Agricultural Water Management 104 (2012) 235–243

Contents lists available at SciVerse ScienceDirect

Agricultural Water Management

journal homepage: www.elsevier.com/locate/agwat

Assessing nutrient losses of reclaimed wastewater irrigation in paddy fields for

sustainable agriculture
T.I. Jang a,b , H.K. Kim c,∗ , C.H. Seong d , E.J. Lee e , S.W. Park b,f
a
Department of Biological and Agricultural Engineering, University of Georgia, 2360 Rainwater Road, Tifton, GA 31793, USA
b
Research Institute for Agricultural & Life Sciences, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul 151-921, Republic of Korea
c
Department of Agricultural and Biological Engineering, Mississippi State University, 130 Creelman Street, Starkville, MS 39762, USA
d
Department of Biological Systems Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA
e
Water Environment Research Department, National Institute of Environmental Research, Environmental Research Complex, Kyungseo-dong, Seo-gu, Incheon 404-708, Republic of
Korea
f
Department of Rural Systems Engineering, Seoul National University, 599 Gwanak-ro, Gwanak-gu, Seoul 151-921, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: An experimental field study was performed during the growing season to assess water and nutrient
Received 7 June 2011 balances in a paddy field over a 3-year period. The plots were separated according to irrigation water:
Accepted 16 December 2011 groundwater (TR#1), wastewater (TR#2), and filtered wastewater with ultraviolet (UV) treatment (TR#3).
Available online 31 December 2011
The hydrology and water quality of rainfall, irrigation, surface water, and infiltration were monitored
throughout the crop stages. More than half of the total water inflow of about 1840 mm in each treatment
Keywords:
was contributed by precipitation and the remainder by irrigation. The water balance analysis indicated
Water reuse
that approximately 13% of the total outflow was lost by surface drainage, 30% was consumed by plant
Mass balance
Drainage
uptake, and 57% was lost by evapotranspiration and infiltration. The nitrogen (N) levels in the irrigation
Water management water in the mass inputs for TR#1 and TR#3 were 22% and 49%, respectively, while the output balances
Fertilization in the drainage water for TR#1 and TR#3 averaged 2% and 6%, respectively. The N in the crop harvest for
TR#1 and TR#3 occupied 59.62 and 121.35 kg ha−1 , respectively. The N in the fertilizer comprised a large
proportion of the N in TR#1 while the N in the irrigation water and fertilizer were the major inputs in
TR#3. The major P input was fertilizer in TR#3, and crop harvest was a main output in the P balance. In
contrast, surface drainage and infiltration were relatively small components, due to the high drainage
outlet height. The difference between inputs and crop harvest shows that it is possible to improve water
quality by reducing the fertilization rates in paddy fields irrigated with reclaimed wastewater.
© 2011 Elsevier B.V. All rights reserved.

1. Introduction
Extensive lowland areas in Asia and other parts of the world
are under rice paddy cultivation (Odhiambo and Murty, 1996). In
the Republic of Korea, paddy fields cover an area of 11,500 km2 ,
which comprises 61% of the nation’s total cultivated area (Kim et al.,
2009). Korea suffers from a limited water supply, including agricul-
tural irrigation water, due to population growth, urbanization, and
economic development (Jang et al., 2008). A recent national survey
on the future Korean water supply and demand reported that the
country can expect a shortage of over 0.44 billion m3 of water by
2030 (MLTM, 2006).
Because paddy rice production requires large volumes of water,
agricultural irrigation is allocated upwards of 47% of the total
annual water use in Korea. A paddy field is blocked by a levee to
∗ Corresponding author. Tel.: +1 6623253365; fax: +1 6623253853.
E-mail addresses: hk300@msstate.edu, hkkimbest@gmail.com (H.K. Kim).
maintain a ponded water condition and the water level is depen-
dent on the outlet height. The irrigation amount is determined
from the optimal ponding water depth (Kim et al., 2008) and a
total of about 1250 mm of water is generally required per rice
paddy during the growing season, which is primarily supplied by
irrigation.
Reclaimed wastewater can be an alternative water resource for
supplementary irrigation in areas that suffer from water shortages
or unsatisfactory water quality since agricultural irrigation water
does not usually require the same high grade of water quality as
drinking water (Kang et al., 2007; Jang et al., 2010b). Guidelines
for reclaimed wastewater irrigation have been adopted for paddy
fields in Korea (MOE, 2005). As a result, more data are available to
help clarify potential human health problems and assess the envi-
ronmental effects associated with reclaimed wastewater irrigation
of paddy fields. However, few reports of practical wastewater reuse
for rice paddies have appeared (Kang et al., 2007; Jang et al., 2008,
2010b). In contrast, wastewater reuse for upland crop irrigation is
being practiced in many countries (Cooper, 1991; Asano and Levine,

