Results
Precipitation and water
table
The total rainfall from April to October in 2019 and 2020 was 160.8 and
179.2mm, respectively. During the 2 years, the number of days with
rainfall larger than 1 mm was 26, and 24 days, and the maximum daily
rainfall was 25.7, and 19.9 mm (Fig. 2a). The maximum monthly rain was
in July and August, which accounts for 55% and 45% of annual
precipitation. In general, the rainfall was characterized by small rain
and moderate rain.
During the measurement period,
water
table depth (WTD) ranged from 233 to 436cm (Fig. 2b). The minimum depth
to water occurred in May 2020, whereas, the maximum occurred in July
2019. In general, during the long dry period between the beginning of
winter and early summer, WTD continuously increased to approximate 434
cm. In the rainy season from May to July, WTD decreased due to the
irrigation or precipitation. The variation of WTD was primarily a
response to irrigation and local ET. The highest peak water level was
observed after the irrigation event of 2 July 2019. In addition, the
maximum increase (118 cm) of WTD appeared in 8 November 2019 due to the
irrigation water percolation to groundwater. This is illustrated by data
for the irrigation events of 2 July and 8 November 2019. WTD was 348 cm
before the 2 July irrigation event and 428 cm before the 8 November
irrigation event (Fig. 2b). The minimum WTD reached 234 cm and 310 cm,
respectively. After irrigation, the water table fell fast to the level
at the beginning of irrigation and then slowly declined because of
evaporation and plant water uptake.
Soil water content (SWC)
The mean, standard deviation (SD), and coefficients of variation (CV)
values of the soil water content at different depth were used to
represent the spatial-temporal variability in soil moisture (Table 2).
During the observation period, the mean SWC ranged from 0.247 to 0.285
for the 0-50cm soil layer and 0.275 to 0.371 for the 50-270cm soil
layer, respectively. The mean SWC increased with soil depth, reaching a
maximum at 50 cm. After an increase of water content with depth down
from surface to 50 cm, another increase with depth could be also
observed at about 50–270 cm depth. High soil moisture SD value (0.050)
and CV value (0.203) were mainly detected in the shallow layers
(0–20 cm), and these values decreased with increased soil depth.
Although both variables decreased along the profile, the decrease in CV
was much quicker than that in SD, from 0.203 to 0.014 and 0.050 to
0.005, respectively.
The statistical data showed that the SWC varied significantly at depth
of 20-50cm, but not at the deeper soil layer (150-270cm) during the
observation period. Fig. 3(a,b) shows the dynamic changes of SWCs with
time at different depths. The SWC had apparent seasonal changes and was
fluctuant with precipitation and irrigation processes. According to
these processes, the whole observation period could be divided into
three stages: change period of gains and losses (from April to
September), relatively stable period (from October to mid-November),
period of slow decline after autumn irrigation (from mid-November to
March) at 20,30 and 50cm.
The first period was an obvious gain and loss period, in which the SWC
increased greatly after precipitation or irrigation and decreased slowly
because of percolation and evapotranspiration. The total rainfall was
59.2 mm from 25 to 27 June, 2019 and 39.8 mm on from 29 to 30 August,
2020. The SWC at 30 cm increased to approximately 0.358 and 0.341,
respectively. The SWC sharply increased at depth of 20-50cm, from about
0.196 to 0.376 after irrigation on July 14, 2019 and from about 0.146 to
0.364 after irrigation on July 6, 2020. However, the SWC recession at
soil depths above 70 cm in 2020 was higher than that in 2019. This was
mainly due to stronger evaporation and water use in 2020, when the plot
1 was cropped with maize. The SWC fluctuated with a variation of 0.55
from October to mid-November due to little precipitation and
non-irrigation, which could be considered a relatively stable period.
Lastly, the SWC decreased gradually from mid-November after autumn
irrigation to March of the next year, which could be considered a losing
period.
Soil water potential(SWP)
Fig. 3(c,d) shows the SWP dynamics at different depths during the
observed period in 2019 and 2020. The mean values of SWP at eight
different depths of soil layers were between -115.7 and -219.5 cm
H2O in 2019 and between -151.1 and -236.2 cm
H2O in 2020. The maximum value was -23.7 cm
H2O at 20 cm layer in 2019 and -21.4 cm
H2O at 20 cm layer in 2020. The minimum value was -269.8
cm H2O at 270 cm layer in 2019 and -438.5 cm
H2O at 70 cm layer in 2020. This difference was probably
caused by the maize of which roots mainly extracted water from the 70cm
soil. The SWPs at 20, 30 and 50 cm ranged from -23.7 to -222.15 cm
H2O in 2019 and from -21.3 to 402.1 cm
H2O in 2020, with greater oscillation due to the
evaporation and irrigation between May and September. In October,
because no irrigation was performed, the SWPs at 20-50 cm decreased with
time. The SWPs at 200 and 270 cm, close to the groundwater capillary
zone, were affected by the water table fluctuation. With increasing
water table during irrigation periods, the SMPs at 270 cm decreased with
time and were even close to 0.
