SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 15 February 2008
Published in Soil Sci Soc Am J 72:514-523 (2008)
DOI: 10.2136/sssaj2006.0403
© 2008 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Agricola
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Related Collections
Right arrow Rice
Right arrow Residue management

SOIL FERTILITY & PLANT NUTRITION

Nitrogen Supply in Rice-Based Cropping Systems as Affected by Crop Residue Management

Nguyen Hong Thuya, Yuhua Shanb, Bijay-Singhc, Kairong Wangd, Zucong Caie, Yadvinder-Singhc and Roland J. Bureshf,*

a Tien Giang Dep. of Science and Technology, 39-Hung Vuong St., My Tho City, Tieng Giang Province, Vietnam
b College of Environ. Science and Eng., Yangzhou Univ., Yangzhou 225009, China
c Punjab Agricultural Univ.,Ludhiana 141004, India
d Institute of Subtropical Agriculture, Chinese Academy of Sciences, Changsha 410125, China
e Institute of Soil Science, Chinese Academy of Sciences, P.O. Box 821, Nanjing 210008, China
f International Rice Research Institute (IRRI), DAPO Box 7777, Metro Manila, Philippines

* Corresponding author (r.buresh{at}cgiar.org).


   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Concerns regarding the detrimental effects of burning crop residues on human health and the environment have increased interest in alternative uses of crop residues. We examined the in situ use of crop residue as a source of supplemental N for succeeding crops in rice (Oryza sativa L.)-based cropping systems at three sites during 3 yr. The experiments included a rice–wheat (Triticum aestivum L.) rotation at Yixing, Jiangsu Province, China; a rice–wheat rotation at Ludhiana, Punjab, India; and double-rice cropping at Taojiang, Hunan Province, China. The supply of N from crop residues was assessed in the absence of fertilizer N as the difference in total plant N between plots with and without residue. At Yixing, incorporation of wheat residue before rice significantly increased the N supply to the rice by 14 kg N ha–1 averaged across 3 yr. At Ludhiana, incorporation of rice residue before wheat reduced the N supply by 3 kg N ha–1 to the wheat, but increased the N supply by 5 kg N ha–1 to the rice crop following the wheat. In all cases, the return of crop residues had no net benefit on crop yield when fertilizer N was supplied at rates sufficient to eliminate N deficiency. The incorporation of crop residues did not increase the N supply to the succeeding crop during its vegetative growth phase, but the N supply to the crop at later growth stages was often increased. Adjustments in the timing and rate of fertilizer N are probably necessary to optimally supply N to crops receiving residues.

Abbreviations: AEN, agronomic efficiency for applied fertilizer nitrogen • DAS, days after sowing • DAT, days after transplanting • INS, indigenous nitrogen supply • LCC, leaf color chart


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
High yields and intensive land use in irrigated rice-based cropping systems result in production of huge quantities of crop residues, which Asian farmers often burn. The field burning of crop residues is a major contributor to reduced air quality and human respiratory ailments in intensive rice-production areas of Asia, especially in northwestern India and portions of China (Streets et al., 2003). An immediate priority is implementing uses of crop residues that avoid burning. One option is incorporation of residues into fields to supply subsequent crops with essential plant nutrients, thereby saving fertilizer, and to supply C, thereby maintaining or building soil fertility (Yadvinder-Singh et al., 2005).

Nitrogen is typically the nutrient most limiting crop production in rice-based systems. Crop residue is a poor-quality source of N, and the incorporation of crop residue with a high C/N ratio into either aerobic or submerged soil typically results in microbial N immobilization and a temporary decrease in plant-available N. This initial period of several weeks of net N immobilization is followed by net N mineralization (Yadvinder-Singh et al., 2005). The duration of net N immobilization and the net supply of N from crop residue to a subsequent crop depend on the decomposition rate, residue quality, and environmental conditions. The application of crop residue in the absence of fertilizer N can depress total N accumulation and yield of a crop (Thuy, 2004). Long-term incorporation of rice residue can increase readily mineralized organic soil N, suggesting the potential after several years for reducing fertilizer N rates for optimal rice yield (Bird et al., 2001; Eagle et al., 2001).

