Draft
Cloud regulations on the gross primary productivity of a
poplar plantation under different environmental conditions
Journal: Canadian Journal of Forest Research
Manuscript ID cjfr-2016-0413.R1
Manuscript Type: Article
Date Submitted by the Author: 16-Dec-2016
Complete List of Authors: Xu, Hang; Beijing Forestry University, institute of soil and water conservation Zhang, Zhiqiang; Beijing Forestry University, College of Soil and Water Conservation Chen, Jiquan; Michigan State University, Landscape Ecology & Ecosystem Science (LEES) Lab, Center for Global Change and Earth Observations
(CGCEO), and Department of Geography Zhu, Mengxun; Beijing Forestry University, College of Soil and Water Conservation Kang, Manchun; Beijing Forestry University, institute of soil and water conservation; China Three Gorges University, College of Hydraulic and Environmental Engineering
Keyword: cloudiness, gross primary productivity, poplar plantation, environmental conditions, path analysis
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
1
Cloud regulations on the gross primary productivity of a poplar plantation under different 1
environmental conditions 2
3
4
Hang Xu1, Zhiqiang Zhang
1, Jiquan Chen
2, Mengxun Zhu
1, Manchun Kang
1,3 5
1. Key Laboratory of Soil and Water Conservation and Desertification Combating, Ministry 6
of Education, College of Soil and Water Conservation, Beijing Forestry University, 7
Beijing 100083, PR China 8
2. Landscape Ecology & Ecosystem Science (LEES) Lab, Center for Global Change and 9
Earth Observations (CGCEO), and Department of Geography, Michigan State University, 10
East Lansing, MI 48823, USA 11
3. College of Hydraulic and Environmental Engineering, China Three Gorges University, 12
Yichang 443002, PR China 13
Corresponding author: Dr. Zhiqiang Zhang (email: [email protected]). 14
Tel.: +86 10 62338097; fax: +86 10 62337873. 15
Page 1 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
2
Abstract: Cloud regulates the gross primary productivity (GPP) of forest ecosystems by 16
changing the radiation component and other environmental factors. In this study, we used an 17
open-path eddy covariance system and microclimate sensors installed over a poplar plantation in 18
northern China to measure the carbon exchange and climate variables during the mid-growing 19
seasons (June to August) in 2014 and 2015. The results indicated that the GPP of the plantation 20
peaked when the clearness index (CI) was between 0.45 and 0.65, at which point diffuse 21
photosynthetically active radiation (PARdif) had reached its maximum. Cloudy skies increased 22
the maximum ecosystem photosynthetic capacity (Pmax) by 28% compared with clear skies. 23
PARdif and soil moisture were the most and the least crucial driver for photosynthetic 24
productivity of the plantation under cloudy skies, respectively. The ecosystem photosynthetic 25
potential was higher under lower vapor pressure deficit (VPD < 1.5 kPa), lower air temperature 26
(Ta < 30°C), and non-stressed conditions (REW > 0.4) for cloudy skies due to effects of Ta and 27
VPD on stoma. Overall, our research highlighted the importance of cloud-induced radiation 28
component change and environmental variation in quantifying the GPP of forest ecosystems. 29
Key words: cloudiness, gross primary productivity, poplar plantation, environmental conditions, 30
path analysis 31
32
Page 2 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
3
1. Introduction 33
The global dimming phenomenon, caused by the drastic increase in clouds and atmospheric 34
particulates produced by fossil fuel combustion and other human activities, has continuously 35
decreased the amount of global solar radiation reaching the ground since the 1950s by 0.51 ± 36
0.05 W m-2
yr-1
(IPCC 2007; Wang et al. 2014). Despite this trend, there have been 37
improvements for most areas of Europe and North America (Barnes et al. 2014), yet a sustaining 38
decline in global radiation exists in the developing countries of the Northern Hemisphere (Streets 39
et al. 2006; Wang and Yang 2014). Clouds and particulates can directly decrease total radiation 40
while increasing diffuse radiation (Oliphant et al. 2011; Zhang et al. 2011), consequently 41
affecting the growth and development of terrestrial ecosystems (Bai et al. 2012; Cheng et al. 42
2015). 43
Numerous studies have indicated that increased diffuse radiation under cloudy sky 44
conditions can drastically enhance the gross primary productivity (GPP) of forest ecosystems and 45
light use efficiency (LUE; defined as GPP/PAR) in regions that are not saturated with light 46
(Kanniah et al. 2012, 2013; Cheng et al. 2015). On a global scale, the carbon sink of terrestrial 47
ecosystems between 1960 and 1999 were estimated to increase by ~25% due to the increase in 48
the diffuse fraction largely associated with the ‘global dimming phenomenon’ (Mercado et al., 49
2009). The underlying mechanisms for this increase are multi-fold; the increase in diffuse 50
radiation can 1) lead to the removal of canopy photoinhibition in sunlit leaves (i.e., leaves above 51
the canopy) at high levels of radiation under clear sky conditions (Gu et al. 2002; Knohl and 52
Baldocchi 2008), 2) uniformly distribute the radiation in the canopy columns (Greenwald et al. 53
2006; Alton et al. 2007), and 3) stimulate plant stomatal opening, which is sensitive to the ratio 54
of blue to red light (Dengel and Grace 2010). 55
Page 3 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
4
Cloudiness and its temporal changes are often coupled with different environmental 56
conditions, such as vapor pressure deficit (VPD), air temperature (Ta) and soil moisture (Kanniah 57
et al. 2012, 2013), all of which produce complex, coupled effects on ecosystem photosynthesis 58
and production (Cai et al. 2008). A few studies found that Ta and VPD had no significant effects 59
on controlling the ecosystem photosynthesis response to diffuse radiation (Jing et al. 2010; 60
Kanniah et al. 2011; Oliphant et al. 2011), while other studies found that these factors played a 61
crucial role (Yamasoe et al. 2005; Zhang et al. 2011). Yamasoe et al. (2005) reported that the 62
increase in VPD under clear sky conditions could constrain carbon exchange in a tropical forest. 63
Zhang et al. (2011) found that decreases in VPD and Ta could depress canopy photosynthesis 64
under cloudy skies in an irrigated maize cropland. Clearly, the significance of the radiation 65
component and other environmental factors on ecosystem GPP remain unclear within these 66
complex, interactive environmental conditions. Moreover, there are few studies that have 67
quantified the effects of cloud-induced radiation components and other environmental factor 68
changes on the ecosystem photosynthetic productivity of plantations. For example, recent studies 69
based on a path analysis of the direct correlation between environmental factors and ecosystem 70
