plots showing (a) PBL heights computed using 4 K/km (in the x-axis)

709

and 4.5 K/km (in the y-axis) as critical potential temperature gradient threshold (GT),

710

and (b) the PBL heights computed using 4 K/km (in the x-axis) and 3.5 K/km (in the y-

711

axis) as critical GT. The results are based on 5399 soundings across China in summer

712

(May-June-July-August) from 2012 to 2016.

713

Figure 2. Frequency distribution (a) and cumulative frequency distribution (b) of PBL

714

heights for SBL (in black), NBL (in red), and CBL (in blue) at 1400 BTJ in summer for

715

the period of 2012-2016. The number of soundings (N) and mean value at each

716

observed time are also given.

717

Figure 3. Spatial distributions of (a) # of soundings from CMA radiosonde sites and

718

the frequency (in percent) for PBL states classified by atmospheric stability: (b) SBL,

719

(c) NBL and (d) CBL at 1400 BJT in summer for the period of 2012-2016, respectively.

720

Figure 4. Spatial distribution of mean PBL height for (a) all types, (b) SBL, (c) NBL,

721

and (d) CBL at 1400 BJT in summer for the period of 2012-2016.

722

Figure 5. The occurrence frequency of SBL, NBL and CBL in six ROIs, including (1)

723

the North China Plain (NCP), (2) the Yangtze River Delta (YRD), (3) the Pearl River

724

Delta (PRD), (4) the Taklamakan Desert (TKD), (5) the Tibetan Plateau (TBP), and (6)

725

the Sichuan Basin (SCB), at 1400 BJT during in summer (May-June-July-August) for

726

the period of 2012 to 2016, respectively. The bars show the frequency of different PBL

727

types (SBL in black, NBL in gray, and CBL in white). The pie charts show the frequency

728

percentage distribution for each month.

729

Figure 6. The bar charts showing the mean PBL height (a) and the occurrence

730

frequency (b) for three PBL types, including SBL (in gray), NBL (in red), and CBL (in

731

blue) under different cloud cover (CLD) conditions (with a bin size of 20%) in summer

732

for the period of 2012-2016. Note that the error bar in (a) denotes the standard deviation

733

of PBL height. 31

734

Figure 7. The spatial distributions of mean PBL heights at 1400 BJT in summer for the

735

period 2012-2016 for CBL days under clear-sky (0 ≤ CLD ≤ 20% (a) and overcast

736

conditions (80%≤CLD≤100% (b). Panel (c) shows the spatial distribution with respect

737

to the differences of PBL heights between clear-sky and overcast conditions.

738

Figure 8. The box plots comparing such meteorological factors as (a) cloud cover

739

(CLD), (b) surface temperature (Tsfc), (c) relative humidity (RHsfc), (d) 10-m wind

740

speed (WS10), and (e) surface sensible heat flux (SHFsfc) between the days with SBL

741

and those with CBL, as averaged over 67 radiosonde sites with valid SBL observations.

742

Figure 9. The correlation coefficient of PBL heights and meteorological parameters in

743

six ROIs under SBL (in black), NBL (in red), and CBL (in blue) conditions. The

744

meteorological variables include surface temperature (Tsfc), surface pressure (Psfc),

745

surface relatively humidity (RHsfc), 10-m wind speed (WS10), V component of 10-m

746

wind speed (WS(V)10), U component of 10-m wind speed (WS(U)10). The gray

747

shadings indicate the correlations are statistically significant at 95% confidence level.

748

Figure 10. The box plots comparing several meteorological factors between clear-sky

749

days and overcast days, including (a) surface temperature (Tsfc), (b) surface relative

750

humidity (RHsfc), (c) 10-m wind speed (WS10), and (d) surface sensible heat flux

751

(SHFsfc). Note that only the radiosonde sites that have more than 30 observations for

752

both clear-sky and overcast days were considered.

32

753

Table list

754

Table 1. The statistics of mean value and standard deviation (sd) of PBL heights (in

755

unites of m AGL) for SBL, NBL and CBL in six ROIs, including (1) the North China

756

Plain (NCP), (2) the Yangtze River Delta (YRD), (3) the Pearl River Delta (PRD), (4)

757

the Taklamakan Desert (TKD), (5) the Tibetan Plateau (TBP), and (6) the Sichuan Basin

758

(SCB) at 1400 BJT in summer (May-June-July-August) for the period of 2012 to 2016. NCP mean (sd)

YRD mean (sd)

PRD mean (sd)

TKD mean (sd)

TBP mean (sd)

SCB mean (sd)

SBL

109(110)

199(184)

174(163)

114(98)

211(197)

-

NBL

1529(712)

1241(455)

999(330)

1982(1020)

2188(752)

1300 (338)

CBL

1703(739)

1297(474)

1067(365)

2248(1004)

2222 (751)

1317(409)

759

33

(a) Critical GT(4 vs. 4.5)

PBL height(m, GT=4.5)

2000

SBL(R=0.73)

1500 1000 500 0

0

5000

NBL(R=0.98)