In rice–wheat cropping systems of India, wheat residue is often fed to animals; but rice residue is considered poor feed for animals due to its high silica content, and it is often burnt by farmers (Mandal et al., 2004). Alternatives to burning are consequently needed more for rice than wheat residues. The incorporation of rice residue before wheat planting is challenging for farmers because of the short interval between rice harvest and wheat planting. Yadvinder-Singh et al. (2004) found no adverse effect on wheat yield when rice residue was incorporated at least 10 d and preferably 20 d before sowing wheat. Nitrogen release from rice residue ranged from 6 to 9 kg N ha–1 during the wheat season (Yadvinder-Singh et al., 2004). Yadvinder-Singh et al. (2005) concluded, based on a review of literature, that application of rice residue to wheat typically has a small or even negative effect on wheat yields during the short term of 1 to 3 yr.

Zeng et al. (2001, 2002), in an examination of diverse rice–rice, rice–wheat, rice–rape (Brassica napus L. var. napus), rice–rice–wheat, and rice–rice–rape cropping systems in China, reported that residues were typically mulched in the wheat and rapeseed seasons and incorporated in the rice season. In a general study of 107 sites throughout main agricultural areas in China, the return of crop residues increased the supply of available soil N by 2.6 mg N kg soil–1 (Zeng et al., 2002). Chinese scientists rather consistently have reported positive effects of crop residues on crop yields and soil fertility in rice-based systems. For example, rice residue applied to wheat in rice–wheat systems or to rice in rice–rice systems reportedly increased yield by 0.2 to 0.8 Mg ha–1 (Li et al., 2003; Yang et al., 2003; Zhu et al., 2004). The effect of crop residues on rice yield is often influenced by fertilizer application, but relatively little is known about the optimal timing of fertilizer N and the feasible savings in fertilizer N when crop residues are applied. The effective retention and incorporation of rice residue in the brief interval between two rice crops remains a challenge because of probable adverse effects on early growth of the second rice crop established soon after residue incorporation (Wang et al., 2007).

The retention and incorporation of crop residues in rice-based cropping systems often has little or no short-term benefits on rice yield (Bijay-Singh et al., 2001; Samra et al., 2003). Long-term experiments have often shown a delayed effect of crop residues—with little benefit on yield increase until after 3 to 4 yr of continuous residue application (Wang et al., 2007). Yadvinder-Singh et al. (2005), in a literature review, indicated that the time of fertilizer N application could be important in enhancing crop yields on residue-treated soils. They concluded that despite the large body of literature on crop residues, very little information is available to enable the proper evaluation of residues for their fertilizer value. There are a few reports of beneficial effects of initial, starter doses of fertilizer in straw-amended soils, but Yadvinder-Singh et al. (2005) concluded that this topic required further evaluation.

We investigated the N-supplying capacity of incorporated crop residues in a rice–wheat cropping system in China and India and a rice–rice cropping system in China. Our study focused on the first 3 yr of residue application because the willingness of farmers to adopt residue retaining technologies depends more on the short- than long-term effects of the residue. The supply of N from crop residues was determined in the absence of fertilizer N as the difference in total plant N between plots with and without residue. Our first objective was to determine during a 3-yr period the N contributions of rice residue to subsequent wheat or rice crops and of wheat residue to a subsequent rice crop. Our second objective was to assess the probable modification in fertilizer N timing and rates required to optimize the supply of N to rice and wheat grown on residue-treated soil. We used temporal differences in plant N between plots with and without residue to assess the effect of residue on the crop's need for fertilizer N.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Site Description
Three field experiments of 3-yr duration were conducted at Yixing and Taojiang in China and at Ludhiana in India. Yixing (31°17' N, 119°54' E) is located in the lower reaches of the Yangtze River, and Taojiang (28°13' N, 112°08' E) is located in the lower reaches of the Zijiang River. Ludhiana (30°56' N, 75°52' E) is in the Indo-Gangetic Plain in northwestern India. The three sites have subtropical climates. Mean monthly temperature and rainfall for the sites are shown in Fig. 1 .


Figure 1
View larger version (32K):
[in this window]
[in a new window]

  		Fig. 1. Rainfall pattern and average temperature based on the monthly averages for a 25-yr period (1981–2005) at Ludhiana, India, and a 3-yr period (2003–2005) at Yixing and Taojiang, China.

 
The experiments were implemented with cropping systems common at the sites. Rice–wheat is the typical rotation at Yixing, with rice from June to October and wheat from November to May. Rice is typically established soon after the harvest of wheat. The common rotation at Taojiang is rice–rice–upland crop with two short-duration rice crops grown within May to October and fallow or upland crops such as vegetables or rapeseed from November to April. At Ludhiana, rice is grown between June and October, and wheat is grown between November and March. Wheat is planted soon after the harvest of rice, whereas land is fallow from April to June between the harvest of wheat and transplanting of rice.