GPP have demonstrated the complex nature of the above relationships (Shao et al. 2016). 71
Here, we hypothesized that the diffuse PAR (PARdif), mainly controlled by clouds, would be 72
the dominant factor regulating forest ecosystem GPP under cloudy skies and that clouds would 73
affect the GPP distinctively under different environmental conditions. To test the hypotheses, we 74
used diurnal 30-min data (PAR > 4 µmol m-2
s-1
) collected during the mid-growing season (June 75
to August) in combination with ancillary meteorological observations from an eddy covariance 76
tower over a poplar plantation in northern China, to 1) quantify the cloudiness effect on 77
ecosystem GPP and LUE, 2) explore GPP responses to cloudy skies under different 78
Page 4 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
5
environmental conditions, and 3) identify the direct and indirect effects of environmental 79
variables on ecosystem photosynthesis under various cloudy skies. 80
2. Materials and Methods 81
2.1 Site description 82
This study was conducted over a poplar plantation (Populus × euramericana) in Gongqing 83
Forest Farm (116°42′41″E, 40°06′27″N, 29 m a. s. l.), which is located on a fluvial plain near the 84
Chaobai River in Shunyi District, Beijing, China. The trees were planted in ~1996 at a density of 85
4 m × 3 m. The average height and diameter at breast height (DBH) of the plantation were 17.5 ± 86
1.6 m (mean ± SD) and 25.7 ± 1.6 cm, respectively, in 2014 and 2015. The leaf area index (LAI) 87
varied between 2.15 and 3.41 during the mid-growing season in 2014 and 2015. The understory 88
vegetation was dominated by Swida alba Opiz., Pinus bungeana Zucc., Caryophylli Flos., 89
Sabina vulgaris., and Gaillardia aristata Pursh. The region is typical of the sub-humid warm 90
temperate monsoon climate, with a mean annual temperature of 11.5°C and an average daily 91
temperature of 4.9°C in January and 25.7°C in July. According to the long-term (1990–2010) 92
meteorological Shunyi Weather Station (116°37′E, 40°08′N, 28.6 m a. s. l.), the mean annual 93
precipitation is ~576 mm, with 75% falling between July and September. The mean annual 94
relative humidity is 50% and the total hours of sunshine are ~690 h in the summer (i.e., June to 95
August). The soil texture is sandy with high permeability and a low water-holding capacity. The 96
ground water level during the study was ~2 m. 97
2.2 Fluxes and meteorological measurements 98
An eddy covariance (EC) system was installed on a 30 m tall tower in November 2013 to 99
continuously measure the exchange of carbon, water and energy between the plantation and the 100
atmosphere. An open-path gas analyzer (EC150, Campbell Scientific Inc., USA) and a 3-D sonic 101
Page 5 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
6
anemometer (CSAT3, Campbell Scientific Inc., USA) were installed at 26 m with a north 102
declination of 165° (i.e., the prevailing wind direction). 103
The global shortwave solar radiation (G), photosynthetically active radiation (PAR) and net 104
radiation (Rn) were measured at the top of the tower (30 m) by a pyranometer (LI-200x, LI-COR 105
Inc., USA), a light quantum sensor (LI-190SB, LI-COR Inc., USA), and a net radiometer (CNR1, 106
Kipp and Zonen., NL), respectively. Two tipping-bucket rain gauges (TE525-L, Campbell 107
Scientific Inc., USA) were mounted above the canopies to measure the cumulative precipitation 108
over 30 min periods. Air temperature and relative humidity were measured at 0.5, 1.5, 5, 15 and 109
30 m above the ground using temperature and humidity sensors (HC2S3, Campbell Scientific 110
Inc., USA), from which VPD was calculated. Soil moisture was measured by TDR (CS616, 111
Campbell Scientific Inc., USA) at 5, 20, 50, 100, 150 and 200 cm below the soil surface. Soil 112
temperature (Ts) and soil heat flux (G) were measured with two thermocouples (TCAV107, 113
Campbell Scientific Inc., USA) and two soil heat transducers (HFT3, Campbell Scientific Inc., 114
USA) installed at depths of 5 and 25 cm. All data were logged into CR3000/1000 dataloggers 115
(Campbell Scientific Inc., USA) at 2/10 Hz. The leaf area index was measured twice a month 116
using the plant canopy analyzer (LAI-2200, LI-COR Inc., USA). 117
2.3 Flux corrections and data processing 118
Eddypro (https://www.licor.com/env/support/) was used to process the 10 Hz raw data, 119
including spike filtering, rotating the coordinates’ axis using the planar fit method (Foken et al. 120
2012) and WPL corrections (Webb 1982). Stationary and integral turbulent tests were performed 121
(Foken 2006), splitting all data into 9 classes. The data of classes 9 and 8 with wind directions 122
between > 350° and < 45° (north declination) were discarded because of insufficient turbulence 123
and because they fell under the tower’s shade. The data with friction velocity below the threshold 124
Page 6 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
7
0.20 m s-1
for 2014 and 0.25 m s-1
for 2015 (Reichstein et al. 2005) were filtered out to eliminate 125
the underestimation under stable stratifications. Simultaneously, the KM model was applied to 126
remove data from beyond the study area (Kormann and Meixner 2001). With these quality 127
controls and other system malfunctions (e.g., power failure, instrumentation malfunctioning and 128
heavy rain conditions), 49% in 2014 and 52% in 2015 of data were finally reserved. Daytime 129
gaps less than < 7 days were filled by using the marginal distribution sampling (MDS) approach 130
(Reichstein et al. 2005). In addition, the larger gaps (i.e., more than 7 days) were filled by the 131
mean diurnal variation (MDV) method with consecutive 15-day windows because the largest gap 132
was less than 10 days (Krishnan et al. 2012). 133
The energy balance ratio (EBR), defined as effective energy/available energy, was 0.93 and 134
0.85 based on daytime 30-min fluxes in 2014 and 2015, respectively. Although there was energy 135
storage in the canopy and air, our results were similar to the average value (0.84) of 173 136
FLUXNET sites (Stoy et al. 2013). We applied a non-linear regression method to calculate the 137
ecosystem respiration (ER, i.e., nighttime NEE) and GPP (Lloyd and Taylor 1994) as follows: 138
(1)
)T)t(T
1
TT
1(E
fre0soil0ref
short,0
eR)t(ER−
−−
⋅
⋅= 139
(2) NEE-ERGPP = 140
where Rref is the ER under the reference temperature, E0,short denotes the activation energy 141
parameter that determines the short-term temperature sensitivity (4 days in this study), Tref is the 142
reference temperature (10°C), T0 is a constant (-46.02°C) and Tsoil is the soil temperature at 5 cm. 143
NEE is the net ecosystem exchange. 144
To compare the GPP responses to sky conditions under different environmental conditions, 145
we divided the VPD into VPD < 1.5 kPa and VPD ≥ 1.5 kPa (Zhou et al. 2013), Ta into Ta < 146
Page 7 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
8
