Journal of Hydrology 529 (2015) 1901–1908

Contents lists available at ScienceDirect

Journal of Hydrology
journal homepage: www.elsevier.com/locate/jhydrol

Paddy rice ecohydrology pattern and influence on nitrogen dynamics

in middle-to-high latitude area
Wei Ouyang a,⇑, Siyang Chen b,a, Xuelei Wang c, Yushu Shan d,a
a
School of Environment, State Key Laboratory of Water Environment Simulation, Beijing Normal University, Beijing 100875, China
b
Marine Monitoring and Forecasting Center of Zhejiang, Hangzhou 310007, China
c
Satellite Environment Center, MEP, Beijing 100094, China
d
Taihu Basin Authority, Ministry of Water Resources, Shanghai 200434, China

a r t i c l e i n f o s u m m a r y

Article history: Paddy rice in middle-to-high latitude areas has a longer growing period with lower temperature.
Received 6 September 2014 Therefore, it was hypothesized to have a different soil ecohydrology pattern and nitrogen (N) efficiency.
Received in revised form 30 April 2015 Based on daily monitoring of soil moisture and N concentrations in soil water at four depth layers, the
Accepted 7 August 2015
characteristics and their ecohydrological interactions, soil environmental indexes and N dynamics were
Available online 13 August 2015
This manuscript was handled by Geoff
analyzed. The temporal–vertical observations of soil moisture and soil water acquisition rates demon-
Syme, Editor-in-Chief strated the existence of an impermeable soil layer at 30 cm, which had lower soil organic carbon content.
The soil moisture greater in the upper layer had higher soil organic carbon content, which provided the
Keywords:
critical conditions for the rice tillage. Lower than the 60 cm depth, the soil moisture increased and had a
Soil water peak acquisition rate of approximately 0.0075 cm3 s 1. The concentration of NO3 -N in the water at the
Paddy rice nitrogen soil subsurface was approximately 2.4 times than that of NH+4-N. The NO3 -N had a larger concentration
Ecohydrology in cold area at the deeper layer as a result of downing leaching, but the NH+4-N had a greater concentration in the tops
Soil microclimate soil layer than the other depths. The soils under the influence of freeze–thawing had the largest TN con-
Middle-to-high latitude area centration in surface water at the end of April, which was about three times than that of the peak during
the growing period. This study aligned the interactions between the N dynamics with soil microclimatic
factors such as temperature, water pH and moisture across the 90 cm profile. The temporal–vertical pat-
tern of soil moisture and the N in soil water provided evidence for the hypothesis that the soil ecohydrol-
ogy dynamics in this area. These findings were of particular significance to understanding the impacts of
paddy rice ecohydrology with N dynamics in middle-to-high latitude areas, which can help to optimize
the N availability and water efficiency.
Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction
Paddy rice cultivation in the middle-to-high latitude regions of
China occurs in the most northern latitude paddy rice area the
world. This cultivation is based on the long interactions of water
and nutrients under cold conditions (Chen et al., 2013). Nitrogen
(N) is a major yield-limiting nutrient factor of crop production in
the agro-ecosystem (Möller et al., 2006). Nitrogen loss is also the
main environmental issue for long term agricultural activities,
which is critical for environmental quality (Arauzo et al., 2011).
The impacts of irrigation, fertilizer, and crop rotation on N leaching
have generally been explored from the perspective of agricultural
management (Fuller et al., 2010; Gheysari et al., 2009). Previous
⇑ Corresponding author. Tel.: +86 10 58804585.
E-mail address: wo43@cornell.edu (W. Ouyang).
studies found that the soil ecohydrology feature is the most impor-
tant variable for the N transportation (McIsaac and Libra, 2003;
Shrestha et al., 2007). It has been hypothesized that the paddy rice
cultivation had different ecohydrology dynamics under the cold
conditions and had special impacts on the N movement. In this
study, the temporal–vertical dynamics of soil water and their
impacts on N were observed and analyzed.
The rice cultivated in the middle-to-high latitude ranges
requires special practices that impact the paddy land ecohydrology
and the N cycle in the rice fields (Singh et al., 2006). The soil mois-
ture and N availability in soil are controlled by patterned tillage
practices and fertilization application principles, which also impact
the relationship between the soil microclimate and N (Zvomuya
et al., 2006; Joseph and Henry, 2008). The ecohydrology and inter-
actions with N in middle-to-high latitude paddy rice are still not
thoroughly understood and have not been well documented by

http://dx.doi.org/10.1016/j.jhydrol.2015.08.007
0022-1694/Ó 2015 Elsevier B.V. All rights reserved.
1902 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908

