In Japan, the amount of sewage sludge (SS) generated each year is continuously increasing; in 2002, it reached 428 M m-3 (Japan Sewage Works Association, 2003). The development of recycling techniques for SS has decreased the rate of landfill, sea dumping and incineration and has promoted the use of sludge as a fertilizer and a construction material. Since SS is generally rich in N, P and other plant nutrients, it is desirable to utilize SS in farmlands from the viewpoint of conserving resources and the environment.
Several studies have reported an increase in microbial and enzymatic activities in surface soil after a single or repeated application of organic wastes (Albiach et al., 2000; Zaman et al., 2002). The application of SS may also have beneficial effects on soil physiochemical and biological properties. However, the application of SS to farmlands has received considerable attention since it not only contains nutrients for plants but also sometimes has a significant level of heavy, potentially toxic metals such as Zn, Cu, Pb and Cd; these pose a risk to human health through accumulation in the soil and food chain (Wilkinson et al., 2001). It has also been reported that heavy metal concentrations in SS depend on regional characteristics. Thus, SS from a region with a low concentration of heavy metals can be applied to farmlands with less risk.
With regard to fertilization, organic materials can be potentially important sources of N in crop production. N is often the most limiting element for plant growth and quality. Therefore, it is important to estimate the amount of N supply and plant N uptake from the applied organic materials. However, in comparison to chemical fertilizers, it is difficult to appropriately control the amount of nutrient supply to crops from organic materials during cultivation; this is because this amount depends on the rate of decomposition and the mineralization processes of organic materials through which nutrients are made available to plants in inorganic or low-molecular-weight forms. Mineralization is a well-known complex process that is regulated by many environmental factors such as soil physicochemical properties, temperature, soil moisture and biota (Oyanagi et al., 2002). Several studies have been conducted on N mineralization in SS (Matsuoka et al., 2006; Zaman et al., 2002, 2004), N and P uptake (Moritsuka et al., 2006) and plant N uptake separately from soil and SS by using δ15N (Goto et al., 1996). However, information on the rate of plant N uptake from SS (N use efficiency) is limited.
In this study, we investigated and discussed (1) the fate of N derived from the applied SS and soil separately in the cultivation of rice by using the indirect 15N approach and (2) the effects of SS on the growth indices of rice, microbial activity and soil chemical properties.
MATERIALS AND METHODS
Soil, Treatment and Plant Cultivation
A green house experiment was conducted at the University Farm, Ehime University,
Matsuyama, Japan. Wagner pots with an area of 0.02 m2 were used.
Each pot contained 3.37 kg of dry soil (Soil Taxonomy: Typic Udorthents; 1.6%
clay, 3.5% silt, 94.9% sand; pH (H2O) 6.2; electrical conductivity
(EC): 0.065 dS m-1; total carbon (TC): 1.32%; total nitrogen (TN):
0.09%; Bray-2 extractable P: 30 mg kg-1; exchangeable K: 0.13 cmol
kg-1; exchangeable Ca: 18.1 cmol kg-1 and exchangeable
Mg: 0.75 cmol kg-1).
The experiment included three N fertilization treatments: (1) no fertilizer (NF) application, (2) application of only the chemical fertilizer (CF) NH4Cl and (3) application of sewage sludge (SS) fertilizer). SS was obtained from Nishida Industry Inc., Ehime, Japan and its chemical properties were as follows: pH (H2O) 6.1, 5.8 C/N, 42.3% TC, 7.3% TN, 2.5% P, 0.5% K, 0.8% Ca, 0.9% Mg, 49 mg kg-1 Zn, 720 mg kg-1 Cu, 0.43 mg kg-1 Hg, 5.6 mg kg-1 As and 0.4 mg kg-1 Cd.
In CF-treated pots, N was applied to the soil at the rate of 4, 3 and 3 g m-2 of N as 15NH4Cl (0.942 atom%) at the basal fertilization (June 10, 2004), tillering (June 29, 2004) and young panicle formation stages (July 27, 2004), respectively. On June 10, P and K were applied to the soil at the rate of 4 g m-2 of P as fused magnesium phosphate and 8 g m-2 K as KCl. In SS-treated pots, SS was applied to the soil at the rate of 208 g m-2 (= 15, 5 and 1 g m-2 of N, P and K, respectively) and a solution containing 0.1 g m-2 of 15NH4Cl (99.8 atom%) was carefully injected into the soil as 15N tracer. K was supplemented to the soil at the rate of 8 g m-2 of K as KCl.