Soils were a moderately acid loam (Typic Endoaquept) at Yixing, an acid silt loam (Gleyic Fluvisol by FAO classification system) at Taojiang, and a mildly alkaline loamy sand (Typic Ustipsamment) at Ludhiana. Initial soil samples collected from plots in each field experiment were mixed, combined by field replication, air dried, sieved, and analyzed in the Analytical Service Laboratory (ASL) of IRRI, Philippines. Total C and N using a CN analyzer, pH (H2O, 1:1), bicarbonate-extractable P, exchangeable K, cation exchange capacity with NH4OAc at pH 7, and particle size by the hydrometer method are shown in Table 1 .


View this table:
[in this window]
[in a new window]

  		Table 1. Properties of soil in the top 20 cm at the two sites in China and in the top 15 cm at the site in India.

 
Experiment Description
In this study, we selected treatments from within experiments with a split-plot design and three replications at Yixing and four replications at Taojiang and Ludhiana. Main-plot treatments in the experiments were management of crop residue on the succeeding crop, and subplot treatments were N management on the crop receiving the residue. The experiment at Yixing examined the effect of wheat residue to the succeeding rice. The selected treatments were a factorial combination of two methods for managing wheat residue (removal except for 5–10-cm height of standing stem after harvest or incorporation of chopped residue) and two rates of fertilizer N applied to rice (0 or 200 kg N ha–1). Fertilizer N was applied three times with 38% at 1 d before rice transplanting, 27% at 10 d after transplanting (DAT), and 35% at 53 DAT. The plus-N subplot was 100 m2 and the minus-N subplot was 25 m2.

The experiment at Taojiang examined the effect of rice residue on the succeeding rice crop. The treatments were a factorial combination of two residue management practices (removal except for 5–10-cm height of standing stem after harvest or incorporation during tillage) and two rates of fertilizer N (0 for each rice crop or 80 kg N ha–1 for early rice and 105 kg N ha–1 for late rice). Tillage was with conventional plowing, which incorporated residue within the top 15-cm soil layer. The plus-N subplot was 64 m2 and the minus-N subplot was 16 m2. All treatments were imposed on each of the two rice crops in each year. Early rice was established by throwing about 20-d-old seedlings onto the plots, and late rice was established by manual transplanting of seedlings.

The experiment at Ludhiana examined the effect of rice residue to the succeeding wheat crop. The treatments were a factorial combination of conventional tillage practices after rice harvest with two management practices for rice residue (removal or incorporation) and two rates of fertilizer N applied to wheat (0 or 120 kg N ha–1). Fertilizer N for wheat was applied 50% at sowing and 50% at tillering at 20 to 21 d after sowing (DAS). The plus-N subplot was 70 m2. The minus-N subplot was 18 m2 rotated each year within a 70-m2 area in which fertilizer N was applied before initiating the minus-N treatment.

Treatments were repeated during 3 yr for rice at Yixing, late rice at Taojiang, and wheat at Ludhiana (Table 2 ). Treatments for early rice at Taojiang and rice at Ludhiana were repeated for 2 yr. In the rice–wheat system at Yixing, wheat was harvested immediately (6 d) before rice transplanting. About 3.7 Mg ha–1 of wheat residue on a dry-weight basis was applied to rice each year at 1 d before transplanting rice. Rice residue was used as mulch for wheat in all treatments, but its influence on wheat was not examined. In the rice–rice system at Taojiang, late rice was transplanted 4 to 10 d after harvest of early rice. About 3.0 Mg ha–1 of residue from early rice was applied. The period from harvest of late rice to establishment of early rice was 186 to 187 d. About 3.5 Mg ha–1 of residue from late rice was left in the field through the long winter season and partly decomposed before establishment of early rice. At Ludhiana, the interval between rice harvest and wheat sowing was 29 to 41 d, which provided only a brief period for decomposition of rice residue before wheat. About 8 to 9 Mg ha–1 of rice residue was returned to wheat and its influence on both wheat and succeeding rice was examined. Wheat residue was removed in all treatments.


View this table:
[in this window]
[in a new window]

  		Table 2. Cropping patterns, residue management, and time intervals between crops in experiments at the three sites.

 
Measurements
Grain yield was determined on harvest areas of 1 m2 at Yixing, 3 m2 at Taojiang (1 m2 in the minus-N plots), and 16 m2 (7.7 m2 in the minus-N plots) at Ludhiana. Plant biomass at different growth stages at Yixing and Taojiang was measured from five hills, which had an average tiller number calculated from 20 hills for each plot. Plant biomass at different growth stages at Ludhiana was measured from a 1-m row length. The N concentration in plant parts was determined by micro-Kjeldahl digestion with H2SO4 and subsequent analysis of NH4 by either steam distillation and titration or colorimetry (Bremner and Mulvaney, 1982).