30°C and Ta ≥ 30°C (Kanniah et al. 2013) and relative extractable soil water (REW) into REW < 147
0.4 and REW ≥ 0.4, representing soil water stress and non-stress, respectively (Granier et al. 148
1999). REW was calculated as follows: 149
(3) minmax
min
VWCVWC
VWCVWCREW
−−
= 150
where VWCmax and VWCmin are the maximum and minimum soil moisture content measured at 5 151
cm. 152
2.4 Defining sky conditions 153
The clearness index (CI) that is the ratio of solar radiation (G) observed above the canopies 154
to the extraterrestrial global horizontal solar radiance (G0) (Kanniah et al. 2013), was used to 155
represent the sky conditions (Gu et al. 2002) as follows: 156
(4) 0G
GCI =
157
(5) βsin365
d360cos033.01GG sc0
+= 158
where Gsc is the solar constant (1370 W m-2
), d is the day of the year, and β is the solar 159
elevation computed by using the Solar Position Online Calculator operated by the National 160
Renewable Energy Lab, USA (http://www.nrel.gov/midc/solpos/solpos.html). 161
A distinct CI threshold (CIt) to define the sky conditions cannot be found because CI 162
changes not only with clouds and aerosol concentration but also with the variation of solar 163
elevation (Gu et al. 1999). CIt at a given solar elevation must be smoothly increased with sinβ 164
and form an envelope curve in the lumped scatter plot of CIt against sinβ (Gu et al. 1999). Due to 165
the asymmetry between mornings and afternoons, we calculated the CIt and its relationship with 166
Page 8 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
9
sinβ separately (Fig. 1) with a cubic polynomial equation (Eq. 6) as follows: 167
(6) dβsincβsinbβsinaCI 23
t +++=
168
where a, b, c and d were estimated regression coefficients. Cloudy sky conditions were defined 169
when CI was less than CIt at a given solar elevation. To eliminate the effect of solar elevation on 170
the relationship between ecosystem photosynthesis and CI, we divided the 30-70° solar elevation 171
data (β) into 5° intervals (Zhang et al. 2010; Bai et al. 2012). 172
2.5 Diffuse PAR 173
PARdif was calculated by using CI and solar zenith angle as (Reindl et al. 1991) follows: 174
(7) fDPARPARPAR dif ×=
175
(8) ( )[ ]
( ) ( ) βcosβ90cosq11
qq13.01fDPAR
32
2
−−+−+
=
176
(9) CIG
Gq
0
f
⋅=
177
when 0 ≤ CI ≤ 0.3; restrain Gf/G0 ≤ CI, 178
(10)
( )βsin0123.0CI254.0020.1CIG
G
0
f +−= 179
when 0.3 < CI≤ 0.78; restrain 0.1CI ≤ Gf/G0 ≤ 0.97CI, 180
(11) )βsin177.0CI749.1400.1(CIG
G
0
f +−= 181
when 0.78 < CI; restrain Gf/G0 ≥CI, 182
Page 9 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
10
(12)
)βsin182.0CI486.0(CIG
G
0
f −=
183
where fDPAR is the fraction of diffuse PAR and Gf is the diffuse fraction of the global solar 184
radiation (W m-2
). 185
2.6 Light response model 186
The rectangle hyperbola light response model, known as the Michaelis-Menten equation, 187
which can simulate the process of leaf and canopy level photosynthesis (Michaelis and Menten 188
1913), is widely used for quantifying the effects of cloudiness on ecosystem photosynthesis 189
under different sky conditions. The formula is as follows: 190
(13) max
max
PαPAR
PARαPGPP
+= 191
where α is the ecosystem apparent quantum yield and Pmax is the maximum ecosystem 192
photosynthetic capacity (µmol m-2
s-1
). 193
2.7 Statistical analysis 194
We developed our path coefficient model by focusing on the interactive environmental 195
factors, including PARdif, direct PAR (PARdir), VPD, Ta and REW, to identify their direct and 196
indirect effects on the GPP of the plantation under different environmental conditions. In 197
simplifying the model structure, we only accepted the significant paths (p < 0.05) and 198
standardized path coefficients (ρ) that were out of the range of -0.1 to 0.1. All observed data were 199
standardized before putting into the path model, and the ρ and discrepancy were calculated based 200
on the maximum likelihood method (Shao et al. 2016). The final models were adopted when the 201
comparative fit index (CFI) was more than 0.9 and the root mean square error of approximation 202
(RMSEA) was less than 0.05 (Kline 2011). The path analysis and data conversion were applied 203
Page 10 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
11
within AMOS (version 24.0, Chicago, IL). All selected regression curves were statistically 204
significant (p < 0.05) and based on several related studies (Zhang et al. 2010, Oliphant et al. 2011, 205
Bai et al. 2012, Kanniah et al. 2013), which were conducted by SigmaPlot (version 12.5, 206
California, USA). 207
3. Results 208
3.1 Microclimate and CI 209
The variations in the mean daily CI, REW, Ta, VPD, and daily total global solar radiation 210
(G), daily total precipitation (P) during the mid-growing season of 2014 and 2015 were shown in 211
Fig. 2. The mean daily CI in 2014 was 0.41 ± 0.141 (mean ± SE), which was greater than that in 212
2015 (0.36 ± 0.133). Total daily G showed consistent changes, corresponding to the change in CI 213
(Fig. 2c/d). The mean daily G in 2015 (16.4 ± 0.59 MJ m-2
) was less than that in 2014 (17.0 ± 214
0.53 MJ m-2
). The total G values during the same period from 2014 and 2015 were 1561.9 MJ 215
m-2
and 1507.3 MJ m-2
, respectively. The total rainfall during the mid-growing season of 2014 216
was 200.6 mm, 54% less than the long term average (432.1 mm) (Fig. 2a/b). As a result, the 217
plantation experienced 68 days of soil water stress (REW < 0.4). In contrast, the rainfall in 2015 218
was 439.4 mm, resulting in 53 days of drought stress. The mean air temperature (Ta) during the 219
mid-growing season in 2015 (27.9 ± 0.07°C) was lower than that in 2014 (28.1 ± 0.09°C) (Fig. 220
2e/f). The maximum mean daily Ta of 2014 and 2015 was 30.8°C and 33.2°C, respectively, with 221
both occurring in July. During the study period, Ta under clear sky conditions (29.6 ± 0.23°C) 222
was 8% higher than that of cloudy sky conditions (27.1 ± 0.08°C). The mean daily VPD in 2015 223
was 1.6 ± 0.02 kPa, which was lower than that in 2014 (1.8 ± 0.03 kPa). The low Ta and high P in 224
2015 resulted in a lower VPD than that in 2014. 225
In addition, sky conditions indicated by CI during the mid-growing seasons of two years 226
Page 11 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
12
were related to the changes in VPD, Ta, PAR and PARdif (Fig. 3). VPD, Ta and PAR linearly 227
increased with increasing CI (p < 0.05), however, the variation in PARdif against CI presented a 228
downward-parabolic trend under selected solar elevations during the study period. Sensitivities 229
of environmental factors (i.e., the slopes of lines) to CI were higher under the solar elevations of 230
65-70° compared to the solar elevations of 45-50°, similarly, higher coefficients of determination 231
were also found at higher solar elevations. 232
3.2 GPP and LUE against CI 233
The variations in the GPP to CI were conic across all ranges of β in the study period (Fig. 4). 234
The GPP peaked when CI ranged between 0.45 and 0.65 and was reduced when the skies were 235