On June 10 in 2004, after the application of CF, SS was mixed well in the soil and the pots were flooded with tap water; then three seedlings per pot of rice-Oryza sativa L. cv. Koshihikari-were transplanted in each pot. The cultivation was conducted in the greenhouse with eight replicates per treatment.
Sampling and Chemical Analysis
The rice growth (plant height, number of tillers and leaf chlorophyll content
index) was measured at 0, 17, 30, 48, 58, 74 and 116 days after transplanting
(DAT). The chlorophyll content index was measured with a chlorophyll meter (SPAD-502;
Minolta Co. Ltd., Japan). Total microbial activity that was evaluated by measuring
the hydrolysis of fluorescein diacet ate (FDA) was measured at 0, 7, 15, 34,
66 and 106 DAT as described by Adam and Duncan (2001).
Plant tops and roots were collected at harvest, oven-dried at 70°C and ground into powder. The total N concentration in the subsamples of the rice plants in the NF treatment was determined using an NC analyzer (Sumigraph NC-80; SCAS, Japan). The total N concentration and abundance of 15N in the subsamples of the plants in the CF and SS treatments were determined using a stable isotope mass spectrometer (ANCA-SL; Europa Scientific Co. Ltd., Cheshire, UK). Soil pH (H2O) and EC were measured with a pH and conductivity meter (D-24; Horiba Ltd., Japan), respectively, using a suspension of soil and deionized water in a ratio of 1:5 (shaken on a reciprocal shaker for 30 min).
To measure P, K, Ca, Mg, Zn and Cu concentrations, the subsamples of the soil and plants were digested with HNO3 and HClO4. After digestion, the filtrate was subjected to P analysis by using the molybdate blue method and to K, Ca, Mg, Zn and Cu analyses by atomic absorption spectrometry (AA-6200; Shimadzu Co. Ltd., Japan).
Estimation of N Dynamics by Using the Indirect 15N Method
To estimate the fate of N derived by the rice plants from SS, we employed
the A value method-one of the indirect 15N approaches. It is assumed
that the available nutrient in the soil (A value) definition implies that when
two sources of a given nutrient are present in the soil, the plant will absorb
from each of these sources in proportion to the respective quantities available
(Fried and Dean, 1952). The A value is a time-integrated estimate of
the plant-available N pool (Stevenson et al., 1998). In the CF treatment
(control), 15N-labeled CF was applied. In the organic matter treatment
(SS), 15N tracer and unlabeled organic matter were applied.
The A value of the soil, i.e., AS, in the CF treatment can be determined from Eq. 1.
where: % Ndfcf is the percentage of plant N uptake from CF applied in the CF treatment and calculated as follows:
= (15N atom% excess in plant/15N atom% excess in
The A value of the chemical fertilizer (ACF) is the amount of 15N-labeled
fertilizer applied in the CF treatment.
The A value of the soil and organic matter (AS + OM) in the SS treatment can be determined from Eq. 3.
where %Ndft is the percentage of plant N uptake from the 15N tracer in the SS treatment and is calculated as follows:
= (15N atom% excess in plant/15N atom% excess in
15N tracer) x 100
The A value of 15N tracer (AT) is the amount of 15N
tracer applied in the SS treatment.
It is assumed that the A value of the soil is constant, regardless of the quantity
of material added to the soil. Therefore, the A value of the organic matter
(AOM) is calculated by subtracting AS from AS + OM.
The percentage of plant N uptake from the organic matter applied in the SS
treatment (%Ndfom) can be calculated from Eq. 6.
The fertilizer use efficiency (FUE) is the recovery rate of N by rice from
the applied N source. The FUE rates of the CF (FUECF) and organic
matter (FUEOM) were calculated from Eq. 7 and 8,
We also computed the relative efficiency-the relative uptake rate of organic matter N to that of CF N-according to Nishida et al. (2004) as described in Eq. 9.
First, the normality of the data was assessed using the chi-square test
for goodness of fit and the differences among means were then analyzed by Tukeys
test or the Steel-Dwass test by using the software KyPlot (KyensLab Inc., Japan).