0378-3774/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.agwat.2011.12.022
236 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243
1996; Al-Lahham et al., 2003; Kiziloglu et al., 2007; Brunetti et al.,
2007; Palese et al., 2009).
Rice paddy culture requires large amounts of water and nutri-
ents for producing staple foods. At present, farming in Korea uses
high inputs of fertilizers (Harris, 1998), substantial amounts of
which can be lost through surface drainage unless there is a careful
balance between the input and the amount that rice plants actually
use (Yoon et al., 2006). In addition, water quality in Korea suffered
seriously during the rapid industrialization of the late 1980s, but,
due to ongoing restoration efforts, has since improved (Yoon et al.,
2003). However, the water quality of many streams and lakes often
does not meet established standards, and periodic algal blooms in
most reservoirs indicate that further efforts to safeguard quality
are needed (Yoon et al., 2003). Furthermore, the possibility that
water quality will deteriorate as reclaimed wastewater is used to
irrigate paddy fields is a matter of concern. Therefore, it is necessary
to determine the nutrient losses that result from surface drainage
from paddy fields. Recently, studies have examined this question
in Korea, with varying results (Han et al., 1999; Cho and Choi, 2001;
Cho and Han, 2002; Cho, 2003; Cho et al., 2003; Yoon et al., 2003,
2006). However, few studies have examined practical wastewater
reuse or high-nutrient-concentration irrigation of paddy fields.
The objectives of the current study are to investigate water
and nutrient balances in paddy fields irrigated by groundwater,
wastewater, and reclaimed wastewater and to assess nutrient
losses with different agricultural management practices in paddy
fields irrigated with water from a typical stream or agricultural
reservoir and reclaimed wastewater.
2. Materials and methods
2.1. Experiment site and design
The experimental plots are in the Byoung-Gem (BG) paddy field
(E 37◦ 12 32 , N 127◦ 01 18 ) located near the Suwon wastewater
treatment plant in Gyeonggi-do, Korea. The present study was con-
ducted for three crop-years from 25 May, 2006 to 21 October, 2008.
A randomized complete block design with split plot arrangements
was used with three treatments and four replicates on 5 m × 5 m
plots (Fig. 1). The three treatments were separated according to
the irrigation water: groundwater (TR#1), untreated wastewater
(TR#2), and filtered wastewater with ultraviolet (UV) treatment
(TR#3). A small scale wastewater reclamation system with a LCHE-
WRT filter system (Maeng et al., 2006), UV treatment unit, and
pipelines supply irrigation water from the wastewater effluents.
A groundwater well was installed to supply water for TR#1. There
was one rain gauge, three infiltrometers, and three five-set ceramic
porous cups at three plots. The flume was designed with a V-
notch weir with the water gauge and the V-notch was positioned
above 15 cm from the base in order to maintain a constant out-
let height, with water levels controlled to be 1–10 cm according to
rice growth stage (Kim et al., 2008). Three weirs were initialized
to the same level for measuring the water depth at each treat-
ment.
Average annual rainfall and evaporation in the study area are
1259 and 1091 mm, respectively. The average annual temperature
in this area is 11.6 ◦ C. The annual mean temperature during the
irrigation periods range from 16.7 to 25.2 ◦ C. The soil in the experi-
mental field is Gangseo series (coarse loamy, mixed, nonacid, mesic
family of Aquic Fluvaquentic Eutrudepts) (NIAST, 2000).
For this experiment, 1-month-old rice seedlings (Oryza sativa
cv. Chucheongbyeo) were transplanted in May and harvested in
October during 2006–2008. Fertilizers are typically applied three
times, during the pre-plant, tilling, and panicle growing stages,
but for this experiment they were applied one time during
the pre-plant (N:P:K = 55:45:40 kg ha−1 ) based on high-nutrient-
concentration irrigation. Insecticides were sprayed in June of every
year to exterminate rice water weevils, and weeds were controlled
manually.
2.2. Measurements and analysis methods
Ponded water depth in each experimental plot was continuously
measured by an automatic float type water level recorder. Inflow
was measured using a water gauge at the inlet pipe, and outflow
was measured using weirs installed at the outlet of three plots. The
samples of irrigation and ponded water were collected on a weekly
or biweekly basis. Three sets of the double ring infiltrometer were
installed on each plot to measure daily water loss by deep percola-
tion. Percolation water quality was sampled by five ceramic porous
cups embedded at a depth of 100 cm below the soil surface. Rain-
fall was recorded using a tipping bucket rain gauge at the site, and
evapotranspiration was estimated by the Penman–Monteith equa-
tion (Allen et al., 1998). Rainfall water was sampled when storms
occurred. Water samples were analyzed by the standard methods of
APHA (1995) for conventional parameters including total nitrogen
(T-N) and total phosphorus (T-P) concentrations.
Soil samples were collected from experimental plots before
transplanting and after harvesting. After clearing organic matter
from the surface, three soil sub-samples were taken from the root
zone (2–30 cm below the soil surface) of each plot. The soil samples
were analyzed for physical and chemical properties with the Ameri-
can Society of Agronomy (ASA) and Soil Science Society of America
(SSSA) methods for soil analysis (Chapman and Pratt, 1961). The
collected soil samples were analyzed for pH, concentrations of T-
N and T-P, and the main cations. The control plot in the Gi-cheon
paddy fields, which is near a suburb of Suwon, was located approx-
imately 6 km northeast of the BG paddy field. These plots were
irrigated with water from the Gi-cheon reservoir, which satisfies
the current agricultural water quality criteria of Korea. An extensive
hydrologic and water quality monitoring system was established
there in 1996 (Kim et al., 2008).
Rice plant samples of root and grain were collected at each plot
after harvesting, and these were analyzed for T-N and T-P con-
centrations using the micro-Kjeldahl method (Jackson, 1967). Mass
loads of T-N and T-P were calculated by multiplying water volume
and corresponding concentration.
Water balance in the paddy fields is estimated by the variation
in ponded water depth (WD), expressed as
WDi = WDi−1 + IRi + PRi − (DRi + ETi + INi ) + Uni (1)
where, WD is ponded water depth, IR is the amount of irrigation
water, PR is the amount of precipitation, DR is the amount of surface
drainage through a weir, ET is the amount of evapotranspiration,
IN is the amount of infiltration water, and Un is the unmeasured
amount. The subscript i represents the ith day.
The nutrient input to the paddy fields is categorized as either
natural supply or fertilization, where natural supply includes atmo-
spheric deposition and irrigation water, and fertilization includes
mineral and organic sources (Yoon et al., 2003). Nutrient output
includes surface drainage through the weir, deep percolation, and
plant uptake. The general mass balance equation for both T-N and
T-P is approximated in this study as
IIR + IPR + IFR = ODR + OIN + OHR (2)
where, IIR is input from irrigation, IPR is input from rainfall, IFR is
input from fertilization, ODR is output by surface drainage, OIN is
output by deep percolation, and OHR is output by crop harvest.
The nutrient balance in a paddy field is also affected by other
factors, including acidic deposition, biological N fixation, miner-
alization, immobilization of soil N and P, ammonia volatilization,
 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243
Fig. 1. Location and schematic representation of BG study site and field experiments including the reuse system, water sampling gauges, and weirs.
and denitrification. These other parameters in the mass balance
are generally either estimated by closure or neglected (Yoon et al.,
2006; Maruyama et al., 2008). In this study, we evaluated nutrient
input by precipitation, irrigation, chemical fertilizer, and nutrient
losses by surface drainage, infiltration, and crop harvest.
3. Results and discussion
3.1. Water balance
Table 1 shows the annual water balance of the experimental
paddy fields during the irrigation period. The monitoring results
show rainfall of 1033–1111 mm, ET of 526–545 mm, and infiltration
of 957–1085 mm. Average inflow amounts to the paddy fields for
TR#1, TR#2, and TR#3 were 1816, 1877, and 1834 mm, respectively,
and about 13% of this was lost to surface drainage. The average
drainage amount for each treatment ranged from 215 to 236 mm
with an average of 227 mm. The water balance analysis indicated
that approximately 30% was consumed by crop harvest and 57% was
lost by evapotranspiration and infiltration. The average unmea-
sured amount of the differences between input and output systems
was from −7 to 135 mm when the water balance was calculated
during the study period. This may have been affected by differ-
ent hydrologic conditions and different irrigation intensities of the
three irrigation systems. These values are consistent with previous
findings of unmeasured amounts that ranged from 6 to 132 mm
(Yoon et al., 2006).
Paddy field water input during the irrigation period generally
varies from 500 to 800 mm and can reach more than 3000 mm (De
Datta et al., 1973; Hukkeri and Sharma, 1980). The amount of irri-
gation in this study area (774 mm) was less than in other regions
in Korea experiencing similar amounts of rainfall (see Table 2; e.g.,
1250 mm in reference 1, 1497 mm in reference 5). However, the
quantity of irrigation water in reference 6 was far less than that
generally required, probably due to unusually high precipitation
(July: 541.8 mm, August: 284 mm, September: 284.2) during the
rice paddy irrigation (Cho et al., 2008). The reason for the low irri-
gation amount in the current study is that the average drainage was
237
far lower than in other studies (see Table 2; e.g., reference 1, 2, 3,
and 5).
3.2. Nutrient concentrations of water samples and soil
Nutrient concentration results for the observed water samples
are summarized in Table 3. In this study, we focus on two treat-
ments, TR#1 and TR#3, because the irrigation of TR#2, wastewater
effluent, is not used for actual irrigation, and the nutrient concen-
trations are only slightly higher in TR#2 than in TR#3. In addition,
we are not currently addressing a primary objective for evaluating
irrigation with wastewater effluent, which is to assess potential
human health problems and to secure a practical reuse system for
paddy fields (Jang et al., 2008, 2010b). The observed mean con-
centrations of T-N and T-P in rainfall were 1.45 and 0.02 mg l−1 ,
respectively. For TR#3, the average T-N influent concentration was
7.80 mg l−1 , compared to 3.45 mg l−1 at the ponding paddy fields.
This average irrigation concentration of T-N was 344% higher in
TR#3 than in TR#1. The average reduction rate of T-N in TR#3 was
56%. Similarly, the average reduction rate of T-N in TR#1 was 53%.
The average T-P influent concentrations over 3 years for TR#1 and
TR#3 were 0.05 and 0.40 mg l−1 , respectively, while at the ponding
paddy fields they were 0.10 and 0.28 mg l−1 , respectively. The aver-
age reduction rate of T-P for TR#3 was 30%, compared to −100% for
TR#1 (that is, the water quality in TR#1 deteriorated after irrigat-
ing). Field data showed that the concentration of soluble nutrient
constituents was significantly reduced by the paddy fields, which
functioned like a treatment wetland with water purification capac-
ity. General purification rates of the N and P in paddy fields have
been estimated to be in the ranges of 52–66 and 27–65%, respec-
tively (Eom, 2001). The present study showed similar trends. The
high nutrient concentration of ponded water can be consumed by
rice plant uptake, soil absorption, nitrification and denitrification,
and other purification processes (Yoon, 2009).
Figs. 2 and 3 show the temporal trend by month of T-N and T-P
concentrations in the drainage water for each treatment. The nutri-
ent concentrations in the drainage water were high in the early
stages of rice growth in June, but subsequently decreased due to
high precipitation. This is similar to the results seen in other Korean
238 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243