4000

5000

3000

3000

2000

2000

1000

1000

500 1000 1500 2000 PBL height(m, GT=4)

0

0

CBL(R=0.99)

4000

1000 2000 3000 4000 5000 PBL height(m, GT=4)

0

0

1000 2000 3000 4000 5000 PBL height(m, GT=4)

PBL height(m, GT=3.5)

(b) Critical GT(4 vs. 3.5) 1500

SBL(R=0.92)

1000 500 0

0

760

5000 4000

NBL(R=0.98)

5000 4000

3000

3000

2000

2000

1000

1000

500 1000 1500 PBL height(m, GT=4)

0

0

1000 2000 3000 4000 5000 PBL height(m, GT=4)

0

CBL(R=0.98)

0

1000 2000 3000 4000 5000 PBL height(m, GT=4)

(c) Subsample of radiosonde profile

762

1000

5000

5000

800

4000

4000

600

3000

3000

200

1000

1000

Figure 1. SBL(R=1) Scatter plots showing (a)NBL(R=1) PBL heights computed using 4 K/km (in the x-axis) CBL (R=1) PBL height (m)

761

and 4.5 K/km (in the y-axis) as critical potential temperature gradient threshold (GT), 400 2000 2000

763

and (b) the PBL heights computed0 using 4 K/km (in the x-axis) and 3.5 K/km (in the y0 0

764

axis) as critical GT. The results are based on 5399 soundings across China in summer

765

(May-June-July-August) from 2012 to 2016.

0

200 400 600 800 1000 PBL height(m)

0

1000 2000 3000 4000 5000 PBL height(m)

766

34

0

1000 2000 3000 4000 5000 PBL height(m)

30

(a)

Relative Frequency(%)

Frequency (%)

25

100 SBL (N=217, Mean=152 m) NBL (N=1404, Mean=1656 m) CBL (N=3778, Mean=1803 m)

20 15 10

80

60

40

20

5 0

0 0

767

(b)

1000

2000 3000 4000 PBL height (m)

5000

0

1000

2000 3000 4000 PBL height (m)

5000

768

Figure 2. Frequency distribution (a) and cumulative frequency distribution (b) of PBL

769

heights for SBL (in black), NBL (in red), and CBL (in blue) at 1400 BTJ in summer for

770

the period of 2012-2016. The number of soundings (N) and mean value at each

771

observed time are also given.

35

(a)

(b) 50

Latitude (°N)

Latitude (°N)

50 40 30 150 - 180 120 - 150 90 - 120 60 - 90 30 - 60

20

75

85

40 30 20

95

105

115

125

15% - 20% 10% - 15% 5% - 10% 0% - 5%

75

135

85

(c)

Latitude (°N)

Latitude (°N)

115

125

135

125

135

50

40 30 80% - 100% 60% - 80% 40% - 60% 20% - 40% 0% - 20%

75

772

105

(d)

50

20

95

Longitude (°E)

Longitude (°E)

85

40 30 20

95

105

115

125

135

Longitude (°E)

80% - 100% 60% - 80% 40% - 60% 20% - 40% 0% - 20%

75

85

95

105

115

Longitude (°E)

773

Figure 3. Spatial distributions of (a) # of soundings from CMA radiosonde sites and

774

the frequency (in percent) for PBL states classified by atmospheric stability: (b) SBL,

775

(c) NBL and (d) CBL at 1400 BJT in summer for the period of 2012-2016, respectively.

36

(a)

(b) 50

Latitude (°N)

Latitude (°N)

50 40 30 20

40 30 400 300 200 100 0 -

20

75

85

95

105

115

125

135

75

- 500 - 400 - 300 - 200 100

85

Longitude (°E)

115

125

135 Unit:m

(d) 50

Latitude (°N)

50

Latitude (°N)

105

Longitude (°E)

(c)

40 30

40 30 20

20 75

776

95

85

95

105

115

125

135

75

Longitude (°E)

85

95

105

115

125

135

3000 - 3200 2800 - 3000 2600 - 2800 2400 - 2600 2200 - 2400 2000 - 2200 1800 - 2000 1600 - 1800 1400 - 1600 1200 - 1400 1000 - 1200 800 - 1000 600 - 800 400 - 600 200 - 400 0 - 200

Longitude (°E)

777

Figure 4. Spatial distribution of mean PBL height for (a) all types, (b) SBL, (c) NBL,

778

and (d) CBL at 1400 BJT in summer for the period of 2012-2016.

37

779

100

59

Frequency (%)

80

28

28

N=468 6

44

19

22

N=171

N=433

N=352

7

18

52

47 53

87

N=70

N=691

30

14

49

37 (%)

60

May June July August SBL NBL CBL

40

20

0

780

NCP

YRD

PRD

TKD

TBP

SCB

781

Figure 5. The occurrence frequency of SBL, NBL and CBL in six ROIs, including (1) the North

782

China Plain (NCP), (2) the Yangtze River Delta (YRD), (3) the Pearl River Delta (PRD), (4) the

783

Taklamakan Desert (TKD), (5) the Tibetan Plateau (TBP), and (6) the Sichuan Basin (SCB), at 1400

784

BJT during in summer (May-June-July-August) for the period of 2012 to 2016, respectively. The

785

bars show the frequency of different PBL types (SBL in black, NBL in gray, and CBL in white). The

786

pie charts show the frequency percentage distribution for each month.