more cloudy or sunny. The mean diurnal GPP rate under cloudy sky conditions (21.93 ± 0.744 236
µmol m-2
s-1
) in two years was higher than that in clear skies (20.01 ± 1.386 µmol m-2
s-1
). More 237
pronounced changes in the GPP against CI were observed at relatively higher solar elevations, 238
demonstrating that ecosystem GPP enhanced more by cloudiness at higher solar elevations. In 239
addition, LUE linearly increased with cloudiness among the solar elevations (Fig. 5) and was 240
more efficient under cloudy skies (0.054 ± 0.001 µmol CO2 µmol-1
PAR) than under clear skies 241
(0.048 ± 0.006 µmol CO2 µmol-1
PAR). 242
3.3 Light response under different sky conditions 243
The canopy photosynthetic productivity responded differently to PAR for clear and cloudy 244
skies (Fig. 6). The apparent quantum efficiency (α) was ~9% lower under cloudy skies (0.106 ± 245
0.003 µmolCO2 µmol-1
PAR) than that under clear skies (0.117 ± 0.009 µmol CO2 µmol-1
PAR), while 246
ecosystem photosynthetic potential (Pmax) was elevated by 28% under cloudy sky conditions 247
(41.51 ± 0.684 µmol m-2
s-1
). Furthermore, with a lower Ta (Ta < 30°C) and VPD (VPD < 1.5 248
kPa), the Pmax of the plantation was increased by 30% and 33% due to the cloudiness (Fig. 7), 249
Page 12 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
13
respectively. These increase rates were higher than that under higher VPD and Ta groups. In 250
addition, Pmax was more sensitive to clouds under non-stressed conditions compared with water 251
stress conditions (Fig. 7). Under cloudy sky conditions, the Pmax value was 33% higher for 252
VPD < 1.5 than that for VPD ≥ 1.5, 12% higher for Ta < 30°C compared with that for Ta ≥ 30°C, 253
and 29% higher for non-stressed soil than that for water stress soil. Our results indicated that 254
lower VPD, lower Ta and non-stressed REW further promoted the ecosystem photosynthetic 255
potential under cloudy skies. 256
3.4 Direct and indirect influences of environmental factors on GPP 257
There were different standardized path coefficients from climate variables to ecosystem 258
photosynthetic productivity (Fig. 8). PARdif was the most important driver for ecosystem GPP 259
under all environmental conditions, with ρ hardly changing with the variation of environmental 260
groups (ρ = 0.54-0.65). Interestingly, soil REW was irrelevant to the GPP across all groups. Ta 261
and direct PAR (PARdir) played positive roles (i.e., ρ > 0) on the GPP, while VPD suppressed 262
ecosystem photosynthesis in most cases. 263
An environmental variable had different direct effects on GPP under different environmental 264
conditions. The variation of path coefficients between radiation (i.e., PARdif and PARdir) and GPP 265
was minimal while the variation between VPD and GPP was maxed (ρ = -0.52-0) among the 266
environmental groups. A non-correlation between VPD and GPP was found when VPD was 267
lower than 1.5 kPa. In addition, the influence of VPD on GPP under higher VPD (VPD ≥ 1.5 kPa) 268
and higher Ta (Ta ≥ 30°C) conditions was more negative than that under lower VPD and lower Ta. 269
VPD depressed photosynthesis more severely under soil water stress conditions than 270
non-stressed periods, and Ta was more relevant to GPP when it was lower compared to higher, 271
while there were no discrepancies within the other environmental groups. 272
Page 13 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
14
The indirect effects of Ta on GPP were negative, indicating that the increase in Ta would 273
indirectly lead to the decrease in GPP via increasing VPD. There were more significant indirect 274
effects under higher VPD (-0.32) and water-stress free conditions (-0.41) compared to lower 275
VPD and water stress conditions. Similarly, the indirect effects of Ta on GPP under high Ta 276
conditions (-0.26) were less than that (-0.14) under low Ta. Consequently, the total effects (i.e., 277
sum of direct and indirect effects) of Ta on GPP were positive only under lower VPD, lower Ta or 278
non-stressed conditions. 279
4. Discussion 280
4.1 Effects of cloudiness on GPP 281
Moderate cloud coverage (0.45-0.65) and corresponding diffuse radiation fractions between 282
0.50 and 0.80 presented optimal conditions that enhanced ecosystem productivity in the poplar 283
plantation. Similar results were also observed within a temperate broadleaf forest in North 284
America (Oliphant et al. 2011) and a temperate broadleaved pine forest in Northwestern China 285
(Zhang et al. 2010). However, Kanniah et al. (2013) found that GPP rate was highest under clear 286
sky conditions in a tropical savanna. Radiation component changes and light distribution within 287
an ecosystem influences the canopy’s photosynthetic productivity under cloudy sky conditions 288
(Knohl and Baldocchi 2008). Leaf area index and leaf orientation/arrangements within canopies 289
are the primary factors determining scattered radiation usage in an ecosystem (Greenwald et al. 290
2006; Kanniah et al. 2012). Leaf inclination angles of the sunlit leaves of the poplar plantation 291
were less than those of the sunshade leaves, which utilizes direct light and diffuse light 292
simultaneously (Dickmann et al. 1990). 293
Ecosystem LUE consistently increased with cloudiness without any discrepancy among the 294
solar elevations (p > 0.05) (Fig. 5), which further demonstrates the positive effects that 295
Page 14 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
15
cloudiness has on GPP. Cloudiness reduced the direct PAR for sunlit leaves and increased the 296
PARdif for shade leaves. These changes in PAR will reduce photosynthesis in the canopy’s sunlit 297
foliage (within the photo prohibition range) but elevate the GPP of the shade leaves. Shade 298
foliage and understory plants received more radiation to enhance their LUE and GPP (Alton et al. 299
2007; Kanniah et al. 2011). As total radiation continued to decline against cloudiness, ecosystem 300
GPP began to decrease because the radiation on the sunlit leaves was less than the light 301
saturation point (Fig. 6). The reduction in GPP could not be compensated by the enhanced 302
photosynthesis within the shade foliage or understory plants, which benefited from the diffuse 303
radiation (Misson et al. 2005). Our data showed that cloudiness had strong effects on the 304
plantation’s GPP and LUE, and that while LUE was enhanced by cloudiness, it could not lead to 305
an increase in ecosystem GPP under conditions dominated by scattered radiation (i.e., fDPAR > 306
0.5) in this plantation. 307
4.2 Effects of CI on environmental factors 308
Canopy GPP and other physiological processes were jointly influenced by multiple 309
environmental factors, such as radiation, VPD, air temperature and soil moisture status (Yuan et 310
al. 2009; Tian et al. 2011). Ta linearly decreased with CI (Fig. 3c) due to reduced incident 311
radiation (Gu et al. 1999). Likewise, VPD was lowered by cloudiness (Fig. 3a). Overall, clouds 312