Effects on Rice Growth and Dry Weight
The plant height, number of tillers and leaf chlorophyll content index in
the SS and CF treatments were higher than those in the NF treatment during cultivation
and significant differences were observed (Fig. 1). The plant
height in the SS treatment was significantly lower than that in the CF treatment
at 17, 48, 58, 74 and 116 DAS, while the number of tillers in the SS treatment
was significantly lower than that in the CF treatment at 17 and 58 DAT. The
change in the leaf chlorophyll content index in the SS treatment was similar
to that in the CF treatment. Eventually, at harvest, the dry weights of the
plants (tops and roots) in the CF and SS treatments were significantly higher
than those in the NF treatment by 12.3- and 8.4-fold, respectively (Table
1). This value in the SS treatment was significantly lower than that in
the CF treatment by 0.7-fold. These results indicate that the application of
SS and CF apparently increased the growth indices and yield of rice, although
the application of SS in this study resulted in lower levels of growth and yield
than in the CF treatment. This was despite the fact that the amount of N applied
in SS was 1.5 times that in CF. The dry weight of plant tops and roots differed
widely among the three treatments and showed significant differences. The ratios
of the dry weight of the plant top to that of the root in the NF, CF and SS
treatments were 2.2, 2.0 and 2.8, respectively. This ratio was the highest in
the SS treatment.
||Change in the rice growth (plant height (A), No. of tillers
(B), leaf chlorophyll content index (C)) Period bars denote standard error.
Symbols with the same letter in each time indicate no significant differences
(p = 0.05)
||Comparison of the rice (Oryza sativa L. cv. Koshihikari)
dry weight and N uptake from fertilizer and soil at harvest
|Ndff and Ndfs refer to N derived from the
applied chemical fertilizer or sewage sludge, from the soil, respectively.
Values in parentheses indicate percentages of N uptake from the respective
pools to the total plant N. Values with the same letter in each column indicate
no significant differences (p = 0.05)
Plant N Uptake from the SS, Cf and Soil
We calculated the amount of N taken up by rice from the CF, SS and soil
(Table 1), by using the indirect 15N method. The
amount of plant N uptake in the CF and SS treatments was significantly higher
than that in the NF treatment; the value in the SS treatment was significantly
lower than that in the CF treatment. These were in accordance with the results
of dry weight (Table 1). In the CF treatment, N uptake from
CF (0.137 g pot-1) was higher than that from the soil (0.054 g pot-1).
In the SS treatment, N uptakes from SS and soil were 0.130 and 0.017 g pot-1,
respectively. N uptakes from the soil were lower than those from the fertilizers
in the CF and SS treatments. The ratios of N uptake from the soil to the total
plant N were 28.3% in the CF and 11.7% in the SS treatments.
The amount of N uptake from SS (0.130 g pot-1) was 0.95 times lower than that from CF (0.137 g pot-1), but no significant difference was observed. On the other hand, N uptake from the soil in the CF treatment was significantly higher than that in the NF and SS treatments and the amounts of N taken up from the soil in the NF, CF and SS treatments were 0.009, 0.054 and 0.017 g pot-1, respectively. The ratios of N uptake from the soil in the CF and SS treatments to that from the soil in the NF treatment were 5.97 and 1.92, respectively.
Table 2 shows the A value and the FUE estimated using the indirect 15N method. The A values of the soil, CF and SS applied were estimated to be 79, 200 and 588 mg pot-1, respectively. The A values of CF and SS were greater than that of the soil by 2.53- and 7.44-fold, respectively. The available N in SS during cultivation was 2.94 times higher than that in CF.
The rates of N uptake derived from the A values of the CF, SS and soil differed widely among all treatments (Table 2). The rates in the NF, CF and SS treatments were 11.9, 68.3 and 22.2, respectively and showed significant differences. The rate of N uptake in the SS treatment was 0.33- and 1.87- fold that in the CF and NF treatments, respectively.
The N use efficiencies of CF and SS were 68.3 and 43.0 of the applied N, respectively (Table 2). The rate in the CF treatment was significantly higher than that in the SS treatment. Therefore, the relative efficiency of SS to CF was 62.9%.