38

PBL Height (m)

4000

(a)

SBL NBL CBL

3000 2000 1000 0 0

Frequency(%)

80

20

40

CLD (%)

60

(b)

80

100

80

100

SBL NBL CBL

60 40 20 0 0

20

40

60 CLD (%)

787 788

Figure 6. The bar charts showing the mean PBL height (a) and the occurrence

789

frequency (b) for three PBL types, including SBL (in gray), NBL (in red), and CBL (in

790

blue) under different cloud cover (CLD) conditions (with a bin size of 20%) in summer

791

for the period of 2012-2016. Note that the error bar in (a) denotes the standard deviation

792

of PBL height.

793

794

39

Latitude (°N)

(a) 50

50

40

40

30

30

20

20 75

Unit:m

(b)

85

95

105

115

125

135

75

Longitude (°E)

3000 - 3200 2800 - 3000 2600 - 2800 2400 - 2600 2200 - 2400 2000 - 2200 1800 - 2000 1600 - 1800 1400 - 1600 1200 - 1400 1000 - 1200 800 - 1000 600 - 800 400 - 600 200 - 400 0 - 200

85

95

105

115

125

135

Longitude (°E)

(c)

Latitude (°N)

50

40

1600 - 1800 700 - 800 600 - 700 500 - 600 400 - 500 300 - 400 200 - 300 100 - 200 0 - 100

30

20 75

795

85

95

105

115

125

135

Longitude (°E)

796

Figure 7. The spatial distributions of mean PBL heights at 1400 BJT in summer for the

797

period 2012-2016 for CBL days under clear-sky (0 ≤ CLD ≤ 20% (a) and overcast

798

conditions (80%≤CLD≤100% (b). Panel (c) shows the spatial distribution with respect

799

to the differences of PBL heights between clear-sky and overcast conditions.

800

40

(a) All sites 100

100 40

20

(a) SBL

0

CBL

(b)

500

WS10(m/s)

2

SHFsfc (W/m )

15 10 5

40

0

SBL

600

(d)

60

20

10

20

20

RHsfc (%)

o

40

0

80

30

60

Tsfc ( C)

CLD (%)

80

CBL

(c) SBL

CBL

(e)

400 300 200 100 0

0

SBL

CBL

-100

SBL

CBL

801 802

Figure 8. The box plots comparing such meteorological factors as (a) cloud cover

803

(CLD), (b) surface temperature (Tsfc), (c) relative humidity (RHsfc), (d) 10-m wind

804

speed (WS10), and (e) surface sensible heat flux (SHFsfc) between the days with SBL

805

and those with CBL, as averaged over 67 radiosonde sites with valid SBL observations.

41

R

0.8 (c)PRD 0.4 0.0 -0.4 -0.8

0.8 (d)TKD 0.4 0.0 -0.4 -0.8

0.8 (e)TBP 0.4 0.0 -0.4 -0.8 Tsfc

0.8 (f)SCB 0.4 0.0 -0.4 -0.8 Tsfc WS(V)10WS(U)10

R

0.8 (b)YRD 0.4 0.0 -0.4 -0.8

R

SBL 0.8 (a)NCP 0.4 0.0 -0.4 -0.8

806

Psfc

RHsfc WS10

Psfc

NBL

CBL

RHsfc WS WS(V) WS(U)10 10 10

807

Figure 9. The correlation coefficient of PBL heights and meteorological parameters in

808

six ROIs under SBL (in black), NBL (in red), and CBL (in blue) conditions. The

809

meteorological variables include surface temperature (Tsfc), surface pressure (Psfc),

810

surface relatively humidity (RHsfc), 10-m wind speed (WS10), V component of 10-m

811

wind speed (WS(V)10), U component of 10-m wind speed (WS(U)10). The gray

812

shadings indicate the correlations are statistically significant at 95% confidence level.

42

40

100

(a)

80 RHsfc (%)

o

Tsfc ( C)

30 20

40 20

Clear-sky

Clear-sky

Overcast

Clear-sky

Overcast

(d)

400

2

SHFsfc (W/m )

WS10(m/s)

500

(c)

10 5 0

0

Overcast

15

813

60

10 0 20

(b)

300 200 100 0

Clear-sky

Overcast

814

Figure 10. The box plots comparing several meteorological factors between clear-sky

815

days and overcast days, including (a) surface temperature (Tsfc), (b) surface relative

816

humidity (RHsfc), (c) 10-m wind speed (WS10), and (d) surface sensible heat flux

817

(SHFsfc). Note that only the radiosonde sites that have more than 30 observations for

818

both clear-sky and overcast days were considered.

819

43