The statistics of SWP gradients are presented in Table 3. The SWP
gradients were almost negative in 2019. The soil water displayed an
upward trend of movement with the mostly positive SWP gradients at
20-30cm, 70-100cm and 150-200cm depth ranges in 2020. When evaporation
continues after the precipitation or irrigation, the soil water in the
upper part moved upwards to the soil surface (KHALIL et al., 2003; Li et
al., 1990). As the SWP gradients decrease with increasing depth, at some
level it is no longer large enough to drive the soil water upward and
the flux is essentially zero. This level is known as the zero flux plane
(ZFP), below which the soil water at lower depth moved downwards towards
the water table due to gravity (Sadeghi et al. 1985). Fig 4 shows an
example of the SWCs and SWPs at
various depths before and after the precipitation or irrigation event in
2019 and 2020. The SWP profile of June 28, 2019 showed that ZFP was at
30cm (Fig. 4b). By June 30, the ZFP has passed below the 50 cm. After
the irrigation on August 4, 2019, the ZFP was at 20 cm depth on the
first day, then extended from 30 to 70 cm in the next four days, and
stabilized at 70 cm on the last three days (Fig. 4d). The ZFP was more
rapidly developed after irrigation on August 2020 than irrigation on
August 2019, which suggests that the roots take up more soil water (Fig.
4d, h).
Isotopic composition of
water
The δ18O and δD of precipitation, irrigation water,
soil water and groundwater are presented in Fig. 5 and table 4. In
total, seventeen precipitation samples were collected from April to
December in 2019 and 2020. The isotopic values of precipitation
were highly variable. In 2019, The
δD
(δ18O) in precipitation ranged from -63.5 (-10.26) to
-18.1 (-4.5) ‰, with an average value of -43.6 (-6.91) ‰ and standard
deviation (SD) of 15.75 (2.12) ‰. In 2020, the range of δD
(δ18O) was from -51.0 (-6.87) to -1.9 (-0.09) ‰, with
a mean value of -28.5 (-4.46) ‰ and SD of 17.7 (2.49) ‰ (Table 4).
Meanwhile, lower δD and δ18O were observed during
heavy and continuous (e.g., June 25-27, 2019) rainfall events,
indicating a strong rainout effect (Clark and Fritz, 1997; Liu et al.,
2011). The local meteoric water line (LMWL) was established as δD =
7.11δ18O + 5.14(R2 =0.91). The
slope of the LMWL was similar with that of
δD = 7.21δ18O + 5.50(R2 =0.96),
based on the monthly data from IAEA collected at Yinchuan station
between 1988 and 2000.
The irrigation water and groundwater
δD and δ18O varied
little in the range of −68.1 to −71.8‰, −9.11 to −10.37‰, and −72.6 to
−64.5‰, −10.55 to −8.66‰, respectively (Table 4). The δD and
δ18O of the two water bodies were very similar, which
implies that the groundwater was mainly recharged by irrigation water.
The range of δD values in soil water were between -75.0 and -24.1 ‰,
with a mean value of -61.4 ‰ and SD of 5.1 ‰, whereas
δ18O ranged from -11.4 to -4.63 ‰, with a mean value
of -8.74‰ and SD of 0.80 ‰ in 2019
(Table
4). The isotopic compositions at 20, 30 and 50cm were more variable.
The mean values of
δD
and δ18O were approximately -62.6‰ and −9.0‰. The
maximum and minimum values appeared at 20cm in June and in July,
respectively. This variation was mainly caused by kinetic fractionation
of evaporation and infiltration of precipitation or irrigation. The
values for the deeper soil layers were relatively stable with an average
of -60.5‰ for δD and -8.59 ‰ for
δ18O. Regarding the isotopic compositions of the soil
water samples in 2020, they ranged from −72.2 to −52.0‰ (mean −64.2‰)
for δD and from −10.44 to −6.91‰ (mean −8.83‰) for
δ18O, respectively. The isotopic compositions of soil
water at 20, 30 and 50cm were closer to that of irrigation water
compared to that at other depths. The isotopic compositions at 270cm
were more enriched relative to the stable isotopes of groundwater.
The evaporation line of soil water
(ELSW) equation was δ D = 4.27 δ 18O-25.19
(R2 = 0.41). The stable isotopes of soil water
were scattered and the slope was lower compared to the LMWL, indicating
that the δ D and δ 18O in soil water were
mainly influenced by mixing with precipitation or irrigation and
undergoing evaporation.