Fate of Fertilizer N
Nitrogen distributions from CF and SS at rice harvest are illustrated in
Table 3. The N in CF and SS were distributed to plant uptake
(68.3 and 43.0%), N that remained in the soil (5.1 and 14.1%) and unaccounted
for N (26.6 and 42.9%). The rate of N uptake from SS was lower than that from
CF. The rates of N that remained in the soil and unaccounted for N in the SS
treatment were higher than the corresponding rates in the CF treatment. Therefore,
the rates of plant N uptake, N that remained in the soil and unaccounted for
N in the SS treatment were 0.63, 2.76 and 1.61 times greater, respectively,
than the corresponding rates in the CF treatment.
||The soil and fertilizer A values, the ratio of rice (Oryza
sativa L. cv. Koshihikari) N uptake and N use efficiency, and the relative
efficiency of the chemical fertilizer, sewage sludge, and no fertilizer
|Percentage of the amount of N uptake in the A
value (fertilizer + soil) of no fertilizer, chemical fertilizer and sewage
sludge treatments. Percentage of the amount of N uptake from the applied
N. §The relative value of the N use efficiency rate in the sewage sludge
application to that in the chemical fertilizer application. Values with
the same letter in each column indicate no significant differences (p =
|| Comparison of N distribution from the chemical fertilizer
and sewage sludge at harvest
|Values with the same letter in each column indicate no significant
differences (p = 0.05)
Microbiological and Chemical Properties of the Soil
The effect of application of SS on soil microbial activity was evaluated
by measuring the hydrolysis of FDA. In each soil treatment, microbial activity
at a later growth stage (from the maximum tiller number stage to the ripening
stage 34, 66 and 106 DAT; Fig. 2B) was higher than that at
the early growth stage (initial growing stage -7 and 15 DAT; Fig.
2A). The activity in the SS-treated soil tended to be higher than that in
the NF- and CF-treated soil during cultivation and the value in the SS-treated
soil at the later growth stage was significantly higher than that in the NF-treated
soil. The SS application activated soil microorganisms and their metabolisms.
The pH of SS-treated soil (7.51) was higher than those of NF- (7.16) and CF-treated
(7.02) soils and they were significantly different from each other (Fig.
3). The EC of CF-treated soil (0.780 dS m-1) was significantly
higher than those of NF- (0.263 dS m-1) and SS-treated (0.275 dS
||Comparison of soil microbiological activity determined by
fluorescein diacetate esterase activity. Means of the activities are separately
indicated in (A) at early growth stage (7 and 15 days after transplanting
(DAT)) and in (B) at later growth stage (34, 66 and 106 DAT). Bars denote
standard error. Columns with the same letter in each stage indicate no significant
differences (p = 0.05)
||Comparison of soil pH (A) and electrical conductivity (EC)
(B) at harvest. Bars denote standard error. Columns with the same letter
indicate no significant differences (p = 0.05)
||Comparison of the mineral concentration in the rice stems
and leaves at harvest. Bars denote standard error. Columns with the same
letter in each mineral indicate no significant differences (p = 0.05)
The EC value of the CF-treated soil was 2.97 and 2.84-fold the EC values of
the NF- and SS-treated soils, respectively. There was no significant difference
in the EC values between the NF- and SS-treated soils.
At harvest, no significant differences in the concentrations of P, Ca, K, Mg, Zn and Cu in the soil were observed among the three treatments (data not shown). The concentration of Ca in the stems and leaves of rice in the NF treatment was significantly higher than that in the CF and SS treatments (Fig. 4), although the concentrations of P, K, Mg, Zn and Cu in the rice stems and leaves collected at harvest showed no significant difference among the three treatments. In this study, the contamination of soil and rice stems and leaves with Cu and Zn derived from SS was negligibly low.
Plant Growth and Fate of Fertilizer N
Because of the low fertility of the soil, the rice growth and N uptake in
the NF treatment were the lowest among the three treatments (Fig.
1 and Table 1). It is assumed that in the NF treatment,
macronutrient limitation led to the poor growth of roots and possibly resulted
in an extremely low amount of N uptake. Based on the above assumption, the A
value and N uptake by rice in the SS treatment were calculated from the distribution
of tracer N in the CF but not in the NF treatment.
The plant growth, dry weight and the amount of N uptake in the SS treatment
were markedly and significantly higher than those in the NF treatment. Therefore
SS could supply sufficient amounts of nutrients to the crop. However, the number
of tillers in the SS treatment was significantly lower than that in the CF treatment
at 17 DAT (Fig. 1A and B). This may be caused
by a slower N mineralization from the SS than that from the CF at the early
growth stage of rice or by the decrease in the redox potential of soil following
harmful metabolite production during soil microbial fermentation under anaerobic
conditions as described by Tanaka and Ono (2000). Moroyu et al. (1981)
also reported that the increase in rice height and tiller number was retarded
by the rice or barley straw applications.