Total plant N at crop maturity was the sum of accumulated N in all aboveground plant parts. Indigenous N supply (INS), which represents the supply of N from all sources other than fertilizer N, was estimated as the total plant N in mature crops not receiving fertilizer N. The agronomic efficiency for applied fertilizer N (AEN) was calculated as follows (Cassman et al., 1998):

Formula
where YN is the rice grain yield (kg ha–1) at a certain level of applied fertilizer N (FN, kg ha–1), and Y0 is the rice grain yield (kg ha–1) measured with no N application. The optimum fertilizer N rate (FN') was calculated from crop response to N divided by a target AEN:

Formula
The target AEN was set at 15 kg kg–1 for China sites and 20 kg kg–1 for the India site, which represent values intermediate between the AEN often obtained with farmers' management of fertilizer N and the AEN obtainable with improved site-specific nutrient management practices (IRRI, 2007).

Statistical Analyses
Treatment effects were determined using the Mixed Procedure of SAS Version 9.1.2 (SAS Institute, 2003). A combined ANOVA was performed on data for grain yield, total plant N, AEN, and crop response to fertilizer N from all years for (i) rice at Yixing, (ii) early rice at Taojiang, (iii) late rice at Taojiang, (iv) wheat at Ludhiana, and (v) rice at Ludhiana. In these combined analyses, the year variable was a repeated measurement factor. Rice at Yixing was repeated in 2003, 2004, and 2005. Early rice at Taojiang was repeated in 2004 and 2005, and late rice at Taojiang was repeated in 2003, 2004, and 2005. Wheat at Ludhiana was repeated in 2003–2004, 2004–2005, and 2005–2006, and rice at Ludhiana was repeated in 2004 and 2005. Residue and year were handled as fixed effects in the ANOVA to evaluate residue effects on the variables measured across years.

Other combined ANOVAs were run with plant biomass and plant N across different growth stages and years. The analyses included (i) rice crops in 2003, 2004, and 2005 at Yixing; (ii) early and late rice crops in 2005 at Taojiang; and (iii) wheat crops in 2003–2004, 2004–2005, and 2005–2006 at Ludhiana. The repeated factor in the combined ANOVAs was year for Yixing and Ludhiana and growth stage for Taojiang. Residue, growth stage, and year were handled as fixed effects in the ANOVAs to evaluate residue effects on the variables measured across years and crop growth stages. Mean comparison was with Fisher's protected least significant difference (LSD) at the 0.05, 0.01, and 0.001 levels of probability.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Grain Yield with Fertilizer Nitrogen
The incorporation of crop residue never increased the grain yield of rice or wheat at any site when fertilizer N was applied at rates sufficient to meet the crop requirements for supplemental N (Table 3 ). At Ludhiana, the incorporation of rice residue to wheat had no effect on the yield of wheat or the succeeding rice. Our results are consistent with the findings of Yadvinder-Singh et al. (2006), who found wheat and rice yields were not significantly affected by residue from the previous crop in a 4-yr experiment on a comparable sandy loam soil in Punjab, India.


View this table:
[in this window]
[in a new window]

  		Table 3. Effect of crop residue application on grain yield of the succeeding rice or wheat crop in the presence and absence of fertilizer N.

 
Indigenous Nitrogen Supply
Incorporation of rice residue in the absence of fertilizer N had no effect on yield of the subsequent rice crop at Taojiang (Table 3), whereas at Yixing the application of wheat residue in the absence of fertilizer N slightly increased the yield of subsequent rice by 0.4 Mg ha–1 (P = 0.12). At Ludhiana, rice residue applied about 20 d before sowing wheat significantly decreased the yield of wheat by 0.2 Mg ha–1 (P = 0.02). The rice residue incorporated to wheat significantly increased the yield of the following rice in 1 of 2 yr.

The incorporation of cereal crop residues often results in a short-term immobilization of inorganic N and a temporary decrease in plant-available N followed by net N mineralization (Yadvinder-Singh et al., 2005). The net effect of incorporated crop residues on the supply of N to a crop therefore depends on the duration and magnitude of the initial net immobilization of N and the subsequent duration and magnitude of net N mineralization. We used total N in the aboveground biomass of mature crops in plots not receiving fertilizer N as a measure of N supply from all sources other than fertilizer N. The difference in plant N between plots with and without residue was used to represent the net effect of residue on N supply during the entire cropping season (Table 4 ).