Table 1
Water balance of the treatment paddy fields during the irrigation period.

Year Inflow (mm) Outflow (mm) Unmeasured (mm)

PR IR ET IN DR

TR#1 TR#2 TR#3 TR#1 TR#2 TR#3 TR#1 TR#2 TR#3

2006 1033.0 858.2 898.5 885.6 531.3 999.6 299.7 315.7 329.7 60.7 84.9 58.1
2007 1035.6 796.4 887.8 814.1 544.6 1084.7 139.7 158.8 100.6 62.9 135.3 119.8
2008 1110.5 613.7 666.9 623.2 525.5 957.2 204.7 234.4 258.3 36.8 60.3 −7.3
Ave. 1059.7 756.1 817.7 774.3 533.8 1013.8 214.7 236.3 229.5 53.5 93.5 56.9

PR is the amount of precipitation, IR is the amount of irrigation water, ET is the amount of evapotranspiration, IN is the amount of infiltration water, and DR is the amount of
surface drainage water.

Table 2
Comparison of water balance, fertilization, and T-N and T-P losses in drainage from paddy fields.

References Water balance (mm) Fertilization (kg ha−1 ) Losses (kg ha−1 )

PR IR ET IN DR T-N T-P T-N T-P

1. Cho and Choi (2001) 1288.9 1250.3 532.3 613.1 1329.2 173 62.0 57.8 2.33
2. Cho and Han (2002) 1231.0 816.3 447.4 513.8 1027.0 184 39.3 113.7 3.70
3. Yoon et al. (2003) 511.3 670.3 488.5 99.9 593.2 110 45.0 12.7 1.21
4. Feng et al. (2004, 2005) 522.0 1781.0 528.0 – 1699.0 62 28.0 41.7 2.74
5. Yoon et al. (2006) 1183.4 1497.3 593.5 540.0 1617.8 125 15.7 59.7 1.43
6. Cho et al. (2008)a 1030.4 282.2 330.0 52.7 885.6 200 51.0 30.3 1.45
7. Maruyama et al. (2008) 1900.7 327.4 764.4 244.6 1219.1 57 39.4 10.9 1.50
8. TR#3 of this study 1059.7 774.3 533.8 1013.8 229.5 55 45.0 8.5 0.31

PR is precipitation, IR is irrigation water, ET is evapotranspiration, IN is infiltration water, and DR is surface drainage water through a weir.
a
This plot is the reclaimed paddy fields from the tideland.

studies (Cho and Choi, 2001; Cho, 2003; Yoon et al., 2006). The
annual trend in T-N was similar to that of T-P, which high con-
centrations during the early stages of rice growth after fertilizer
application, and decreasing thereafter. The peak concentration of
T-N occurred after basal fertilizer application. Korean farmers typi-
cally apply 50% of total N fertilizer just before the transplantation of
a paddy, and the remainder in two subsequent fertilizations. How-
ever, only the first remainder fertilizer was applied in this study,
because the irrigation water had a high nutrient concentration.
The changes in the concentrations of T-N and T-P in the infil-
tration water that moved through the buried ceramic porous cups
were measured in the experimental field. The averaged T-N con-
centrations in the infiltration water for TR#1 and TR#3 were 1.85
and 2.38 mg l−1 , respectively, that is, TR#3 was 28% higher than
TR#1. However, for T-P, there were no apparent trends of difference
between the treatments.
In the infiltration water, the NH4+ concentration is higher than
the NO3− concentration, because NH4+ is easily leached downward
with adsorption to soil particles (Seol et al., 2000). In contrast, the
NH4+ concentration presents higher in surface drainage than in sub-
surface infiltration water, because downward movement of NH4+
is retarded by adsorption onto soil particles. The concentrations of
NH4+ and NO3− were higher in TR#3 than in TR#1, due to the high
influent concentration, similar to what was observed for T-N.
Annual average physico-chemical properties of each paddy soil
are given in Table 4, and the results are as follows: the pH of the

Table 3
Water quality characteristics of water samples from 2006 to 2008.

Samples T-N (mg l−1 ) Ammonia-N (mg l−1 ) Nitrate-N (mg l−1 ) T-P (mg l−1 )

TR#1 TR#2 TR#3 TR#1 TR#2 TR#3 TR#1 TR#2 TR#3 TR#1 TR#2 TR#3

PRa Mean 1.45 0.48 0.35 0.02

Max 4.59 1.89 1.12 0.05
Min 0.28 0.00 0.19 0.00
SD 1.11 0.68 0.27 0.02
No. 11
IR Mean 2.27 8.62 7.80 0.65 3.16 2.62 0.58 5.14 4.61 0.05 0.49 0.40
Max 9.00 23.85 24.06 4.37 13.57 13.00 1.37 21.68 16.58 0.62 1.83 1.84
Min 0.09 0.58 0.13 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00
SD 1.91 5.12 5.81 1.00 3.34 3.13 0.46 5.26 5.22 0.11 0.49 0.44
No. 36
DR Mean 1.06 2.47 3.45 0.43 1.09 0.84 0.05 1.39 2.31 0.10 0.26 0.28
Max 3.10 9.05 10.34 2.82 5.24 4.57 0.63 6.36 9.11 0.23 0.72 0.74
Min 0.09 0.04 0.03 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
SD 0.91 2.86 3.25 0.85 1.78 1.22 0.16 2.08 2.93 0.07 0.22 0.21
No. 18 17 17 18 17 17 18 17 17 18 17 17
IN Mean 1.85 2.08 2.38 1.01 1.26 1.54 0.00 0.00 1.32 0.20 0.19 0.19
Max 10.03 15.02 21.32 6.55 11.88 16.80 0.00 0.00 16.80 1.02 0.95 0.91
Min 0.01 0.01 0.06 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
SD 2.92 4.15 5.97 1.98 3.38 4.81 0.00 0.00 4.46 4.46 0.26 0.25
No. 12

PR is precipitation, IR is irrigation water, DR is surface drainage, and IN is infiltration water.

a
The rainfall sampling period was during 2008.
 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243 239

Table 4
Physico-chemical properties of the soil in the experimental paddy fields.

Properties Unit TR#1 TR#2 TR#3 Gi-cheona

Sand % 48.7 48.7 48.7 31.7

Silt 32.3 35.3 30.8 45.3
Clay 19.0 16.0 20.5 23.0
Texture class – Loam Loam Loam Loam
Organic matter (%) 1.74 2.11 1.82 3.18
pH (1:5) – 5.55 5.52 5.53 4.44
CEC cmol+ kg−1 13.09 12.11 13.29 13.05
EC ␮S m−1 0.473 0.319 0.524 0.263
T-N mg kg−1 850.0 930.0 950.0 1310.0
T-P 528.2 557.7 358.3 724.00
Ca 758.5 853.8 745.2 555.89
Mg 66.7 64.6 63.3 47.5
Na 26.5 34.2 30.8 12.8
K 40.9 44.5 49.5 54.5
a
Gicheon paddy fields, which were irrigated from the nearby stream water that satisfied the agricultural water quality criteria of Korea, were selected as a control in order
to compare the test plots.

