Variations in diurnal and seasonal net ecosystem carbon dioxide 1 exchange in a semiarid sandy grassland ecosystem in China ’ s Horqin

Sandy Land 3 Yayi Niu, Yuqiang Li*, Hanbo Yun, Xuyang Wang, Xiangwen Gong, Yulong 4 Duan, Jing Liu 5 a Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou 6 730000, China 7 b University of Chinese Academy of Sciences, Beijing 100049, China 8 c Naiman Desertification Research Station, Northwest Institute of Eco-Environment and Resources, 9 Chinese Academy of Sciences, Tongliao 028300, China 10 d State Key Laboratory of Frozen Soil Engineering, Northwest Institute of Eco-Environment and 11 Resources, Chinese Academy of Sciences, Lanzhou, Gansu 730000, China 12 e Center for Permafrost (CENPERM), Department of Geosciences and Natural Resource 13 Management, University of Copenhagen, DK-1350 Copenhagen, Denmark 14

our knowledge, there has been no report on the intra-annual and interannual variation 5 precipitation amounts can increase SWC in deeper levels of the soil and trigger 104 sequestration processes (Hao et al., 2010). To better understand the effects of 105 precipitation on NEE, we asked the following question: Is there a threshold of "effective 106 precipitation" that determines whether ecosystem carbon fluxes will lead to net 107 sequestration or net emission in sandy grasslands? 108 Precipitation is characterized by discrete events in arid and semiarid regions, with 109 high variability in the amount, duration, and frequency of precipitation at intra-annual 110 (e.g., seasonal) and inter-annual scales (Hao et al., 2010;Ponce Campos et al., 2013). 111 These discrete and largely unpredictable events may lead to pulsed availability of soil 112 water and nutrients, with both spatial and temporal variation (Zhao and Liu, 2011). The 113 response of photosynthesis and respiration to precipitation is seasonally specific 114 because of differences in the depth of soil water infiltration and because these processes 115 differ in their sensitivity to temperature (Li and Zhou, 2012). Spring and autumn . This is particularly true when relatively low temperatures limit soil microbial 120 respiration during certain periods (Knorr et al., 2005). Summer precipitation is thought 121 to primarily influence shallow soil moisture, thereby stimulating the activity of 122 shallowly rooted plants, whereas a combination of high temperatures and high soil 123 moisture stimulate the respiratory response by soil microbes (Sponseller, 2006). The associated with changes in precipitation patterns, and that affect soil water regimes, 133 may be critical for developing strategies to preserve or restore these sandy grasslands.

134
In this paper, we present the results from continuous (14 September 2014 to 31 135 December 2018) in situ monitoring of CO2 dynamics in the Horqin Sandy Land's sandy 136 grassland using the eddy covariance technique, and quantify the temporal variation of 137 NEE and the factors that control it. We had the following goals: (1) To quantify the 138 annual, seasonal, and diurnal variation in NEE, GPP, and Rec. We hypothesized that the 139 sandy grassland is a carbon source at the ecosystem scale, because the sandy grassland   2007). Thus, the grassland had been recovering naturally for nearly 30 years when our 160 study began. At an elevation of 377 m a.s.l., the study area has a continental semiarid 161 7 monsoon temperate climate regime. The mean annual temperature is 6.8 °C, with mean 162 monthly temperatures ranging from -9.63 °C in January to 24.58 °C in July. Average 163 annual precipitation is approximately 360 mm, with 70 % of the precipitation occurring 164 during the growing season, between June and August. Annual mean potential 165 evaporation is approximately 1973 mm. The annual frost-free period is 130 to 150 days. 166 The most common soil type in the study region is a sandy chestnut soil, but most of the 167 soil has been degraded by a combination of climate change and anthropogenic activity 168 (unsustainable grazing or agriculture) into an aeolian sandy soil under the action of 169 wind erosion (Zhao et al., 2007), with coarse sand, fine sand, and clay-silt contents of

192
We cleaned the mirror of the LI-7500 every 15 days to maintain the automatic gain 193 control value below its threshold (55 to 65). All of the instruments were powered by 194 solar panels connected to a battery.     217 We used the EddyPro 6.2.0 software (Li-Cor, Lincoln, NE, USA) to process the 10-

218
Hz raw eddy covariance data. Processing included spike removal, lag correction, 219 secondary coordinate rotation, Webb-Pearman-Leuning correction, and sonic virtual 220 temperature conversion (Webb et al., 1980). We used the data processing method of Lee  We used several strategies to compensate for missing data. We used linear 232 interpolation to fill gaps that were shorter than 2 h. For longer gaps, we handled the gap (1) 238 R0 is the base ecosystem respiration rate when the soil temperature is 0 ℃, b is an 239 empirically determined coefficient, and T10 is the soil temperature at a depth of 10 cm.