View this table:
[in this window]
[in a new window]

  		Table 4. Effect of crop residue application on indigenous N supply for the succeeding rice or wheat crop, as determined from total N in above-ground biomass at maturity in plots not receiving fertilizer N.

 
At Yixing, the incorporation of wheat residue before rice increased (P = 0.03) the accumulation of N in rice by 14 kg N ha–1 averaged across 3 yr (Table 4). This accumulation of N by rice nearly matched the approximately 15 kg N ha–1 added in each season with the wheat residue, and it corresponded to a 0.4 Mg ha–1 increase in rice yield (Table 3). The effect of wheat residue on rice yield and INS was consistent across the 3 yr.

At Taojiang, the applied rice residue contained approximately 20 kg N ha–1 based on N concentrations within the range of 5 to 7 g kg–1 across seasons and years. This incorporation of rice residue had no effect on N supply to either early or late rice (Table 4). As at Yixing, the effect of residue on yield and INS was consistent across all years.

At Ludhiana, the applied rice residue contained approximately 40 kg N ha–1 based on N concentration within the range of approximately 4 to 5 g kg–1 across the years. This incorporation of rice residue before wheat decreased (P < 0.01) the accumulation of N in wheat by 3 kg ha–1 N but increased (P = 0.03) the accumulation of N in rice following wheat by 5 kg N ha–1 (Table 4). The net effect of rice residue on N supply for an entire annual cropping cycle of one wheat and one rice crop was negligible. As at the other sites, the effect of residue on INS was consistent across all years.

The net benefit of residue on N supply reported in some cases in our study could result from a direct effect of residue serving as an N source and an indirect effect in which C from the added residue increased microbial activity. Such increased microbial activity might contribute to increased release of N from soil organic matter (Asten et al., 2005) or increased biological N2 fixation in the soil (Roper and Ladha, 1995). It cannot be determined from our measurements whether these processes occurred in our study.

Patterns in Plant Growth and Nitrogen Accumulation
In the absence of fertilizer N, the rates of crop growth and N accumulation rely on N from indigenous sources, which include soil, irrigation water, crop residues, and manure. We assessed the effect of residue on crop growth and N uptake at specific growth stages from the differences in plant biomass and plant N between plots with and without residue in the absence of fertilizer N.

Yixing
At Yixing, the incorporation of wheat residue had no beneficial effect on early growth and N accumulation of rice (Table 5 ). Rice biomass and N accumulation at panicle initiation and from panicle initiation to flowering were not enhanced by wheat residue, but rice biomass and N accumulation after flowering were markedly increased by incorporation of wheat residue. Wheat residue increased rice biomass by 1.2 Mg ha–1 from flowering to maturity. This corresponded to an increase of 17 kg N ha–1 in aboveground plant biomass. Incorporation of wheat residue resulted in short-term immobilization of N, which was followed by increased release of plant-available N during the ripening phase of rice.


View this table:
[in this window]
[in a new window]

  		Table 5. Effect of wheat residue on biomass production and N accumulation by a succeeding rice crop at specific growth stages in plots with and without fertilizer N at Yixing, China.

 
Rice responded to fertilizer N (Table 3), which confirms that rice growth was limited by insufficient supply of N in the absence of fertilizer N. The application of fertilizer N (200 kg N ha–1) far exceeded the requirement for the mean rice response to N of 0.8 to 1.3 Mg ha–1 (Table 3). Assuming a relatively low AEN of 10 kg grain increase kg–1 N applied, which can occur in farmers' fields in China (Peng et al., 2006), then only about 100 kg N ha–1—half the applied N—would be required to achieve the observed response of approximately 1 Mg ha–1 (IRRI, 2007).

The application of surplus fertilizer N masked the benefit of wheat residue on rice during the ripening phase (Table 5). Fertilizer N applied by 10 DAT (130 kg N ha–1) eliminated any detrimental effect of wheat residue on rice growth during the vegetative stage. The remaining fertilizer N applied at 53 DAT—immediately before panicle initiation—ensured a surplus supply of N in the absence of wheat residue, which negated any beneficial N effect of wheat residue on rice during the ripening phase (Table 5).