paddy soil was about 5.5. The soil electrical conductivity (EC) in
TR#3 ranged from 72.1 to 375.3 ␮S cm−1 . The EC values in TR#1
were maintained at a relatively low level, due to the low EC val-
ues of the TR#1 influent. However, a high concentration of sodium
in wastewater might cause damage to the physico-chemical prop-
erties of paddy soil, and a long-term investigation of the soil
environment is needed. In this study, the soil sodium concentra-
tion (Na) in TR#3 stabilized gradually from 101.2 to 46.2 mg kg−1
during the study period. Total soil N also increased in the spring of
each year in all plots, but declined to the levels of the fallow plots
by the fall of each year. No long term variations in total soil N and
P were apparent.
3.3. Nutrient balance
The N balance of the input–output systems and unmeasured
difference are summarized in Table 5. The crop harvest in the mass
balance has been monitored to consider the nutrient uptake by rice
since 2007. While N fertilizer was applied at the same rate in TR#1
and TR#3, the input balances were 61% and 40%, respectively. The
input balances in precipitation were 17% in TR#1 and 11% in TR#3,
respectively. The N levels in irrigation water in the mass inputs for
TR#1 and TR#3 were 22% and 49%, respectively. The N in the irriga-
tion water was about two times higher in TR#3 than in TR#1. The
output balances in the drainage water for TR#1 and TR#3 were 2%
and 6%, respectively. This result is a relatively low value compared
to other studies (see Table 2). The N levels in the infiltration water
in the mass output were 22% in TR#1 and 15% in TR#3, and those in
the crop harvest were 76% and 80% for TR#1 and TR#3, respectively.
The N in the crop harvest for TR#1 averaged 59.62 kg ha−1 , and
that for TR#3 averaged 121.35 kg ha−1 . The rice culture efficiency
by crop harvest was higher in TR#3 than in TR#1. The N in the fer-
tilizer comprised a large proportion of the N in TR#1 while the N in
the irrigation water and fertilizer were major inputs in TR#3, 40%
and 49%, respectively. Surface drainage and infiltration were rela-
tively small components. Total N outputs exceeded total N inputs in
TR#3. The differences were caused by unmeasured elements (e.g.,
denitrification-N fixation).
Table 6 shows the annual P mass balance for 3 years, the mean
values of the input and output systems, and the unmeasured dif-
ference. The major P input in each treatment was fertilizer (8% in
irrigation water) and crop harvest was also the main output in the P
balance. Crop harvests in the same output for TR#1 and TR#3 were
76% and 84%, respectively. The cycling of P does not have an atmo-
spheric component like denitrification or N fixation. The P outputs
were larger than inputs by about 30 kg ha−1 in all treatments. This
value is larger than in previous studies, at 4.7 kg ha−1 (Maruyama
et al., 2008). This result indicates that there is a possibility for
improvement by reducing the application rate of P fertilization.
Our results reveal that a significant amount of nutrients can
be lost via infiltration and crop harvest. Therefore, the paddy
fields irrigated with reclaimed wastewater require proper fertilizer
management for agricultural sustainability. Cho et al. (2006) inves-
tigated using domestic wastewater, demonstrating that reducing
the standard application of chemical fertilizer by as much as 50%
reduce the harvesting index only 10%. This implies that proper fer-
tilizer management considering the high concentration of nutrients
in irrigation water can achieve improvement in water quality.
3.4. Crop growth and yield
Crop growth and yield data were investigated and analyzed in
the BG plots from 2006 to 2009. The results indicate that reclaimed
wastewater irrigation caused no adverse effects on the growth
and yield of rice. The average yield for TR#3 (5197 kg ha−1 ) was
about 15% greater than for TR#1, and this was also greater than the
national average yield (4500 kg ha−1 ) in Korea (Jang et al., 2010a,b;
Jung, 2011). Analyses of variance showed that the rice yields in
TR#1 and TR#3 were significantly different (Kang et al., 2007;
Jang, 2009). Jung (2011) reported that the total nutrients supplied
with irrigation water were highly correlated with the rice yield.
Our results similarly show that there was a significant relationship
between rice yield and the nutrient supply.
3.5. Water and fertilizer management for reclaimed wastewater
irrigation
The amounts of water balance, fertilization, and nutrient losses
by surface drainage are summarized in Table 2. In this study,
the irrigation amount was less than the consumptive water in
the rice paddy culture (i.e., evapotranspiration plus infiltration).
That is, rainfall was retained and effectively used, due to the
high drainage outlet height. Recently, Korean farmers have lit-
tle incentive to effectively use rainfall because irrigation water
is provided at no charge to them and irrigation systems are well
developed (Yoon et al., 2006). Therefore they have little concerns
for precise water management for effective rainfall consumption.
However, precise water management should be considered for
reducing nutrient losses from drainage water to the environment
of reclaimed wastewater irrigation.
The water level is adjusted by controlling the drainage outlet
height, and it can generally be as high as 10 cm during continu-
ous flooding irrigation in the growing season (Jang et al., 2010a).
The drainage amounts of reference 3, 6, and 8 are relatively low
240 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243

Table 5
Nitrogen balance in the treatment plots.

Year Classification 2006 (kg ha−1 ) 2007 (kg ha−1 ) 2008 (kg ha−1 ) Average (kg ha−1 )

Input Fertilizer 55.00 55.00 55.00 55.00

Precipitation 15.51 14.43 16.90 15.61
Irrigation TR#1 17.72 28.44 13.42 19.86
water TR#2 59.71 122.88 52.96 78.52
TR#3 57.38 108.43 36.66 67.49
Subtotal TR#1 88.23 97.87 85.32 90.47
TR#2 130.22 192.31 124.86 149.13
TR#3 127.89 177.86 108.56 138.10
Output Drainage water TR#1 2.43 1.54 0.76 1.58
TR#2 6.49 7.57 7.53 7.19
TR#3 12.95 2.66 10.00 8.54
Infiltration TR#1 7.03 43.08 2.22 17.44
water TR#2 13.50 41.67 7.72 20.96
TR#3 6.51 54.61 7.10 22.74
Crop harvest TR#1 – 58.62 60.62 59.62a
TR#2 – 137.27 118.86 128.06a
TR#3 – 135.73 106.96 121.35a
Subtotal TR#1 9.46b 103.24 63.6 78.64
TR#2 19.99b 186.51 134.11 156.21
TR#3 19.46b 193.00 124.06 152.63
Difference TR#1 – 5.37 −21.72 −8.16a
TR#2 – −5.80 9.25 1.80a
TR#3 – 15.14 15.50 15.32a
a
The averages in crop harvest and difference were calculated during 2007 and 2008.
b
The subtotals in the output system during 2006 were calculated without crop harvest.