lowered leaf temperature and increased air humidity, thus reducing VPD. Although VPD, Ta and 313
PAR decreased with cloudiness for selected solar elevations (p < 0.05), the changes in 314
environmental factors could not be well interpreted by CI at a lower solar elevation, indicating 315
that the decrease in VPD and Ta at a lower solar elevation was not solely attributed by cloudiness. 316
This result might be due to the strong positive correlations between VPD, Ta and solar elevations 317
(pVPD < 0.001, pTa < 0.001). Under clear sky conditions, CI increased with sunrise (Perez et al. 318
Page 15 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
16
1990; Misson et al. 2005). Thus, the decrease in VPD and Ta with decreasing CI was partly 319
induced by the decline in solar elevations at lower solar elevations. Unlike the linear relation of 320
PAR with CI, the PARdif was nonlinearly correlated with CI (Fig. 3d), as indicated by previous 321
studies (Alton et al. 2007; Zhang et al. 2011). PARdif initially increased with the increase in CI 322
and peaked until CI reached 0.40–0.60, and PARdif declined afterwards in accordance with GPP 323
(Fig. 4). Thus, PARdif played a crucial role in controlling and regulating the plantation’s GPP 324
under cloudy skies. 325
4.3 Environment regulations on GPP under different conditions of cloudiness 326
Cloudiness induced variations in radiation components and other environmental factors that 327
led to interactive effects on forest ecosystem productivity. Our path analysis results showed that 328
PARdif was the primary factor under cloudy skies controlling photosynthesis in this fast-growing 329
polar plantation, with no obvious discrepancy for the environmental groups (Fig. 8). This 330
indicated that PARdif did not cause the obvious difference in GPP between clear and cloudy skies 331
under lower VPD, lower Ta and non-stressed periods. The increase in VPD limited the poplar’s 332
photosynthesis under different environmental conditions because leaf guard cell behavior is 333
highly controlled by the balance of vapor pressure inside and outside of leaves (Farquhar and 334
Sharkey 1982) and tends to be closed (i.e., decrease stomatal conductance) when there is an 335
increase in VPD (Zhou et al. 2013; Goodrich et al. 2015). The sensitivity of ecosystem 336
photosynthesis to VPD varied by environmental conditions (Day 2000). The discrepancy is due 337
to the degree of deficit in CO2 and H2O for leaf photosynthesis. Ta directly played a positive role 338
in ecosystem photosynthesis because increased Ta enhanced enzyme activity and photosynthetic 339
electron transfer efficiency (Berry and Bjorkman 1980), which the poplar plantation could have 340
used to promote leaf stomatal conductance, transpiration rates, and higher photosynthetic rates 341
Page 16 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
17
(Song et al. 2014). Additionally, increases in Ta also resulted in increased VPD, leading to the 342
indirect depression of ecosystem photosynthesis except for lower VPD conditions. These results 343
confirmed that the regulation of ecosystem photosynthesis by environmental factors was 344
complicated and coupled. Moreover, poplar species consume large quantities of water and are 345
usually sensitive to soil water content (Kang et al. 2015). However, REW appeared to have been 346
the least important factor for GPP under any environmental conditions likely because the 347
taproots of the plants absorbed deep soil water in this plantation’s sandy soil, where sufficient 348
underground water exists during water stress periods (Comas et al. 2013; Su et al. 2014). 349
Furthermore, there might have been a lagged photosynthesis response to the changes in soil 350
water content, which lowers the correlation strength (Jia et al. 2014). 351
Correlation coefficients of the environmental factors among the different groups indicated 352
that the GPP increase in varying degrees by cloudiness was directly ascribed to the discrepancy 353
in the total effects of Ta (i.e., the balance between negative effects of VPD and positive effects of 354
Ta ) on leaf stoma rather than the effects of PARdif (Fig. 8). Clearly, the role of each 355
environmental factor in regulating GPP and photosynthesis needs to be understood in the context 356
of other variables, including cloudiness. 357
5. Conclusions 358
Cloudiness induced changes in radiation components while other environmental factors had 359
coupling effects on the GPP of a fast-growing poplar plantation in northern China. The GPP of 360
the plantation peaked with a CI of 0.45-0.65 when diffuse radiation was the largest component of 361
the total solar radiation. Clouds increased ecosystem Pmax and LUE, leading to the continuous 362
increase of canopy GPP until scattering radiation dominated. Our path analysis indicated that 363
PARdif was the most crucial driver while soil moisture was the least important factor for 364
Page 17 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
18
photosynthetic productivity of the poplar plantation under cloudy sky conditions. In addition, 365
lower VPD (VPD < 1.5 kPa), lower Ta (Ta < 30°C) and non-stressed conditions (REW > 0.4) 366
were more conducive to enhancing ecosystem photosynthesis under cloudy skies due to the 367
comprehensive effects of VPD and Ta to stomatal conductance. Therefore, evaluating the 368
cloudiness effects for quantifying the GPP of forest ecosystems is important. Although we 369
quantified the direct and indirect environmental effects on GPP under cloudy skies, the weakness 370
of path analysis was that empirical relationships between ecosystem GPP and environmental 371
factors were only considered. Thus further studies focusing on more biophysical processes 372
should be conducted to explore more accurately the cloud effect on forest ecosystem 373
productivity. 374
375
Acknowledgments 376
This study was financially supported by the Beijing Education Commission Grant to jointly 377
support the Universities of Ministry of Education in Beijing and the National Special Research 378
Program for Forestry, which was entitled “Forest Management Affecting the Coupling of 379
Ecosystem Carbon and Water Exchange with Atmosphere” (grant no. 201204102). The fourth 380
author also acknowledges the financial support of the Beijing Municipality Educational 381
Committee under the graduate student training program. Partial support by the US–China Carbon 382
Consortium (USCCC) is also acknowledged. The authors thank Gabriela Shirkey for editing the 383
manuscript for language.384
Page 18 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
19
References
Alton, P.B., North, P.R., and Los, S.O. 2007. The impact of diffuse sunlight on canopy light-use efficiency,
gross photosynthetic product and net ecosystem exchange in three forest biomes. Glob. Chang. Biol. 13(4):
776–787. doi:10.1111/j.1365-2486.2007.01316.x.