Taojiang
In the rice–rice system at Taojiang, the incorporation of rice residue had no beneficial effect on growth of either early or late rice in the absence of fertilizer N (Table 6 ). Residue slightly increased the growth of only early rice from flowering to maturity, but the effect (0.3 Mg ha–1, P = 0.35) was not comparable to the pronounced positive effect of residue on rice growth during the ripening phase at Yixing (Table 5). The effect of residue on late growth of rice at Yixing (Table 5) but not at Taojiang (Table 6) can be possibly attributed to the longer growth duration of rice at Yixing, which provided more time for N released after an initial period of immobilization to be utilized by the rice.


View this table:
[in this window]
[in a new window]

  		Table 6. Effect of rice residue on biomass accumulation by the succeeding rice crop at specific growth stages with and without fertilizer N at Taojiang, China, in 2005.

 
The response of early and late rice to fertilizer N at Taojiang (Table 3) confirms that rice growth was limited by an insufficient supply of N in the absence of fertilizer N (Table 6). The response to fertilizer N (about 1.2 Mg ha–1) was comparable to the response at Yixing (Table 3), but the rate of fertilizer N was markedly lower at Taojiang (early rice = 80 kg N ha–1, late rice = 105 kg N ha–1), and fertilizer N was applied earlier at Taojiang than at Yixing. At Taojiang, all the fertilizer N for early rice was applied by 16 DAT and 90% of the fertilizer N for late rice was applied by 18 DAT. This relatively early application of fertilizer N at Taojiang could have resulted in a suboptimal supply of N to rice at late growth stages.

With the application of fertilizer N, residue increased biomass accumulation of early rice but not late rice from flowering to maturity by 0.7 Mg ha–1 (Table 6). This relatively higher biomass accumulation at later growth stages in plots receiving residue supports the supposition that N was relatively more limiting at later growth stages when residue was not applied. Residue had no beneficial effect on late growth of N-fertilized late rice (Table 6), perhaps because a small amount of fertilizer N (12 kg N ha–1) was applied at 44 DAT, which is approximately panicle initiation.

Ludhiana
At Ludhiana, the incorporation of rice residue to wheat reduced biomass and N accumulation for the mature wheat crop in 2 of the 3 yr in the absence of fertilizer N (Table 7 ). Biomass and N accumulation of wheat were enhanced by incorporated rice residue only at later growth stages in the third year (2005–2006). This benefit was negated by a negative effect of residue on early wheat growth (Table 7), and rice residue consequently had no net benefit for yield (Table 3).


View this table:
[in this window]
[in a new window]

  		Table 7. Effect of rice residue on biomass production and N accumulation by a succeeding wheat crop at specific growth stages in plots with and without fertilizer N at Ludhiana, India.

 
Despite the application of all fertilizer N (120 kg N ha–1) to wheat by 21 DAS, early growth and N accumulation of wheat were reduced by incorporation of rice residue (Table 7), presumably because of a reduced supply of N arising from N immobilization. This early suppression of wheat growth was partially negated by slightly enhanced growth at later growth stages. Incorporated rice residue consistently had no net benefit for wheat, and yield of N-fertilized wheat averaged for the 3 yr was reduced (0.2 Mg ha–1, P = 0.02) by incorporated rice residue (Table 3). These findings are consistent with other research in northwestern India, which has typically reported little or even negative effects of incorporated rice residue on wheat in the first 3 yr (Yadvinder-Singh et al., 2005).

Crop Response to Nitrogen and Agronomic Efficiency
The difference in yield between plots with and without fertilizer N reflects the response of the crop to fertilizer N. This response to fertilizer N is reduced when crop yield without fertilizer N increases as a result of increased supply of N from residue and the crop yield with fertilizer N is not increased by the residue. Such an instance of reduced crop response to fertilizer N can reflect a probable opportunity for saving fertilizer N through the use of residue.

Crop response to fertilizer N was not commonly influenced by residue (Table 8 ) because crop yield without fertilizer N was only occasionally increased by residue and the yield of N-fertilized crops was unaffected by residue (Table 3). Relatively higher mean rice yield with incorporated wheat residue in the absence of fertilizer N at Yixing (Table 3) resulted in a significantly lower response of rice to N with incorporated residue in the third of the 3 yr (Table 8).


View this table:
[in this window]
[in a new window]

  		Table 8. Crop response to fertilizer N and agronomic efficiency for applied fertilizer N (AEN) as affected by crop residue.

 
Higher yield at Ludhiana for rice following wheat amended with rice residue without fertilizer N (Table 3) corresponded to a reduced response of rice to fertilizer N (P = 0.06) (Table 8). A reduced crop response to fertilizer N corresponded in our study to reduced AEN (Table 8) because the same rate of fertilizer N was used in plots with and without residue. The use of less fertilizer N with no reduction in yield of the N-fertilized crop would increase AEN.