240
Daytime ecosystem respiration can be estimated by extrapolation from the 241 parameterization derived from Eq. (1). We did not attempt to fill in gaps longer than 7 242 days, and treated those gaps as missing data. Gross primary productivity (GPP) was 243 obtained as follows: We used the standard sign convention for NEE, with NEE > 0 indicating a net loss 246 of CO2 to the atmosphere (source) and NEE < 0 indicating net CO2 uptake by the 247 ecosystem (sink).  252 We performed correlation analysis (Pearson's r) and regression analysis using the 253 SPSS software. Unless otherwise noted, we defined statistical significance at p < 0.05.

254
Pearson's r was applied to confirm the strength of the relationships between parameters.

255
Before regression analysis, we tested for collinearity (using a variance inflation factor 256 of 0 < VIF < 10) using the Kaiser-   (Table   273 1). Zhao and Liu (2010) showed that precipitation less than 5 mm in arid and semiarid 274 areas changes SWC primarily in the near-surface soil, and that precipitation events 275 greater than 5 mm can effectively supplement root layer moisture at greater depths; it 276 is therefore called "effective precipitation". Our result was consistent with this view 277 11 (Fig. 2). The essence of effective precipitation is that precipitation enters the soil below 278 the surface layer, and becomes part of the soil water; that soil water is used either

323
The diurnal cycles of NEE and GPP were also characterized by a single peak, and the 324 ecosystem CO2 uptake reached its peak from around 10:30 to 12:00 (Fig. 5b). The

333
In winter, the grassland ecosystem functioned as a net CO2 source in all years, with 334 an average seasonal NEE of 0.59 ± 0.02 g C m −2 d −1 (Fig. 4d). It should also be noted  Rec) were relatively stable in winter ( Fig. S4 and Fig. 4d). We have therefore focused 352 on the relationships between NEE, its components, and the associated environmental 353 factors in the other three seasons (Fig. 7-8). In the spring, the total monthly precipitation

409
In spring, the sandy grassland was a net CO2 source in all years (Fig. 4a). Before the ). Therefore, the ecosystem was a net CO2 source.

415
In summer, the sandy grassland was a CO2 sink in all years (Fig. 4b). Our results At the diurnal scale, NEE in the spring and summer showed CO2 uptake during the 428 day (06:00-18:00), and CO2 emission during the night (Fig. 5a, b). NEE decreased with 429 increasing light intensity during the day, reached its peak value around noon, then

432
In autumn and winter, the sandy grassland ecosystem showed CO2 emission 433 throughout the day (Fig. 5c, d). At a diurnal scale, the ecosystem showed carbon  with these studies, as the slope of the regression line that relates precipitation to GPP 460 (0.98) was much higher than that for Rec (0.51) (Fig. 6). However, we must improve 461 our understanding of the responses of the ecosystem to precipitation and the underlying 462 mechanisms that control whether it will be a carbon source or sink. To accomplish this, 463 it will be necessary to observe the ecosystem continuously for a longer period of time.

464
The dominant factors varied seasonally. In the spring, NEE was most strongly 465 affected by Tsoil (Fig. 8), SWC (Fig. 8), and the amount of precipitation (Fig. 7). After  (Table 1), and the carbon uptake in 2018 was higher than that in 2015 482 and 2016 (Fig. 4b). Effective precipitation may penetrate deeper into the soil, thereby relationship between NEE and SWC in deeper soil layers was negative (Fig. 8c), which 503 was similar to the relationship in summer.

504
In winter, the annual plants had withered, so there was no GPP and the entire  (Table S1). Previous studies found that when SWC decreases sufficiently to create 508 water stress, it may replace temperature as the main factor that controls soil respiration