compared to other studies. Even more pronounced in this study,
which used controlled weir-type drainage with higher drainage
outlet and pipe irrigation by a field manager, the drainage amount
was about one-fifth that of similar studies due to effective water
management. However, Yoon et al. (2006) reported that water
management of individual paddy plots is not carefully conducted
by farmers in Korea, although they have worked to control drainage
by removing or readjusting the drainage height through earth
dikes, when high rainfall is expected or irrigation water and rain-
fall are stored (Yoon et al., 2006). Drainage management is related
to specific types of drainage design facilities. Drainage systems
for wastewater reuse should be replaced by structured weir type
drainage control outlets, to provide effective water management
and labor efficiency for farmers.
The amount of N fertilization was three to four times lower in
this study than in earlier Korean studies (see Table 2; reference 1,
2, 3, 5, 6). These earlier studies reported that nutrient losses ranged
from 57.8 to 113.7 kg N ha−1 and from 1.21 to 3.70 kg P ha−1 due to
surface drainage, whereas Japanese studies (Feng et al., 2004, 2005;
Maruyama et al., 2008) found that nutrient losses in a rice paddy
field were 10.9–41.7 kg N ha−1 and 1.50–2.74 kg P ha−1 by surface
drainage. The nutrient losses in Korea are higher than in Japan, and
the trend is similar for the rate of fertilization. However, the losses
by drainage in this study were much lower, due to the high drainage
outlet height.
In many countries, including Korea, water quality problems in
rural areas are thought to be largely due to excessive fertilization
and the resulting agricultural runoff (Yoon et al., 2006). In addition,

Table 6
Phosphorus balance in the treatment plots.

Year Classification 2006 (kg ha−1 ) 2007 (kg ha−1 ) 2008 (kg ha−1 ) Averagea (kg ha−1 )