Modern rice cultivars typically respond to recommended fertilizer N practices and achieve an AEN ranging from 15 to 25 kg grain increase kg–1 N applied (Yoshida, 1981; IRRI, 2007). The very low AEN in our researcher-managed experiment at Yixing (3–8 kg grain increase kg–1 N applied) reflects the low crop response to N (0.6–1.6 Mg ha–1) with high use of fertilizer N (200 kg N ha–1) (Table 8). Such low AEN are common with existing farmers' fertilizer N practices in Jiangsu Province and coastal areas of China (Peng et al., 2006). Considerable scope exists through improved fertilizer N management to achieve high rice yields with more efficient use of fertilizer N regardless of the residue management practice (Hu et al., 2007). Specific guidelines on improved N management through the use of the leaf color chart (LCC) to optimally supply fertilizer N to rice are available for farmers (IRRI, 2007).

Optimal Fertilizer Nitrogen Management
We estimated optimal fertilizer N requirements for the observed crop responses to fertilizer N based on the assumption that a target AEN of 15 grain increase kg–1 N applied was achievable in China and an AEN of 20 kg grain increase kg–1 N applied was achievable in India. These levels of AEN are within the range currently targeted for fertilizer N with a widely evaluated and promoted site-specific nutrient management approach for rice (IRRI, 2007).

Because of the negligible effect at Taojiang of rice residue on yield and response of rice to fertilizer N (Table 8), the fertilizer N required for the target AEN was unaffected by residue (Fig. 2 ). Estimated fertilizer N rates for early and late rice were near the 1:1 line, representing equal N rates with and without residue incorporation. At Yixing, because wheat residue reduced the response of rice to fertilizer N in 2 of 3 yr (Table 8), the fertilizer N required for the target AEN was often reduced by residue (Fig. 2). The one observation above the 1:1 line, indicating a greater requirement of fertilizer N with residue incorporation, occurred in the first year. By the third year of the experiment, all estimated fertilizer N requirements for rice were less in the presence of residue. Research in India on a rice–wheat cropping system has similarly shown a reduced fertilizer N requirement for rice following incorporation of wheat residue (Yadvinder-Singh, 2003).


Figure 2
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 2. Effect of rice and wheat residue on fertilizer N required by rice to achieve the attained yield with a target agronomic efficiency for applied fertilizer N (AEN) of 15 kg grain increase kg–1 N applied at Yixing and Taojiang, China.

 
The incorporation of rice residue before sowing wheat at Ludhiana did not increase the supply of N to wheat (Table 4) or the yield of wheat (Table 3). The residue consequently did not result in a savings of fertilizer N for wheat (Fig. 3 ). The incorporated rice residue increased the supply of N in the succeeding rice crop after wheat (Table 4). The fertilizer N required for rice to achieve the target AEN was consequently often reduced when rice residue was incorporated before wheat (Fig. 3). The one observation with a greater requirement of fertilizer N with residue incorporation occurred in the first year. Several years of residue incorporation in a rice–wheat system might be required before the N benefit of rice residue on the following rice is evident. Verma and Bhagat (1992) reported a benefit only from the third rice crop onward.


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		Fig. 3. Effect of rice residue on fertilizer N required by rice and wheat to achieve the attained yield with a target agronomic efficiency for applied fertilizer N (AEN) of 20 kg grain increase kg–1 N applied at Ludhiana, India.

 
General Discussion
The incorporation of cereal crop residues has often been promoted in rice-based systems for maintenance of soil fertility and savings in fertilizer use (Bird et al., 2001; Pathak et al., 2006; Linquist et al., 2006). Our study, representing three major rice-based cropping areas in Asia, found little or no net benefit of incorporated cereal residues on the supply of N to a succeeding rice or wheat crop within 3 yr. Our results provided insights into the characteristics of rice-based cropping systems with relatively greater probability of a net N contribution from incorporated cereal residues—albeit probably small—and how to manage fertilizer N to optimize potential benefits from residues.

We consistently found no net benefit from incorporated rice residue to succeeding rice across five cropping seasons in a continuous rice-cropping system at Taojiang (Table 3). Rice cultivars in the cropping system were relatively short duration for early rice (81–86 d from seedling throwing to harvest) and late rice (87–90 d from transplanting to harvest). Rice residue was incorporated immediately before transplanting (3–9 d), which did not provide sufficient time for the residue to decompose and reach a C/N ratio suitable for net N release. The duration of the rice cultivars was presumably not sufficiently long for rice to measurably benefit from enhanced release of N following an initial period of N immobilization.