Bai, Y.F., Wang, J., Zhang, B.C., Zhang, Z.H., and Liang, J. 2012. Comparing the impact of cloudiness on
carbon dioxide exchange in a grassland and a maize cropland in northwestern China. Ecol. Res. 27(3):
615–623. doi:10.1007/S11284-012-0930-Z.
Barnes, E.A., Dunn-Sigouin, E., Masato, G., and Woollings, T. 2014. Exploring recent trends in Northern
Hemisphere blocking. Geophys. Res. Lett. 41(2): 638–644. doi:10.1002/2013GL058745.
Berry, J., and Bjorkman, O. 1980. Photosynthetic Response and Adaptation to Temperature in Higher Plants.
Annu. Rev. Plant Physiol. 31(1): 491–543. doi:10.1146/annurev.pp.31.060180.002423.
Cai, T., Flanagan, L.B., Jassal, R.S., and Black, T.A. 2008. Modelling environmental controls on ecosystem
photosynthesis and the carbon isotope composition of ecosystem-respired CO2 in a coastal Douglas-fir
forest. Plant, Cell Environ. 31(4): 435–453. doi:10.1111/j.1365-3040.2008.01773.x.
Cheng, S.J., Bohrer, G., Steiner, A.L., Hollinger, D.Y., Suyker, A., Phillips, R.P., and Nadelhoffer, K.J. 2015.
Variations in the influence of diffuse light on gross primary productivity in temperate ecosystems. Agric.
For. Meteorol. 201: 98–110. doi:10.1016/j.agrformet.2014.11.002.
Comas, L.H., Becker, S.R., Cruz, V.M. V, Byrne, P.F., and Dierig, D. a. 2013. Root traits contributing to plant
productivity under drought. Front. Plant Sci. 4(11): 442. doi:10.3389/fpls.2013.00442.
Day, M.E. 2000. Influence of temperature and leaf-to-air vapor pressure deficit on net photosynthesis and
stomatal conductance in red spruce (Picea rubens). Tree Physiol. 20: 57–63. doi:10.1093/treephys/20.1.57.
Dengel, S., and Grace, J. 2010. Carbon dioxide exchange and canopy conductance of two coniferous forests
under various sky conditions. Oecologia 164(3): 797–808. doi:10.1007/s00442-010-1687-0.
Dickmann, D.I., Michael, D. a., Isebrands, J.G., and Westin, S. 1990. Effects of leaf display on light
interception and apparent photosynthesis in two contrasting Populus cultivars during their second growing
season. Tree Physiol. 7(1-2-3-4): 7–20. doi:10.1093/treephys/7.1-2-3-4.7
Farquhar, G.D., and Sharkey, T.D. 1982. Stomatal Conductance and Photosynthesis. Annu. Rev. Plant Physiol.
33(1): 317–345. doi:10.1146/annurev.pp.33.060182.001533.
Foken, T. 2006. Angewandte Meteorologie. Springer: 325. doi:10.1007/978-3-540-38204-1.
Foken, T., Aubinet, M., and Leuning, R. 2012. The Eddy Covariance Method. In Eddy Covariance. pp. 1–19.
doi:10.1007/978-94-007-2351-1.
Goodrich, J.P., Campbell, D.I., Clearwater, M.J., Rutledge, S., and Schipper, L.A. 2015. High vapor pressure
deficit constrains GPP and the light response of NEE at a Southern Hemisphere bog. Agric. For. Meteorol.
203: 112–123. doi:10.1016/j.agrformet.2015.01.001.
Granier, A., Bréda, N., Biron, P., and Villette, S. 1999. A lumped water balance model to evaluate duration and
Page 19 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
20
intensity of drought constraints in forest stands. Ecol. Modell. 116(2–3): 269–283.
doi:10.1016/S0304-3800(98)00205-1.
Greenwald, R., Bergin, M.H., Xu, J., Cohan, D., Hoogenboom, G., and Chameides, W.L. 2006. The influence
of aerosols on crop production: A study using the CERES crop model. Agric. Syst. 89(2–3): 390–413.
doi:10.1016/j.agsy.2005.10.004.
Gu, L.H., Baldocchi, D., Verma, S.B., Black, T.A., Vesala, T., Falge, E.M., and Dowty, P.R. 2002. Advantages
of diffuse radiation for terrestrial ecosystem productivity. J. Geophys. Res. Atmos. 107(D6): ACL 2-1.
doi:10.1029/2001JD001242.
Gu, L.H., Fuentes, J.D., Shugart, H.H., Staebler, R.M., and Black, T. a. 1999. Responses of net ecosystem
exchanges of carbon dioxide to changes in cloudiness: Results from two North American deciduous
forests. J. Geophys. Res. 104(D24): 31421. doi:10.1029/1999JD901068.
IPCC. 2007. IPCC Fourth Assessment Report (AR4). IPCC 1: 976. doi:ISSN: 02767783.
Jia, X., Zha, T.S., Wu, B., Zhang, Y.Q., Gong, J.N., Qin, S.G., Chen, G.P., Qian, D., Kellomäki, S., and Peltola,
H. 2014. Biophysical controls on net ecosystem CO2 exchange over a semiarid shrubland in northwest
China. Biogeosciences 11(17): 4679–4693. doi:10.5194/bg-11-4679-2014.