The A value of SS (588 mg pot-1) was higher than that of CF (200 mg pot-1); however, the rate of N uptake in the A value of SS (22.2%) was significantly lower than that in the A value of CF (68.3%) (Table 2). Consequently, the amount of N uptake in the SS treatment was significantly lower than that in the CF treatment (Table 1). This indicates the relative lower capability of the plant to take up available N in the SS-treated soil than that in the CF-treated soil. This may be caused by the negative effects of the organic matter application on root growth in the SS-treated soil, which is demonstrated by the results that the dry weight of root was lower and the top/root ratio was higher in the SS-treated soil than in the CF-treated soil (Table 1). On the other hand, the relative higher capability of the plant to take up available N in the CF-treated soil may be caused by the positive effects of the CF split application on N use efficiency.
The N use efficiencies of CF and SS were estimated to be 68.3 and 43.0%, respectively (Table 2). Therefore, the relative efficiency of SS to CF was 62.9%. This result suggested from a simple calculation that 1.59 times of N application by SS would be required when compared with that by the CF in order to equalize the uptake amount of N from CF and SS. Recent studies reported widely diverse relative efficiencies of organic fertilizers in paddy soil determined by the 15N direct method, i.e., 70% in rape cake (Uenosono et al., 2004), 16-20% in cattle manure compost, 81% in poultry manure compost and 71% in swine feces (Nishida et al., 2004).
In this study, the recovery rates of N by rice from the applied CF and SS were 68.3 and 43.0%, respectively (Table 3). The recovery rates of N by rice from applied ammonium sulphate and Sesbania aculeata L. were 12.5 and 19.3%, respectively (Azam, 1990). Further, the recovery rates of N by rice were 5.6-7.1% from cattle manure compost, 29.2% from poultry manure compost and 25.4% from swine feces (Nishida et al., 2004) and 24.6% from rapeseed pod-wall, 19.1% from rice straw, 14.9% from cattle manure compost and 7.7% from rice straw compost (Ueno and Yamamuro, 2001). The higher recovery rates of fertilizer N in our experiment may be attributed to the fact that the used soil had considerably low fertility, as shown in the NF-treated pot.
The rates of unaccounted for N may be attributed to losses from the soil-plant system due to NH3 volatilization and/or denitrification. N losses from SS were higher than those from CF. This may be because the higher soil pH (7.51) in the SS treatment led to the higher rate of NH3 volatilization.
Microbiological and Chemical Properties of the Soil
The microbiological activity in the SS-treated soil tended to be higher
than that in the CF-treated soil (Fig. 2). This was in agreement
with the results of previous studies on soil enzyme activity (Albiach et
al., 2000; Zaman et al., 2002) and biomass specific respiration (Aoyama
et al., 2006). The pH of SS-treated soil after harvest was higher than
those of the NF- and CF-treated soils (Fig. 3A). The application
of SS probably increased the release amount of ammonia, exchangeable Ca and
Mg and resulted in the higher pH. The EC of CF-treated soil after harvest was
significantly higher than that of SS-treated soil (Fig. 3B).
It was considered that the excess amount of CF remained in the soil as inorganic
forms. This result was in agreement with the result of a previous study (Zaman
et al., 2004).Care should be taken to avoid pollution of water resources
from the nitrate and nutrient salts derived from the CF.
The concentrations of P, K, Mg, Zn and Cu in the rice stems and leaves at harvest showed no significant differences among the three treatments, although the concentration of Ca in the stems and leaves of the plants in the NF-treated soil was significantly higher than that in the plants in the CF- and SS-treated soils (Fig. 4). The reason for the differences between the plants with regard to Ca uptake is not clear, but it may be related to the relative concentration of Ca in the soil, mycorrhizal effects, or to physiological changes in the root.
Based on these findings, we conclude that SS with a low concentration of heavy metals can be regarded as a useful organic fertilizer from the viewpoint of nutrient supply for soil fertilization and nutrient recycling in the environment; however, further studies are necessary such as the effect of SS on root activity and the evaluation of soil chemical and biological properties in long-term experiments with several soils, crops, climates and soil management techniques, e.g., split application.
This study was supported by the Grant-in-Aid for Scientific Research by the Ministry of Education, Science and Culture, Sports, Science and Technology of Japan (No. 14360041).