Input Fertilizer 45.00 45.00 45.00 45.00

Precipitation 0.21 0.20 0.23 0.22
Irrigation TR#1 0.50 0.78 0.02 0.43
water TR#2 7.13 4.35 1.42 4.30
TR#3 6.65 4.33 0.82 3.94
Subtotal TR#1 45.71 45.98 45.25 45.65
TR#2 52.34 49.55 46.65 49.52
TR#3 51.86 49.53 46.05 49.16
Output Drainage water TR#1 0.52 0.12 0.00 0.21
TR#2 1.28 0.65 0.23 0.72
TR#3 0.60 0.32 0.00 0.31
Infiltration TR#1 1.75 2.60 3.22 2.52
water TR#2 1.45 2.43 2.87 2.25
TR#3 1.62 2.45 2.97 2.35
Crop harvest TR#1 – 9.15 8.33 8.74a
TR#2 – 15.91 13.08 14.50a
TR#3 – 15.73 12.39 14.07a
Subtotal TR#1 2.27b 9.52 10.25 11.47
TR#2 2.73b 18.99 16.18 17.47
TR#3 2.22b 18.50 15.37 16.73
Difference TR#1 – −34.11 −30.47 −32.29a
TR#2 – −30.56 −30.47 −30.52a
TR#3 – −31.03 −30.68 −30.86a
a
The averages in crop harvest and difference were calculated during 2007 and 2008.
b
The subtotals in the output system during 2006 were calculated without crop harvest.
 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243
Fig. 2. Monthly trend of T-N concentrations in the drainage water according to treat-
ment, (a) groundwater (TR#1); (b) wastewater (TR#2); (c) reclaimed wastewater
(TR#3). The blue line represents the monthly change of average T-N concentrations
during 2006–2008.
agricultural nonpoint source pollution occurs due to chemical fer-
tilizer and sediment loss (Cho et al., 2008). As shown in Table 2, the
paddy fields with high fertilization rates show high nutrient loads.
The amount of nitrogen fertilizer generally applied to paddy fields
is from 110.0 to 112.0 kg N ha−1 in Korea; however, in this study,
the amount applied to the reclaimed irrigation paddy field (TR#3) is
only 55.0 kg N ha−1 . Maruyama et al. (2008) reported that the paddy
field has a cleaning or purifying effect. The difference between N
inputs of 138.1 kg ha−1 and crop harvest of 121.4 kg ha−1 in TR#3
was 16.75 kg ha−1 in this study (Table 5). This means that the paddy
system had “purified” 16.75 kg ha−1 of N, reducing potential envi-
ronmental pollution by 16.75 kg ha−1 . This purification potential
creates the possibility for improving water quality degradation of
241
Fig. 3. Monthly trend of T-P concentrations in the drainage water according to treat-
ment, (a) groundwater (TR#1); (b) wastewater (TR#2); (c) reclaimed wastewater
(TR#3). The blue line represents the monthly change of average T-P concentrations
during 2006–2008.
reclaimed wastewater irrigation through fertilizer management
and the uptake efficiency of N.
Kim et al. (2007) reported that T-N and T-P loads from paddy
fields increase by 254% and 534%, respectively, when reclaimed
wastewater is used for irrigation compared to conventional irri-
gation and fertilization. However, the nutrient runoff load from
this paddy field during the rice cultivation period was relatively
low compared to other studies (Table 2). Raising the drainage out-
let height in ponded paddy fields can increase rainwater storage
by reducing excess flow from the field and reducing rainfall loss
by surface drainage (Yoon et al., 2003). This contributes to the T-
N and T-P losses from the field by decreasing inflow irrigation of
the high nutrient concentration in wastewater irrigation. Bouman
and Tuong (2001) reported that reducing the ponded water depth
242 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243
to within 5–10 cm of the soil saturation level did not reduce land
productivity, and they found that a 23% water savings resulted in
only a 6% reduction in yield. Therefore, accurate and timely water
delivery is particularly important in reclaimed irrigation areas.
4. Conclusions
This study assesses the water and mass balances of reclaimed
wastewater irrigation for rice cultivation. The results show that
about 58% of the total water inflow in each treatment was con-
tributed by precipitation and the remainder by irrigation. Water
outflow generally balanced the inflow, with about 12–13% of the
total outflow going to surface drainage, 57–58% to infiltration, and
30% to evapotranspiration. The average drainage amount was far
lower in this study than in other studies, due to the high drainage
outlet height.
The N in the irrigation water was about two times higher in TR#3
than in TR#1. The N in the crop harvest averaged 59.62 kg ha−1
for TR#1 and 121.35 kg ha−1 for TR#3. The N in the fertilizer was
a large proportion of the T-N in the paddy field irrigated with
groundwater (TR#1), while the N in the irrigation water and the
fertilizer were both major inputs in the paddy field irrigated with
reclaimed wastewater (TR#3). The major P input was fertilizer in
TR#3, and crop harvest was also a main output in the P balance.
Surface drainage and infiltration in this study were relatively small
components.
The effects of the high nutrient concentration in the reclaimed
irrigation water in TR#3 on rice yield and plant growth were mon-
itored, and results indicated no adverse effects.
The difference between the N inputs and the crop harvest in
TR#3 was 16.75 kg ha−1 , which means that the paddy system has
reduced potential pollution to the environment. In addition, the P
outputs in this study were larger than the inputs. This indicates
that it is possible to reduce the P fertilization rate. Further research
to investigate the proper fertilization management practices of
reclaimed wastewater irrigation is currently being conducted.
In general, this study showed that reclaimed wastewater
irrigation with appropriate water and fertilizer managements
is potentially useful for sustainable rice cultivation, although
long-term monitoring is needed to determine its effects on soil
characteristics and health risks. In addition, indirect discharge from
the wastewater treatment plant along paddy watercourses can con-
tribute to improved water quality compare to direct discharge to
streams, due to the purification capacity of paddy fields.
Acknowledgements
This project was supported by a grant (code# 4-5-3) from the
Sustainable Water Resources Research Center of the 21st Century
Frontier Research Program. Funding of this research was also pro-
vided by the Technology Development Program for Agriculture
and Forestry, Ministry for Food, Agriculture, Forestry and Fisheries,
Republic of Korea. We gratefully acknowledge the constructive
comments of the editor and two anonymous reviewers.
References
Al-Lahham, O., El Assi, N.M., Fayyad, M., 2003. Impact of treated wastewater irri-