Jing, X., Huang, J., Wang, G., Higuchi, K., Bi, J., Sun, Y., Yu, H., and Wang, T. 2010. The effects of clouds and
aerosols on net ecosystem CO2 exchange over semi-arid Loess Plateau of Northwest China. Atmos. Chem.
Phys. 10(17): 8205–8218. doi:10.5194/acp-10-8205-2010.
Kang, M.C., Zhang, Z.Z., Noormets, A., Fang, X.R., Zha, T.G., Zhou, J., Sun, G., McNulty, S.G., and Chen, J.Q.
2015. Energy partitioning and surface resistance of a poplar plantation in northern China. Biogeosciences
12(14): 4245–4259. doi:10.5194/bg-12-4245-2015.
Kanniah, K.D., Beringer, J., and Hutley, L. 2013. Exploring the link between clouds, radiation, and canopy
productivity of tropical savannas. Agric. For. Meteorol. 182–183: 304–313. Elsevier B.V.
doi:10.1016/j.agrformet.2013.06.010.
Kanniah, K.D., Beringer, J., and Hutley, L.B. 2011. Environmental controls on the spatial variability of
savanna productivity in the Northern Territory, Australia. Agric. For. Meteorol. 151(11): 1429–1439.
doi:10.1016/j.agrformet.2011.06.009.
Kanniah, K.D., Beringer, J., North, P., and Hutley, L. 2012. Control of atmospheric particles on diffuse
radiation and terrestrial plant productivity: A review. Prog. Phys. Geogr. 36(2): 209–237.
doi:10.1177/0309133311434244.
Kline, R.B. 2011. Principles and practice of structural equation modeling. In Structural Equation Modeling.
doi:10.1038/156278a0.
Knohl, A., and Baldocchi, D.D. 2008. Effects of diffuse radiation on canopy gas exchange processes in a forest
ecosystem. J. Geophys. Res. Biogeosciences 113(2): 1–17. doi:10.1029/2007JG000663.
Kormann, R., and Meixner, F.X. 2001. An analytical footprint model for non-neutral stratification.
Boundary-Layer Meteorol. 99(2): 207–224. doi:10.1023/A:1018991015119.
Page 20 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
21
Krishnan, P., Meyers, T.P., Scott, R.L., Kennedy, L., and Heuer, M. 2012. Energy exchange and
evapotranspiration over two temperate semi-arid grasslands in North America. Agric. For. Meteorol. 153:
31–44. doi:10.1016/j.agrformet.2011.09.017.
Lloyd, J., and Taylor, J.A. 1994. On the temperature dependence of soil respiration. Funct. Ecol. 8: 315–323.
doi:10.2307/2389824.
Michaelis, L., and Menten, M.L. 1913. Die Kinetik der Invertinwirkung. Biochem Z 49(2): 333–369.
doi:10.1021/bi201284u.
Misson, L., Lunden, M., McKay, M., and Goldstein, A.H. 2005. Atmospheric aerosol light scattering and
surface wetness influence the diurnal pattern of net ecosystem exchange in a semi-arid ponderosa pine
plantation. Agric. For. Meteorol. 129(1–2): 69–83. doi:10.1016/j.agrformet.2004.11.008.
Oliphant, A.J., Dragoni, D., Deng, B., Grimmond, C.S.B., Schmid, H.P., and Scott, S.L. 2011. The role of sky
conditions on gross primary production in a mixed deciduous forest. Agric. For. Meteorol. 151(7):
781–791. Elsevier B.V. doi:10.1016/j.agrformet.2011.01.005.
Perez, R., Ineichen, P., Seals, R., and Zelenka, A. 1990. Making full use of the clearness index for
parameterizing hourly insolation conditions. Sol. Energy 45(2): 111–114.
doi:10.1016/0038-092X(90)90036-C.
Reichstein, M., Falge, E., Baldocchi, D., Papale, D., Aubinet, M., Berbigier, P., Bernhofer, C., Buchmann, N.,
Gilmanov, T., Granier, A., Grünwald, T., Havránková, K., Ilvesniemi, H., Janous, D., Knohl, A., Laurila, T.,
Lohila, A., Loustau, D., Matteucci, G., Meyers, T., Miglietta, F., Ourcival, J.M., Pumpanen, J., Rambal, S.,
Rotenberg, E., Sanz, M., Tenhunen, J., Seufert, G., Vaccari, F., Vesala, T., Yakir, D., and Valentini, R. 2005.
On the separation of net ecosystem exchange into assimilation and ecosystem respiration: Review and
improved algorithm. Glob. Chang. Biol. 11(9): 1424–1439. doi:10.1111/j.1365-2486.2005.001002.x.
Reindl, D.T., Beckman, W.A., and Duffie, J.A. 1991. Diffuse fraction correlations. Sol. Energy 45(1): 1–7.
doi:10.1016/0038-092X(91)90123-E.
Shao, J.J., Zhou, X.H., Luo, Y.Q., Li, B., Aurela, M., Billesbach, D., Blanken, P.D., Bracho, R., Chen, J.Q.,
Fischer, M., Fu, Y.L., Gu, L.H., Han, S.J., He, Y.T., Kolb, T., Li, Y.N., Nagy, Z., Niu, S.L., Oechel, W.C.,
Pinter, K., Shi, P.L., Suyker, A., Torn, M., Varlagin, A., Wang, H.M., Yan, J.H., Yu, G.R., and Zhang, J.H.
2016. Direct and indirect effects of climatic variations on the interannual variability in net ecosystem
exchange across terrestrial ecosystems. Tellus B 68(0). doi:10.3402/tellusb.v68.30575.
Song, Y., Chen, Q.Q., Ci, D., Shao, X.N., and Zhang, D.Q. 2014. Effects of high temperature on
photosynthesis and related gene expression in poplar. BMC Plant Biol. 14(1): 111.
doi:10.1186/1471-2229-14-111.
Stoy, P.C., Mauder, M., Foken, T., Marcolla, B., Boegh, E., Ibrom, A., Arain, M.A., Arneth, A., Aurela, M.,
Bernhofer, C., Cescatti, A., Dellwik, E., Duce, P., Gianelle, D., van Gorsel, E., Kiely, G., Knohl, A.,
Margolis, H., Mccaughey, H., Merbold, L., Montagnani, L., Papale, D., Reichstein, M., Saunders, M.,
Serrano-Ortiz, P., Sottocornola, M., Spano, D., Vaccari, F., and Varlagin, A. 2013. A data-driven analysis
Page 21 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
22
of energy balance closure across FLUXNET research sites: The role of landscape scale heterogeneity.