field research (Hao et al., 2013). The ecohydrology in rice fields is
characterized by ponding under saturation conditions during the
growth period, but is sometimes unsaturated under intermittent
submerged conditions (Boling et al., 2008). Under this condition,
the amount of N and soil water in each layer of the soil profile
below the arable layer is an important part of soil N storage.
Vertical flows are more effective than horizontal flow systems
because of their high oxygen transfer rates and longer ponding
times (Cooper, 2005).
The N dynamics in paddy rice depend on the varieties of
edaphic, tillage, climatic, and agronomic factors. The soil proper-
ties, such as texture and organic matter content, are the priority
factors that affect the water-holding capacity of soils (Hopmans,
2003; Sleutel et al., 2008). The greater leaching due to higher soil
moisture and the presence of preferential flow pathways may
decrease N availability (Shipitalo et al., 2000). The N transforma-
tion is highly sensitive to soil ecohydrology variations, which is a
controlling variable that affects the redox potential of soil (Yu
and Ehrenfeld, 2009). Furthermore, the paddy land ecohydrology
can include water storage and availability levels, which are the
necessary for rice growth. The temperature sensitivity of net nitri-
fication and N mineralization also vary with soil hydrological sta-
tus (Craine and Gelderman, 2011) and long-term studies show
that N losses by leaching typically increase with higher soil mois-
ture conditions (Bryant et al., 2011).
Understanding the interactions of soil ecohydrology and nitro-
gen (N) is important to improve the efficiency of water use and
N in the farmland, which is more critical in the middle-to-high lat-
itude agricultural systems (Fuentes et al., 2003). Paddy land ecohy-
drology dynamics depend on a combination of micro-variables,
including soil texture, precipitation, evaporation, and crop transpi-
ration (Hupet and Vanclooster, 2005). During N transport in soil
water, dissolved organic N is recognized as the main form in soil
surface water (SW) and leaching in soil subsurface water (SSW)
(Fellman et al., 2008). In addition to serving as transport media
for N transport, soil water also influences the denitrification pro-
cess and the availability of N to plants (Hagedorn et al., 2001).
The paddy rice in this middle-to-high latitude area was
reclaimed from wetlands during agricultural exploitation, and
how to minimize the environmental impact with larger yields is
the priority concern (Hao et al., 2012). In this study, the quantity
of soil water and N dynamics (TN, NO3 -N, and NH+4-N) in SW and
SSW at different depths was monitored daily. The impacts of soil
temperature, pH, and soil ecohydrology on soil water N dynamics
were also identified. In brief, the objectives of the study were to:
(1) discriminate the paddy rice field ecohydrology characteristics
at four soil depths in middle-to-high latitude area, (2) identify
the interactions between the soil ecohydrology and the transporta-
tion of TN, NO3 -N, and NH+4-N at four soil layers, and (3) quantify
the impacts of ponding and soil microclimate on four forms of N
concentrations in the 0–90 cm soil profile.
2. Materials and methods
2.1. Research site description
Field investigations provide the most direct and effective means
for evaluating soil water dynamics and N transportation (Bagarello
et al., 2010; Jalali, 2005). Field observations were conducted on a
farm in a middle-high latitude area, which has latitude more than
45°. The experiment station in Paddy #1 of the Bawujiu Farm
adjoins the Ussuri River in northeast of China, (Fig. 1). The field
was irrigated by well water, where the well location is approxi-
mately 10 m north of the field. The irrigation and drainage canals
embayed the field on two sides with a total length of 50 m. The
topography of the field is defined by a 0–1% slope in east-west
and north-south direction. The regional climate is cold temperate
continental monsoon with a mean annual precipitation (1983–
2010) of 588 mm and a mean annual air temperature (1983–
2010) of 2.94 °C. The monthly average air temperature is below
0 °C for approximately 6 months during the year (Ouyang et al.,
2013).
2.2. Soil characterization and farming practice
The paddy rice growing area in this region has a special vertical
profile and there is an impermeable albic layer at around 30 cm
depth, which is the necessary condition for rice growth (Hao
et al., 2013). The soil texture and chemical characteristics, such
as soil bulk density, pH, organic matter, TN, and organic carbon
of the 90 cm profile is summarized (Table 1). The planting and har-
vesting dates, detailed irrigation events, and tillage and fertiliza-
tion practices are presented in Table 2. Irrigation amounts from
ponding to harvesting in the rice growing season ranged from 20
to 60 mm. The field received N, P, and K from chemical fertilizer
of urea (N 46%), diammonium phosphate (N 18%, P2O5 46%), and
potassium sulfate (K2O 33%).

Well Soil water sampler

Soil moisture sensor
Soil temperature sensor
Paddy #1
DC 15 cm
30 cm
Paddy #2
IC
60 cm
Paddy #3
90 cm
Experimental site
IC: irrigation canal
DC: drainage canal

Fig. 1. Location of the experimental field and associated sensors and samplers.
 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908 1903

Table 1
Soil physical–chemical properties in four layers of tested paddy rice field.
1
Soil depth Soil Organic TN (g kg ) Organic C pH
1
(cm) texture matter (g kg ) (g kg 1)
0–15 Loam 30.3 1.58 17.8 5.50
15–30 Loam 26.5 1.42 16.9 5.98
30–60 Clay 12.0 0.69 7.2 6.10
60–90 Clay 10.2 0.74 5.9 6.57
2.3. Soil water monitoring and laboratory measurements
To evaluate the performance of the irrigation amount, a moni-
toring system (Intelligent Electromagnetic Flowmeter, LDG-150S-
M2X100, Tianjian, China) was employed. To monitor the soil water
N, four vertical samplers (a suction cup: Teflon and quartz powder;
OD 21 mm  L 95 mm; aperture, 2 lm; conductibility,
3.31  10 6 mm s 1, PRENART, Denmark) were installed in a plot
at a 45° angle at depths of 15, 30, 60, and 90 cm. The porous cup
was surrounded with fine quartz sand so that the biggest pores
in the collector were filled with fine soil or silica flour to ensure
a tight capillary contact with the soil. For sampling, the PVC pipe
was separately connected to a PRENART collecting bottle with a
screw cap, and a 0.05 MPa vacuum was applied with aid of a por-
table vacuum pump. Soil water samples were collected at least
once a week and three samples were collected continuously each
time. The precipitation, irrigation water, and recession flow sam-
ples were collected with triplicate samples after each precipitation,
irrigation, or recession event. After each of these events, soil water
samples (15, 30, 60, and 90 cm) were collected at intervals of 1, 2,
3, and 5 days. During the ponding period, water samples were
taken daily. The soil water sample was sealed in plastic bottles.
The water volume, elapsed time, pH, irrigation amount, and
ponding level were determined immediately. The irrigation
amount was calculated by the velocity and duration of irrigation.
The acquisition rate of soil water at a given depth was defined as
the water volume divided by the elapsed time. Water samples
were immediately taken to the lab and stored at 4 °C in an icebox
until N analyses were performed. Water samples were analyzed for
total nitrogen (TN), nitrate (NO3 -N), and ammonia (NH+4-N). The TN
was analyzed by UV spectrophotometer after alkaline potassium
persulfate digestion. The NO3 -N was determined colorimetrically
using the UV spectrophotometric method. The NH+4-N was deter-
mined by salicylic acid spectrophotometry (Caliskan et al., 2008).
2.4. Meteorological monitoring
The meteorological data were recorded on-site by the ZENO
Meteorological Station. Automated soil temperature (Thermistor,
Coastal, Seattle, USA) and moisture (TDR type, Coastal, Seattle,
USA) sensors with data logger were inserted flatly into the undis-
turbed soil layers at the corresponding depth of soil samples. The
set-up of soil temperature and moisture sensors was replicated
at a distance of 0.5 m. The vertical samplers and sensors were
inserted into the soil column and left to equilibrate for 24 h
(Chen et al., 2013).
Bulk
density
(g cm 3) 2.5. Data analysis procedure
1.11
1.24 Statistical analyses were performed using Sigmaplot 10.0
1.58 (Systat Software Inc., San Jose, CA, USA) and SPSS 16.0 (SPSS Inc.,
1.64
Chicago, IL, USA) software packages. Curve regression (exponen-
tial) analysis was employed to describe relationships between TN
concentrations in SW during the ponding period. Pearson correla-
tion coefficients were computed to analyze the relationships
between the soil water acquisition rates with precipitation and
the concentrations of each N form with soil temperature at differ-
ent depths. The MANOVA was employed to analyze the impacts of
soil microclimate factors on N dynamics in SW and SSW. As is nec-
essary for MANOVA, the air temperatures were divided into five
ranges (3–6 °C, 6–10 °C, 10–15 °C, 15–20 °C, and >20 °C); the soil
temperatures were divided into five ranges (3–4 °C, 4–8 °C, 8–
15 °C, 15–20 °C, and > 20 °C); the ponding depths were divided
into five levels (0–10, 10–20, 20–30, 30–40, and 40–50 mm). The
value of soil moisture, pH, and N concentrations at each depth
were divided into five levels.
3. Results
3.1. Temporal–vertical pattern of soil moisture during thawing period
Daily monitoring data was used to analyze the temporal–verti-
cal dynamics of the soil moisture, which observed the similar pat-
terns at the four layers (Fig. 2). The soil moisture increased from
0.20 to 0.45 cm3 cm 3 in April because of the thawing of frozen
soil. Following the early growing period after the rice planting,
the moisture changed slightly from 0.38 to 0.42 cm3 cm 3 and fluc-
tuated noticeably during the pre-harvest period. At the depth of
30 cm, the water storage decreased to 0.35 cm3 cm 3 and then
increased to 45 cm3 cm 3 after a heavy rain. Furthermore, the soil
at the 90 cm began with more water storage than soils at the other
three depths. The water storage reached 0.48 cm3 cm 3, which was
approximately 14% more than that of the 15 cm soil, possibly
because of the leaching of water through the soil profile or the
thawing of frozen soil.
Precipitation and irrigation were the main water sources for the
rice field, and accumulated for the vertical distribution The precip-
itation event on 13 August resulted in the relatively high soil mois-
ture on the soil profile after two (15 cm, 0.43 cm3 cm 3) and seven
days (30 cm, 0.42 cm3 cm 3; 60 cm, 0.43 cm3 cm 3). The ponding
level responded sensitively to the precipitation and irrigation
events (p < 0.01) mainly because of the drainage required for trans-
planting. The precipitation events resulted in a 20 mm increase in
the ponding level. Moreover, the air temperature increased stea-
dily and reached 26.5 °C in late July. Followed by the dry soil sur-
face, this period, referred to as the booting stage, was the prime
time for rice plant growth.

Table 2
Tillage practices in paddy rice field in middle-to-high latitude area.
1
Crop Crop year Planting date Harvesting date Tillage prior to planting ponding Main fertilization (kg ha ) Fertilizationa (kg ha 1
)
N P K
Rice 2011 15/5 5/10 Rotary tilling 588.00 90.30 55.20 52.62
Irrigation (mm)
Date (D/M) 20/4 30/4 2/5 24/5 29/5 28/6 26/7
Amount (mm) 60 40 30 30 20 30 40
a
Based on computational the N, phosphorus, and potassium elements; All the dates are day/month.
1904 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908

Drainage Dry period

0.15 0.25 0.35 0.45

Soil moisture (cm3 cm-3)

Fig. 2. Daily precipitation, 2-m air temperature, irrigation, ponding level and soil moisture pattern in paddy rice.

3.2. Soil water acquisition rate in paddy rice
The soil water-extractable efficiency and pH at each layer were
analyzed with vertical samplers. The acquisition rate of soil water
at 60 cm was higher than at shallower depths, with a peak value of
0.0075 cm3 s 1 (Fig. 3). During the rice growing season, the soil
water acquisition rate at the various depths fluctuated. There
was little variation in the acquisition rates among the depths at
15, 30, and 90 cm, with slight fluctuations in the early season
and a large fluctuation in the late season. For the 60 cm layer,
the acquisition rate decreased sharply in early August after a slight
fluctuation stage. Table 3 shows that there were significant fluctu-
ations in the acquisition rate at the 15 cm deep layer compared
0.008
0.007
Acquisition rate of soil water (cm 3 s-1)
with those in the other layers; the coefficient of variance (CV) value
reached 0.52. The acquisition rate of the soil water and precipita-
tion showed similar trends. The correlation analysis between these
factors indicated that the acquisition rate at 60 cm was mostly
associated with precipitation (P < 0.050). Overall, the acquisition
rate decreased in late season due to the smaller amount of precip-
itation and less irrigation at a time when temperatures were
higher.
Over the growing season, the average pH and standard devia-
tion of SW were 7.51 and 0.52, respectively. The pH increased from
7.1 to 7.7 when the soil was fertilized with a weak alkaline med-
ium. Then, the pH decreased quickly by 0.89 units and increased
smoothly for a long time. The pH did not change significantly
under sustained wetting in June and thereafter decreased to 6.3
in the middle of August under dry and high-temperature condi-
tions. The average pH of the SSW was 7.60, and the pH at the
15 cm 15 cm depth was the highest, followed by the values at the 30,
30 cm

90, and 60 cm depths. The pH of SSW at each soil depth decreased

60 cm in the middle of August when the pH in SW decreased (see Fig. 4).
0.006
90 cm
0.005
3.3. Temporal–vertical pattern of N in soil subsurface water
0.004
The temporal–vertical variations of each N form in the paddy
0.003 rice soil profile are shown in Fig. 5. The TN concentration in each
0.002
Table 3
Statistical analysis of acquisition rates on soil water profile.
0.001
Items Depth (cm)
0.000
15 30 60 90
4-22 5-13 6-3 6-24 7-15 8-5 8-26 9-16
Mean 0.0016 0.0021 0.0048 0.0020
Date Std. error 0.0008 0.0008 0.0017 0.0005
Coefficient of variance 0.52 0.39 0.34 0.25
Fig. 3. Acquisition rate of soil water at different depths.
 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908
9.0
8.5
8.0
pH
7.5
7.0
6.5
SW 15 cm 30 cm 60 cm
6.0
4-22 5-13 6-3 6-24 7-15 8-5
Date
Fig. 4. SW and SSW pH (15, 30, 60, and 90 cm) during rice growing season.
layer reached a similar value late in the season, whereas the N con-
centrations at the 15 cm layer were higher earlier in the growing
season and lower as the rice matured in August and September.
These results suggested that a considerable amount of N was lost
during the growing season, which may be the consequence of crop
absorption, leach and microbial N transformation. The TN concen-
tration at the 15 cm depth decreased and became close to that at
the 30 and 60 cm depths. At the 30 cm layer, the TN concentration
was low at the beginning and increased to a high concentration
level relatively late. The seasonal variation of TN at 60 cm was sim-
ilar to the trend at 90 cm; in both cases the trend was relatively
smooth.
The variation of NO3 -N, was very sensitive to the water input.
After irrigation or precipitation, the concentration of NO3 -N at
the 15 cm layer increased, with peaks on May 30, July 2, and July
28, and then decreased immediately thereafter. In the middle of
July, when there was no precipitation and the temperature was
high, the lower NO3 -N decrease suggested that the root zone
was watered without oxygen input, thereby enhancing the micro-
bial N transformation. The TN in the upper and lower layers had
higher N concentrations than those of the two middle layers. In
the ponding condition, there were high NO3 -N and NH+4-N concen-
trations in the upper layer. The concentration of NH+4-N in the soil
water decreased as the depth increased, suggesting that anaerobic
ammonia oxidation (anammox) may occur in these anaerobic
conditions.
In contrast to NH+4-N, the NO3 -N was inclined to vary through-
out the profile. The NO3 -N concentration was highest at the 15 cm
layer, lowest at the 30 cm layer, and similar moderate values
occurred at the 60 and 90 cm layers. Furthermore, under the influ-
ence of the non-water input-cultivation in early July, the NO3 -N
concentration in soil water at the 15 cm depth increased from mid-
dle July with higher soil moisture due to the precipitation, reaching
2.0 4.0 6.0 8.0 (mg L-1)
Fig. 5. Temporal–spatial variations of N concentration in soil subsurface water of paddy rice fields.
1905
a level of 1.63 mg L 1, compared with 0.8 mg L 1 late in the season.
The considerable concentration decrease in the lower layers may
be due to upward soil water flow from the lower to upper layer.
The soil water continued to migrate upwards, accompanied by
NO3 -N until late July, when irrigation and precipitation occurred.
The correlations of the N concentrations in SSW at different soil
depths were also analyzed (Table 4). The NO3 -N and NH+4-N in the
90 cm
top two soil layers varied in a similar manner. In contrast to NO3 -
N, NH+4-N had the smallest standard (Std.) error and variance, sug-
8-26 9-16
gesting that it did not fluctuate significantly in the entire period.
Relative to NO3 -N, NH+4-N was only found in small concentrations,
fluctuating by approximately 0.5 mg L 1. The NO3 -N concentra-
tions were generally very small, and most of the depths showed
mean concentrations of approximately 0.8 mg L 1. Comparing the
results to the N input, its concentration in the SSW was relatively
low, suggesting that N was sustainably assimilated by the rice
plants or lost in other ways. In general, changes in TN and NO3 -
N were more pronounced than those in NH+4-N with fertilization.
The correlation analysis indicated that there was a slight correla-
tion of NO3 -N at 90 cm depth with the other three layers.
3.4. Soil water N dynamics in surface and recession water
The NO3 -N dynamics in the SW followed not the similar trends
to those for TN and NH+4-N, where the NO3 -N concentration kept
constant during the rice growing period (Fig. 6(a)). The largest
TN concentration in the SW appeared at the end of April, when
the rice was not get cultivated. Overlaying the TN concentration
with the soil temperature, it was found that peak value in SW
was a consequence of soil thawing. The TN reached its peak con-
centration quickly after urea was applied and then decreased
quickly during the growth period. The downward trend of TN fit
the exponential equation (Y = C0  e kt, R2 = 0.971). The NH+4-N
concentration peaked 1–2 days after urea was applied and
decreased soon afterward as NO3 -N increased slightly, which also
leveled off after another 3–4 days. The NO3 -N typically changed
smoothly, with a low concentration of approximately
0.97 mg L 1. However, the results showed that N concentrations
of 28% and 62% were found in SW for NO3 -N/TN and NH+4-N/TN,
respectively. In order to transplant the rice plants from greenhouse
to the field, the SW was drained on 15 May. The recession flow was
investigated over five consecutive days (Fig. 6(b)). The concentra-
tions of TN, NO3 -N, and NH+4-N in the recession water gradually
decreased, with decreases of 22%, 15%, and 50%, respectively, from
the first to the fifth day.
0.0 0.5 1.0 1.5 (mg L-1) 0.2 0.4 0.6 0.8 (mg L-1)
1906 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908

Table 4
1
Statistical features and correlation coefficients of N in vertical SSW (mg L ).

Index Soil depth (cm) Min Max Mean Std. error Variance 15 cm 30 cm 60 cm 90 cm
TN 15 2.09 7.62 3.97 1.27 2.44 1
30 1.83 4.12 2.77 0.60 0.89 0.399a 1
60 1.24 3.50 2.55 0.60 0.82 0.166 0.149 1
90 2.26 5.40 3.39 0.69 1.09 0.060 0.451a 0.491a 1
NO3 -N 15 0.74 1.87 1.29 0.28 0.41 1
30 0.13 1.10 0.60 0.22 0.40 0.110 1
60 0.16 1.42 0.79 0.30 0.27 0.459a 0.293 1
90 0.16 1.41 0.91 0.37 0.57 0.470a 0.667a 0.706a 1
NH+4-N 15 0.37 0.96 0.72 0.16 0.21 1
30 0.24 0.79 0.52 0.14 0.21 0.762a 1
60 0.24 0.73 0.49 0.10 0.26 0.507a 0.529a 1
90 0.22 0.60 0.42 0.09 0.13 0.066 0.162 0.233 1
a
Correlation is significant at the 0.01 level (2-tailed).

The relationships between the TN concentration in SW and the

30
(a) soil microclimate factors were identified with the aid of statistical
TN 7
TN (b) analysis (Table 5). The five-way ANOVA analysis indicated that the
N concentration (mg L-1)

25 NO3--N 6 NO3--N
air temperature, soil temperature, and ponding level did not signif-
N concentration of SW (mg L-1)

NH4+-N
NH4+-N 5 icantly correlate with the concentration of TN in the SW. The soil
20 2 moisture and the pH of the SW affected the TN concentration to
1
some extent (p < 0.05). Statistically, the concentrations of NO3 -N
15
and NH+4-N were not significantly correlated with soil moisture,
0
5-14 5-15 5-16 5-17 5-18 5-19 5-20
pH and temperature.
10
Date
The relationships between the TN concentration in SSW at the
15 cm depth and the soil microclimate factors were also analyzed
5
(Table 6). The three-way ANOVA analysis indicated that the soil
temperature and moisture were closely, to some extent, related
0
to the TN concentration in the 15 cm layer (p < 0.05). The soil
microclimatic factors did not significantly impact the NO3 -N and
4-20 5-4 5-18 6-1 6-15 6-29 7-13 7-27 8-10 NH+4-N concentrations in either the 15 or 90 cm layers. In the
Date 30 cm layer, each N form was correlated with the soil temperature
of each layer (p < 0.05). In the 60 cm layer, the soil moisture had a
Fig. 6. N Concentrations in SW (a) and in recession flow (b) with drainage time significant effect on NO3 -N (p < 0.05), and the pH had a significant
during the ponding period.
effect on NH+4-N (p < 0.05).

3.5. Interaction analysis of soil water N with soil microclimatic factors

4. Discussion

In the paddy rice fields in this middle-to-high latitude area, the

4.1. Soil ecohydrology and N pattern in middle-to-high latitude area
soil temperature varied intensively from April to the end of
September (Fig. 7). The temperature at the 15 cm depth fluctuated
The special soil ecohydrology pattern due to the vertical profile
more than it did at the lower three layers due to the heat exchange
of the soil is the basis for rice cultivation in this cold middle-to-
with the soil surface or air and the soil texture of the profile.
Furthermore, the soil temperature of the 15 cm layer and the air Table 5
temperature were closely correlated with each other throughout ANOVA analysis of soil microclimatic factors on the concentration of TN in SW.
the growing season (PC = 0.971, p = 0.000). Variation Sum of squares Mean squares F P
Soil temperature 9.020 4.510 3.088 0.120
Ponding level 14.601 3.650 1.199 0.375
Soil moisture 16.860 8.430 5.772 0.040
pH 16.264 8.132 5.568 0.043
Error 8.763 1.460
Corrected total 65.984 26.658

0 Table 6
5 ANOVA results of soil microclimatic factors with TN concentration in SSW at 15 cm
10 depth.
15
20 Variation Sum of squares Mean squares F P
25 Soil temperature 12.410 0.886 4.187 0.044
Soil moisture 1.251 1.251 5.910 0.049
Ponding Growing stage pH 0.005 0.005 0.021 0.889
period
Error 1.270 0.212
Corrected total 16.426
Fig. 7. Temporal and vertical pattern of soil temperatures.
 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908
high latitude area. Compared with a sand/gravel medium, clay soil
substrates have a high water retention capacity and porosity (Horn
et al., 1994; Zotarelli et al., 2009), which influence the microbial
community and N pattern in the soil. The lower acquisition rate
of soil water at the 30–60 cm depths meant there was an imperme-
able layer with the weak permeability. The soil moisture and
acquisition rate is different in comparison with other areas. In rice
field in tropical Thailand with loamy sand at depths of 0–60 cm
and sand clay at depths below 60 cm, the soil moisture at
100 cm was always significantly higher than that at 60 cm
(Moroizumi et al., 2009). The soil moisture is closely related to
the soil texture as dispersion theory has show, where a linear rela-
tionship between the coefficient of hydrodynamic dispersion and
the water velocity has been verified (Nützmann et al., 2002). For
similar highly mechanized tillage practices, the soil properties in
the rice field are fairly homogeneous during the agricultural
exploitation, which leads to a clear gradient in the soil water acqui-
sition rate.
In a paddy rice field, the soil has a greater chance of being stored
in anaerobic holding ponds and subjected to the freeze–thawing.
Due to the anaerobia, NO3 -N may not be produced in waterlogged
profiles (Cooper et al., 2007). Thus, the NO3 -N accounts for a small
proportion of TN in the field experiment. With the tillage before
transplanting, the rice field is recharged with aerobic water. This
may cause an alternating aerobic and anaerobic condition that
would induce nitrification and subsequent denitrification.
Precipitation and irrigation intensify soil hydrological processes
and improve the NO3 -N leaching (Sun et al., 2008; Gheysari
et al., 2009), which also contributed slight N to the field with the
irrigation water and rainfall. The N from the irrigation is covered
in the water sample, but this part of N can be explained separately.
4.2. Interactions of soil water with N transportation
The interactions between N concentrations with soil water
could be affected by the N accumulation efficiency of crops and
the soil N concentration. Furthermore, the amount of soil N in
paddy rice fields can be supplemented by the substantial release
of N by the soil (Reddy and Dangelo, 1997). The MANOVA results
indicated that the soil water dynamics significantly affect the N
concentration in the SW and SSW at the 15 cm layer (Tables 5
and 6). The NO3 -N concentration at the 60 cm layer is high in
the early growing season and low in the late growing season under
the seasonal moisture dynamics, which is also controlled by the
effect of soil moisture on net nitrification (Wang et al., 2006).
The nitrification rate is typically rapid under thermic and hyper-
thermic conditions (Thompson et al., 2002), which may explain
the soil temperature’s significant impacts on the N dynamics in
some soil layers. For the SW, the impact of soil temperature on
nitrification is not so significant due to the rapid leaching of
NO3 -N. The decreasing peak value of the NH+4-N concentration is
consistent with ammonia volatilization, which is accompanied by
its quick attenuation (Sommer et al., 1991).
The rice was planted in the beginning of May and harvested in
the end of September (Table 1). This is a longer growth time than
occurs in the tropical areas (Yang and Chang, 1999). Therefore,
the water and N addition face a greater challenge under the longer
growing period. The peak concentration in SSW ranged from
3.50 mg L 1 to 7.62 mg L 1, which was similar to the value in the
SW. Compared with the N concentration and water flow in warm
agricultural areas (Zhao et al., 2014), the N loss percentage was
lower in this middle-to-high latitude area. However, the largest
N concentration in SW after the thawing period and the peak value
in recession water after the fertilization still highlight the risk of N
to the aquatic environment. With respect to soil N conservation,
1907
the accurate fertilization with consideration of soil water and the
SW adjustment after the thawing are the priority issues.
4.3. Paddy rice N optimization in middle-to-high latitude area
The effects of soil water on N redistribution in rice soil are
mainly related to the differences in pH and N transformations,
which also influence N optimization. The SW’s pH will decrease
after fertilization, which is due to the organic acid exudation from
the rice residue impelled by the fertilizer (Surekha et al., 2003).
However, the ponding period and waterlogging events would
accelerate urea hydrolyzation and increase the pH in soil
(Ponnamperuma, 1972), which will occur more rapidly in the
middle-to-high latitude paddy rice area. The pH at the 60 cm depth
has a more direct impact on the NH+4-N concentration, which
increased on June 16, July 12 and July 29 with correspondingly
higher pH values. The SW’s pH can prove an indication of the sorp-
tion and dissolution of N in soil (Wang and Alva, 2000). The pH
should be frequently monitored during tillage and before fertiliza-
tion planning, which also varies under different type of the soil.
The spatial distributions of the horizontal variations cause some
uncertainty on the N optimization, especially for the soil
properties.
Vertical monitoring in Japanese middle latitude areas has indi-
cate that flooded rice fields have higher NH+4-N concentrations in
the upper layers than in the non-flooded columns (Zhou et al.,
2009). The similar findings in our middle-to-high latitude study
area suggest that anaerobic ammonia oxidation (anammox) may
occur under anaerobic conditions. As a result of the vertical pattern
of the soil in this area, the anaerobic condition varies at different
depths and has a larger impact on N concentration than does the
pH in the SSW. Due to soil texture of the impermeable layer, it
has a much stronger soil redox potential under intermittently
flooded vertical flows (Minamikawa and Sakai, 2006). It is our con-
clusion that pH and irrigation adjustments provide effective solu-
tions for N optimization in middle-to-high latitude paddy rice
areas.
5. Conclusions
This study explored the temporal–vertical dynamics of soil
water profile and patterns of three different N forms in paddy rice
land in a middle-to-high latitude area. The analysis compared the
N concentration and transport among different soil microclimate
conditions. The concentrations of each N form in the SW and
SSW were low with the low fertilizer level compared with other
paddy rice areas. The temporal variations in NH+4-N and NO3 -N
were not significant, but the spatial distributions of water and N
were different for the impermeable layer at the 30 cm depth. We
concluded that soil texture played an important role in soil water
and N dynamics in the middle-to-high latitude paddy rice area. It
improved water retention and N capacity of soil, thus improving
conditions for rice growing.
Soil temperature and precipitation significantly affected the
NO3 -N concentration in this middle-to-high latitude paddy rice
field. Before the rice was planted in May, the soil temperature
was the key factor for soil moisture and N loss in the surface flow.
On average, the N concentrations in soil water were partially asso-
ciated with the soil temperature and moisture. Under the relatively
longer growth period, the N had more opportunities to slow down.
Furthermore, the concentrations of the three N forms in the 90 cm
layer had pronounced peaks during the early ponding period while
the layers above 60 cm was in a frozen state. The distributions of
the soil water and N during the freeze–thaw period should be doc-
umented in future studies.
1908 W. Ouyang et al. / Journal of Hydrology 529 (2015) 1901–1908
Acknowledgements
The research discussed in this paper benefited from financial
support from the National Natural Science Foundation of China
(Grant No. 41371018), the Supporting Program of the ‘‘Twelfth
Five-year Plan” for Sci & Tech Research of China
(2012BAD15B05), the Fundamental Research Funds for the
Central Universities, and the Scientific Research Foundation for
the Returned Overseas Chinese Scholars, State Education Ministry.
References
Arauzo, M., Valladolid, M., Martínez-Bastida, J.J., 2011. Spatio-temporal dynamics of
nitrogen in river-alluvial aquifer systems affected by diffuse pollution from
agricultural sources: implications for the implementation of the Nitrates
Directive. J. Hydrol. 411 (1–2), 155–168.
Bagarello, V., Di Stefano, C., Ferro, V., Iovino, M., Sgro, A., 2010. Physical and
hydraulic characterization of a clay soil at the plot scale. J. Hydrol. 387 (1–2),
54–64.
Boling, A.A., Tuong, T.P., Suganda, H., Konboon, Y., Harnpichitvitaya, D., Bouman, B.
A.M., Franco, D.T., 2008. The effect of toposequence position on soil properties,
hydrology, and yield of rainfed lowland rice in Southeast Asia. Field Crop. Res.
106 (1), 22–33.
Bryant, J.R., Snow, V.O., Cichota, R., Jolly, B.H., 2011. The effect of situational
variability in climate and soil, choice of animal type and N fertilisation level on
nitrogen leaching from pastoral farming systems around Lake Taupo, New
Zealand. Agric. Syst. 104 (3), 271–280.
Caliskan, S., Ozkaya, I., Caliskan, M.E., Arslan, M., 2008. The effects of nitrogen and
iron fertilization on growth, yield and fertilizer use efficiency of soybean in a
Mediterranean-type soil. Field Crop. Res. 108 (2), 126–132.
Chen, S.Y., Ouyang, W., Hao, F.H., Zhao, X.C., 2013a. Combined impacts of freeze–
thaw processes on paddy land and dry land in Northeast China. Sci. Total
Environ. 456–457 (1), 24–33.
Chen, W.W., Wang, Y.Y., Zhao, Z.C., Cui, F., Gu, J.X., Zheng, X.H., 2013b. The effect of
planting density on carbon dioxide, methane and nitrous oxide emissions from
a cold paddy field in the Sanjiang Plain, northeast China. Agric. Ecosyst. Environ.
178 (15), 4–70.
Cooper, P., 2005. The performance of vertical flow constructed wetland systems
with special reference to the significance of oxygen transfer and hydraulic
loading rates. Water Sci. Technol. 51 (9), 81–90.
Cooper, R., Thoss, V., Watson, H., 2007. Factors influencing the release of dissolved
organic carbon and dissolved forms of nitrogen from a small upland headwater
during autumn runoff events. Hydrol. Process. 21 (5), 622–633.
Craine, J.M., Gelderman, T.M., 2011. Soil moisture controls on temperature
sensitivity of soil organic carbon decomposition for a mesic grassland. Soil
Biol. Biochem. 43 (2), 455–457.
Fellman, J.B., D’Amore, D.V., Hood, E., 2008. An evaluation of freezing as a
preservation technique for analyzing dissolved organic C, N and P in surface
water samples. Sci. Total Environ. 392 (2–3), 305–312.
Fuentes, J.P., Flury, M., Huggins, D.R., Bezdicek, D.F., 2003. Soil water and nitrogen
dynamics in dryland cropping systems of Washington State, USA. Soil Till. Res.
71 (1), 33–47.
Fuller, K.D., Gordon, R., Grimmett, M., Fillmore, S., Madani, A., VanRoestel, J.,
Stratton, G.W., MacLeod, J., Embree, C., St George, E., 2010. Seasonal and crop
rotational effects of manure management on nitrate-nitrogen leaching in Nova
Scotia. Agric. Ecosyst. Environ. 137 (3–4), 267–275.
Gheysari, M., Mirlatifi, S.M., Homaee, M., Asadi, M.E., Hoogenboom, G., 2009. Nitrate
leaching in a silage maize field under different irrigation and nitrogen fertilizer
rates. Agric. Water Manage. 96 (6), 946–954.
Hagedorn, F., Bucher, J.B., Schleppi, P., 2001. Contrasting dynamics of dissolved