gation on quality attributes and contamination of tomato fruit. Agric. Water
Manage. 61, 51–62.
Allen, R.G., Pereira, L.S., Raes, D., Smith, M., 1998. Crop Evapotranspiration. Guide-
lines for Computing Crop Water Requirements. FAO Irrigation and Drainage
Paper no. 56, Rome, Italy.
APHA, 1995. Standard Methods for the Examination of Water and Wastewater.
American Public Health Association, Washington DC.
Asano, T., Levine, A.D., 1996. Wastewater reclamation, recycling and reuse: past,
present, and future. Water Sci. Technol. 33, 1–14.
Bouman, B.A.M., Tuong, T.P., 2001. Field water management to save water and
increase its productivity in irrigation lowland rice. Agric. Water Manage. 49,
11–30.
Brunetti, G., Senesi, N., Plaza, C., 2007. Effects of amendment with treated and
untreated olive oil mill wastewaters on soil properties, soil humic substances
and wheat yield. Geoderma 138, 144–152.
Chapman, H.D., Pratt., P.F., 1961. Methods of Analysis for Soils, Plants and Waters.
University of California, Division of Agricultural Sciences, CA.
Cho, J.Y., 2003. Seasonal runoff estimation of N and P in a paddy field of central Korea.
Nutr. Cycl. In Agroecosyst. 65, 43–52.
Cho, J.Y., Choi, J.K., 2001. Nitrogen and phosphorus losses from a broad paddy field
in central Korea. Commun. Soil Sci. Plant Anal. 32, 2395–2410.
Cho, J.Y., Han, K.W., 2002. Nutrient losses from a paddy field plot in central Korea.
Water Air Soil Pollut. 134, 215–228.
Cho, J.Y., Han, K.W., Kim, J.H., Son, J.K., Yoon, K.S., 2003. Distribution and sources
of PAHs in Saemangeum reclaimed tidal lands of central Korea. Bull. Environ.
Contam. Toxicol. 71, 182–188.
Cho, J.Y., Park, S.W., Son, J.G., Park, B.J., Lee, Y.G., 2006. Rice growth response and
soil quality by domestic wastewater irrigation on rice paddy field. J. Korean Soc.
Rural Plan. 12, 49–56 (in Korean, with English abstract).
Cho, J.Y., Son, J.G., Choi, J.K., Song, C.H., Chung, B.Y., 2008. Surface and subsurface
losses of N and P from salt-affected rice paddy fields of Saemangeum reclaimed
land in South Korea. Paddy Water Environ. 6, 211–219.
Cooper, R.L., 1991. Public health concerns in wastewater reuse. Water Sci. Technol.
24, 55–65.
De Datta, S.K., Abilay, W.P., Kalwar, G.N.,1973. Water stress effects in flooded tropi-
cal rice. In: Water Management in Philippine Irrigation Systems: Research and
Operations. International Rice Research Institute, Los Banos, Philippines, pp.
19–36.
Eom, K.C.,2001. Environmentally beneficial function of rice culture and paddy soil.
In: Rice Culture in Asia. International Commission on Irrigation and Drainage,
and Korean National Committee on Irrigation and Drainage, Korea.
Feng, Y.W., Yoshimaga, I.Y., Shiratani, E., Hitomi, T., Hasebe, H., 2005. Nutrient bal-
ance in a paddy field with a recycling irrigation system. Water Sci. Technol. 51,
151–157.
Feng, Y.W., Yoshimaga, I.Y., Shiratani, E., Hitomi, T., Hasebe, H., 2004. Characteris-
tic and behavior of nutrients in a paddy field area equipped with a recycling
irrigation system. Agric. Water Manage. 68, 47–60.
Han, K.W., Cho, J.Y., Choi, J.K., 1999. Annual runoff loading of nitrogen and phospho-
rus from a paddy field. Agric. Chem. Biotechnol. 42, 29–23.
Harris, G., 1998. An analysis of global fertilizer application rates for major crops. In:
IFA Annual Conference, Toronto, Ontario, Canada.
Hukkeri, S.B., Sharma, A.K., 1980. Water-use efficiency of transplanted and direct-
sown rice under different water management practices. Indian J. Agric. Sci. 50,
240–243.
Jackson, M.L., 1967. Soil Chemical Analysis. Prentice-Hall, New Delhi.
Jang, T.I., 2009. Environmental effects of reclaimed wastewater irrigation on paddy
fields. Doctoral Dissertation. Seoul National University, Seoul, Korea (in Korean
with an English abstract).
Jang, T.I., Kim, H.K., Im, S.J., Park, S.W., 2010a. Simulations of storm hydrographs in a
mixed-landuse watershed using a modified TR-20 model. Agric. Water Manage.
97, 201–207.
Jang, T.I., Lee, S.B., Sung, C.H., Lee, H.P., Park, S.W., 2010b. Safe application of
reclaimed water reuse for agriculture in Korea. Paddy Water Environ. 8,
227–233.
Jang, T.I., Park, S.W., Kim, H.K., 2008. Environmental effects analysis of a wastewater
reuse system for agriculture in Korea. Water Sci. & Technol. WS 8, 37–42.
Jung, K.W., 2011. Effects of reclaimed wastewater irrigation on paddy rice growth
and yield. Mater thesis. Seoul National University, Seoul, Korea (in Korean with
an English abstract).
Kang, M.S., Kim, S.M., Park, S.W., Lee, J.J., Yoo, K.H., 2007. Assessment of reclaimed
wastewater irrigation impacts on water quality, soil, and rice cultivation in
paddy fields. J. Environ. Sci. Health A 42, 439–445.
Kim, H.K., Jang, T.I., Im, S.J., Park, S.W., 2009. Estimation of irrigation return flow from
paddy fields considering the soil moisture. Agric. Water Manage. 96, 875–882.
Kim, S.M., Im, S.J., Park, S.W., Lee, J.J., Benham, B.L., Jang, T.I., 2008. Assessment of
Wastewater reuse effects on nutrient loads from paddy field using field-scale
water quality model. Environ Model. Assess. 13, 305–313.
Kim, S.M., Park, S.W., Lee, J.J., Benham, B.L., Kim, H.K., 2007. Modeling and assessing
the impact of reclaimed wastewater irrigation on the nutrient loads from an
agricultural watershed containing rice paddy fields. J. Environ. Sci. Health A 42,
305–315.
Kiziloglu, F.M., Turan, M., Sahin, U., Angin, I., Anapali, O., Okuroglu, M., 2007. Effects of
wastewater irrigation on soil and cabbage-plant (Brassica olerecea var. Capitate
cv. Yalova-1) chemical properties. J. Plant Nutr. Soil Sci. 170, 166–172.
Maeng, S.K., Ahn, K.H., Kim, K.P., Song, K.G., Park, K.Y., 2006. Compressible synthetic
dual-medium filtration of wastewater effluents for water reuse. Water Pract.
Technol., 1, doi:10.2166/wpt.2006.050.
Maruyama, T., Hashimoto, I., Murashima, K., Takimoto, H., 2008. Evaluation of N and
P mass balance in paddy rice culture along Kahokugata Lake, Japan, to assess
potential lake pollution. Paddy Water Environ. 6, 355–362.
MLTM, 2006. Water Vision 2020 (2006 National Water Resources Plan Update). Min-
istry of Land, Transportation and Maritime Affaris (MLTM). Kyeonggi-do, Korea.
Available at: www.waterplan.go.kr (accessed 02.04.09).
MOE, 2005. Recommended Water Quality Guidelines for Wastewater Reuse. Min-
istry of Environment (MOE). Gyeonggi-do, Korea.
 T.I. Jang et al. / Agricultural Water Management 104 (2012) 235–243 243

NIAST, 2000. Taxo-nomical Classification of Korean Soils. National Institute of Agri-
cultural Science and Technology (NIAST), Suwon, Korea.
Odhiambo, L.O., Murty, V.V.N., 1996. Modeling water balance components in rela-
tion to field layout in lowland paddy fields (I). Agric. Water Manage. 30,
185–199.
Palese, A.M., Pasquale, V., Celano, G., Figliuolo, G., Masi, S., Xiloyannis, C., 2009. Irriga-
tion of olive groves in Southern Italy with treated municipal wastewater: effects
on microbiological quality of soil and fruits. Agric. Ecosyst. Environ. 129, 43–51.
Seol, S.I., Lee, S.M., Han, G.H., Choi, W.J., Yoo, S.H., 2000. Urea transformation and
nitrogen loss in Waterlogged soil column. Agric. Chem. Biotechnol. 43, 86–93.
Yoon, C.G., 2009. Wise use of paddy rice fields to partially compensate for the loss
of natural wetlands. Paddy Water Environ. 7, 357–366.
Yoon, C.G., Ham, J.H., Jeon, J.H., 2003. Mass balance analysis in Korean paddy rice
culture. Paddy Water Environ. 1, 99–106.
Yoon, K.S., Cho, J.Y., Choi, J.K., Son, J.G., 2006. Water management and N, P losses
from paddy fields in Southern Korea. J. Am. Water Res. Assoc. 42, 1205–1216.