Agric. For. Meteorol. 171–172: 137–152. doi:10.1016/j.agrformet.2012.11.004.
Streets, D.G., Wu, Y., and Chin, M. 2006. Two-decadal aerosol trends as a likely explanation of the global
dimming/brightening transition. Geophys. Res. Lett. 33(15). doi:10.1029/2006GL026471.
Su, H., Li, Y.G., Liu, W., Xu, H., and Sun, O.J. 2014. Changes in water use with growth in Ulmus pumila in
semiarid sandy land of northern China. Trees-Struct. Funct. 28(1): 41–52.
doi:10.1007/s00468-013-0928-3.
Tian, H.Q., Melillo, J., Lu, C.Q., Kicklighter, D., Liu, M.L., Ren, W., Xu, X.F., Chen, G.S., Zhang, C., Pan,
S.F., Liu, J.Y., and Running, S. 2011. China’s terrestrial carbon balance: Contributions from multiple
global change factors. Global Biogeochem. Cycles 25(1). doi:10.1029/2010GB003838.
Wang, R., Balkanski, Y., Boucher, O., Ciais, P., Peñuelas, J., and Tao, S. 2014. Significant contribution of
combustion-related emissions to the atmospheric phosphorus budget. Nat. Geosci. 8(1): 48–54.
doi:10.1038/ngeo2324.
Wang, Y.W., and Yang, Y.H. 2014. China’s dimming and brightening: Evidence, causes and hydrological
implications. Ann. Geophys. 32(1): 41–55. doi:10.5194/angeo-32-41-2014.
Webb, E.K. 1982. On the correction of flux measurements for effects of heat and water vapour transfer.
Boundary-Layer Meteorol. 23: 251-254. doi:10.1007/BF00123301.
Yamasoe, M., von Randow, C., Manzi, A., Schafer, J., Eck, T., and Holben, B. 2005. Effect of smoke on the
transmissivity of photosynthetically active radiation inside the canopy. Atmos. Chem. Phys. 5(4),
pp.5909-5934. doi:10.5194/acpd-5-5909-2005.
Yuan, W.P., Luo, Y.Q., Richardson, A.D., Oren, R., Luyssaert, S., Janssens, I.A., Ceulemans, R., Zhou, X.H.,
Grünwald, T., Aubinet, M., Berhofer, C., Baldocchi, D.D., Chen, J.Q., Dunn, A.L., Deforest, J.L., Dragoni,
D., Goldstein, A.H., Moors, E., Munger, J.W., Monson, R.K., Suyker, A.E., Starr, G., Scott, R.L.,
Tenhunen, J., Verma, S.B., Vesala, T., and Wofsy, S.T.E. 2009. Latitudinal patterns of magnitude and
interannual variability in net ecosystem exchange regulated by biological and environmental variables.
Glob. Chang. Biol. 15(12): 2905–2920. doi:10.1111/j.1365-2486.2009.01870.x.
Zhang, B.C., Cao, J.J., Bai, Y.F., Yang, S.J., Hu, L., and Ning, Z.G. 2011. Effects of cloudiness on carbon
dioxide exchange over an irrigated maize cropland in northwestern China. Biogeosciences Discuss. 8(1):
1669–1691. doi:10.5194/bgd-8-1669-2011.
Zhang, M., Yu, G.R., Zhang, L.M., Sun, X.M., Wen, X.F., Han, S.J., and Yan, J.H. 2010. Impact of cloudiness
on net ecosystem exchange of carbon dioxide in different types of forest ecosystems in China.
Biogeosciences 7(2): 711–722. doi:10.5194/bg-7-711-2010.
Zhou, J., Zhang, Z.Z., Sun, G., Fang, X.R., Zha, T., McNulty, S., Chen, J.Q., Jin, Y., and Noormets, A. 2013.
Response of ecosystem carbon fluxes to drought events in a poplar plantation in Northern China. For. Ecol.
Manage. 300: 33–42. Elsevier B.V. doi:10.1016/j.foreco.2013.01.007.
Page 22 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 1. The change of the clearness index under clear skies (CIt) with the sin of solar elevation (sinβ) from June to August in 2015.
Fig. 1 127x82mm (300 x 300 DPI)
Page 23 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 2. Variation of environmental factors during the mid-growing season in 2014 and 2015. Daily precipitation (P) and relative extractable water (REW) (a/b); mean daily clearness index (CI) and daily total
global solar radiation (G) (c/d); mean daily air temperature (Ta) and vapor pressure deficit (VPD) (e/f).
Fig. 2 143x187mm (300 x 300 DPI)
Page 24 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 3. Changes of vapor pressure deficit (VPD) (a), air temperature (Ta) (b), photosynthetically active radiation (PAR) (c) and diffuse photosynthetically active radiation (PARdif) (d) with clearness index (CI) for
selected intervals of solar elevation during June to August in 2014 and 2015.
Fig. 3 169x145mm (300 x 300 DPI)
Page 25 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 4. Relationship between gross primary productivity (GPP) and the clearness index (CI) at different selected solar elevations from June to August in 2014 and 2015.
Fig. 4
169x145mm (300 x 300 DPI)
Page 26 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 5. Relationship between light use efficiency (LUE) and the clearness index (CI) at different selected solar elevation from June to August in 2014 and 2015.
Fig. 5
169x145mm (300 x 300 DPI)
Page 27 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 6. Ecosystem photosynthesis-radiation relationship of the plantation under different sky conditions from June to August in 2014 and 2015.
Fig. 6 127x84mm (300 x 300 DPI)
Page 28 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 7. The growth rate of canopy photosynthetic potential capability (Pmax) (µmol m-2 s-1) under cloudy sky conditions versus clear sky conditions. Numbers above bars show the growth percentage of Pmax by
cloudy skies compared to the clear skies under different environmental classifications. Fig. 7
127x79mm (300 x 300 DPI)
Page 29 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research
Draft
Fig. 8. The direct and indirect effects of environmental factors on ecosystem photosynthetic productivity (GPP) for different levels of vapor pressure deficit (VPD) (a/b), air temperature (Ta) (c/d) and relative
extractable water (REW) (e/f) under cloudy skies during the study period. The values beside the paths are
standardized path coefficients (ρ: -1-1), ‘ρ < 0’ denotes negative correlation and ‘ρ > 0’ denotes positive correlation.
Fig. 8 170x219mm (300 x 300 DPI)
Page 30 of 30
https://mc06.manuscriptcentral.com/cjfr-pubs
Canadian Journal of Forest Research