inorganic and organic nitrogen in soil and surface waters of forested
catchments with Gleysols. Geoderma 100 (1–2), 173–192.
Hao, F., Lai, X., Ouyang, W., Xu, Y., Wei, X., Song, K., 2012. Effects of land use changes
on the ecosystem service values of a reclamation farm in northeast China.
Environ. Manage. 50 (5), 888–899.
Hao, F.H., Chen, S.Y., Ouyang, W., Shan, Y.S., Qi, S.S., 2013. Temporal rainfall patterns
with water partitioning impacts on maize yield in a freeze–thaw zone. J. Hydrol.
486 (12), 412–419.
Hopmans, J.W., 2003. Soil hydrological properties and processes and their
variability in space and time. J. Hydrol. 272 (1–4), 1–2.
Horn, R., Taubner, H., Wuttke, M., Baumgartl, T., 1994. Soil physical properties
related to soil structure. Soil Till. Res. 30 (2), 187–216.
Hupet, F., Vanclooster, M., 2005. Micro-variability of hydrological processes at the
maize row scale: implications for soil water content measurements and
evapotranspiration estimates. J. Hydrol. 303 (1–4), 247–270.
Jalali, M., 2005. Nitrates leaching from agricultural land in Hamadan, western Iran.
Agric. Ecosyst. Environ. 110 (3–4), 210–218.
Joseph, G., Henry, H.A.L., 2008. Soil nitrogen leaching losses in response to freeze–
thaw cycles and pulsed warming in a temperate old field. Soil Biol. Biochem. 40
(7), 1947–1953.
McIsaac, G.F., Libra, R.D., 2003. Revisiting nitrate concentrations in the Des Moines
River: 1945 and 1976–2001. J. Environ. Qual. 32 (6), 2280–2289.
Minamikawa, K., Sakai, N., 2006. The practical use of water management based on
soil redox potential for decreasing methane emission from a paddy field in
Japan. Agric. Ecosyst. Environ. 116 (3–4), 181–188.
Möller, K., Habermeyer, J., Zinkernagel, V., Reents, H.-J., 2006. Impact and
interaction of nitrogen and Phytophthora infestans as yield-limiting and
yield-reducing factors in organic potato (Solanum tuberosum L.) crops. Potato
Res. 49 (4), 281–301.
Moroizumi, T., Hamada, H., Sukchan, S., Ikemoto, M., 2009. Soil water content and
water balance in rainfed fields in Northeast Thailand. Agric. Water Manage. 96
(1), 160–166.
Nützmann, G., Maciejewski, S., Joswig, K., 2002. Estimation of water saturation
dependence of dispersion in unsaturated porous media: experiments and
modelling analysis. Adv. Water Resour. 25 (5), 565–576.
Ouyang, W., Xu, Y.M., Hao, F.H., Wang, X.L., Lin, C.Y., 2013. Effect of long-term
agricultural cultivation and land use conversion on soil nutrient contents in the
Sanjiang Plain. CATENA 104 (5), 243–250.
Ponnamperuma, F.N., 1972. The chemistry of submerged soils. Adv. Agron. 24, 29–
96.
Reddy, K.R., Dangelo, E.M., 1997. Biogeochemical indicators to evaluate pollutant
removal efficiency in constructed wetlands. Water Sci. Technol. 35 (5), 1–10.
Shipitalo, M.J., Dick, W.A., Edwards, W.M., 2000. Conservation tillage and macropore
factors that affect water movement and the fate of chemicals. Soil Till. Res. 53
(3–4), 167–183.
Shrestha, R.R., Bardossy, A., Rode, M., 2007. A hybrid deterministic-fuzzy rule based
model for catchment scale nitrate dynamics. J. Hydrol. 342 (1–2), 143–156.
Singh, R., van Kroes, J.G., Dam, J.C., Feddes, R.A., 2006. Distributed ecohydrological
modelling to evaluate the performance of irrigation system in Sirsa district,
India: I. Current water management and its productivity. J. Hydrol. 329 (3–4),
692–713.
Sleutel, S., Moeskops, B., Huybrechts, W., Vandenbossche, A., Salomez, J., De Bolle, S.,
Buchan, D., De Neve, S., 2008. Modeling soil moisture effects on net nitrogen
mineralization in loamy wetland soils. Wetlands 28 (3), 724–734.
Sommer, S.G., Olesen, J.E., Christensen, B.T., 1991. Effects of temperature, wind-
speed and air humidity on ammonia volatilization from surface applied cattle
slurry. J. Agric. Sci. 117, 91–100.
Sun, B., Chen, D., Li, Y., Wang, X., 2008. Nitrogen leaching in an upland cropping
system on an acid soil in subtropical China: lysimeter measurements and
simulation. Nutr. Cycl. Agroecos. 81 (3), 291–303.
Surekha, K., Kumari, A.P.P., Reddy, M.N., Satyanarayana, K., Cruz, P.C.S., 2003. Crop
residue management to sustain soil fertility and irrigated rice yields. Nutr. Cycl.
Agroecos. 67 (2), 145–154.
Thompson, T.L., Doerge, T.A., Godin, R.E., 2002. Subsurface drip irrigation and
fertigation of broccoli: I. Yield, quality, and nitrogen uptake. Soil Sci. Soc. Am. J.
66 (1), 186–192.
Wang, F.L., Alva, A.K., 2000. Ammonium adsorption and desorption in sandy soils.
Soil Sci. Soc. Am. J. 64 (5), 1669–1674.
Wang, C.H., Wan, S.Q., Xing, X.R., Zhang, L., Han, X.G., 2006. Temperature and soil
moisture interactively affected soil net N mineralization in temperate grassland
in Northern China. Soil Biol. Biochem. 38 (5), 1101–1110.
Yang, S.S., Chang, H.L., 1999. Diurnal variation of methane emission from paddy
fields at different growth stages of rice cultivation in Taiwan. Agric. Ecosyst.
Environ. 76 (2–3), 75–84.
Yu, S., Ehrenfeld, J.G., 2009. The effects of changes in soil moisture on nitrogen
cycling in acid wetland types of the New Jersey Pinelands, USA). Soil Biol.
Biochem. 41 (12), 2394–2405.
Zhao, Z., Zhang, H.L., Li, C.S., Zhao, Q., Cao, L.K., 2014. Quantifying nitrogen loading
from a paddy field in Shanghai, China with modified DNDC model. Agric.
Ecosyst. Environ. 197 (1), 212–221.
Zhou, S., Nakashimada, Y., Hosomi, M., 2009. Nitrogen transformations in vertical
flow systems with and without rice (Oryza sativa) studied with a high-
resolution soil-water profiler. Ecol. Eng. 35 (2), 213–220.
Zotarelli, L., Dukes, M.D., Scholberg, J.M.S., Munoz-Carpena, R., Icerman, J., 2009.
Tomato nitrogen accumulation and fertilizer use efficiency on a sandy soil, as
affected by nitrogen rate and irrigation scheduling. Agric. Water Manage. 96 (8),
1247–1258.
Zvomuya, F., Rosen, C.J., Gupta, S.C., 2006. Nitrogen and phosphorus leaching from
growing season versus year-round application of wastewater on seasonally
frozen lands. J. Environ. Qual. 35 (1), 324–333.