The scope for saving fertilizer N through residue incorporation in such continuous rice-cropping systems is limited because of the short time for residue decomposition before rice transplanting and the relatively short duration of the rice cultivars. Climate and cropping intensity typically prevent using a longer time for in situ residue decomposition before transplanting or using longer duration rice cultivars. One possible management alternative to achieve greater benefit from residue is controlled irrigation to alternately wet and dry soil to hasten residue decomposition (Kongchum et al., 2006) and reduce possible formation of organic acids (Johnson et al., 2006), while not reducing soil water to levels that adversely affect rice yield. Another alternative is shallow incorporation or surface placement of residue to minimize N immobilization and detrimental effects of partially decomposed crop residue in the root zone of young rice.

The incorporation of wheat residue to rice at Yixing consistently increased N supply to rice in a wheat–rice rotation common in China. As in the continuous rice system at Taojiang, the residue was incorporated immediately before rice establishment, thereby leading to short-term immobilization of N during growth of young rice. The rice cultivar, however, was longer duration (132 d from transplanting to harvest) than at Taojiang. As a result, rice had more time to benefit at later growth stages from the N released from the residue. In N-fertilized rice, the N supply from the residue did not contribute to added yield or savings in fertilizer N because the fertilizer N rate was excessive and not adjusted to account for N supplied by the residue.

The net N contribution of wheat residue in Yixing (14 kg N ha–1) was relatively small compared with the total N taken up by rice (130 kg N ha–1) (Table 3). Achieving a savings in fertilizer N use through exploitation of the net N contribution from residue requires fine-tuning of fertilizer N to optimally match the temporal demand of rice for supplemental N (Cassman et al., 2002). In residue-amended soil, a relatively larger portion of the fertilizer N would probably be needed for young rice to counter N immobilization, which is followed by increased N release from residue at later growth stages when relatively less fertilizer N would probably be needed. The LCC can be used for fine-tuning the supply of fertilizer N to optimally supplement N supplied by indigenous sources including crop residues (Witt et al., 2005; IRRI, 2007).

Our findings on the incorporation of rice residue in a rice–wheat rotation at Ludhiana are consistent with other reports (Yadvinder-Singh et al., 2005) that incorporated rice residue seldom has a beneficial effect on wheat and can even have a slight detrimental effect. In our study, the incorporated rice residue occasionally contributed significant N to the succeeding rice crop in a rice–wheat rotation. This suggests an opportunity to save fertilizer N for rice when residues of the preceding rice crop are incorporated before wheat. The N contribution for rice residue to the succeeding rice crop was small in our study, suggesting that a savings in fertilizer without a loss in yield would be small and require applying fertilizer N at times and doses to optimally match the temporal need of the rice crop for supplemental N. Recent developments of machinery for simultaneously mulching rice straw while sowing wheat (Kukal et al., 2005; Sidhu et al., 2007) provide the option of surface-applied rice residue rather than incorporation and burning. Mulched rice residue is more likely than incorporated residue to have a beneficial effect on wheat. It is less likely to result in N immobilization, and mulched residue can also provide non-N benefits such as conservation of soil water and control of weeds.

In our study, the incorporation of rice and wheat residues solely as a source of supplemental N in rice-based cropping systems did not provide an attractive alternative to the farmers' practice of burning crop residues. The supply of N from incorporated cereal residues was small or negligible, suggesting that Asian rice farmers can often expect no savings in fertilizer N in the first 3 yr of incorporating residue in rice–rice and rice–wheat cropping systems. Incorporated residue through effects on soil processes including N immobilization, N mineralization, and biological N2 fixation can alter the amount and time of N release to crops. In instances where incorporated residue increases the supply of plant-available N, the application of fertilizer N to crops must be skillfully adjusted in time and amount to optimally supplement N from the residue and soil to save fertilizer N with no loss in crop yield.


   ACKNOWLEDGMENTS
 
The research was funded by the Federal Ministry for Economic Cooperation and Development (BMZ), Germany, through the project "Managing Crop Residues for Healthy Soils in Rice Ecosystems" with the International Rice Research Institute (IRRI).


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

Received for publication November 27, 2006.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Agricola
Right arrow Articles by Thuy, N. H.
Right arrow Articles by Buresh, R. J.
Related Collections
Right arrow Rice
Right arrow Residue management

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome