| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Bangladesh Rice Res. Inst. (BRRI), Gazipur 1701, Bangladesh
b Dep. Agric. Extension (DAE), Bangladesh
c PETRRA funded Project, BRRI, Gazipur 1701, Bangladesh
d IRRI, CG Block, NASC Complex, DPS Marg, Pusa, New Delhi 110012, India
e Int. Rice Res. Inst., DAPO Box 7777, Metro Manila, Philippines
* Corresponding author (r.buresh{at}cgiar.org)
Received for publication July 29, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
0.3 Mg ha–1 at about half the combinations of villages and seasons when the LCC was combined with recommended P, K, S, and Zn fertilization. Grain yields were further increased by another 0.4 Mg ha–1 at about half the combinations of villages and seasons when the LCC and recommended P, K, S, and Zn fertilization were combined with improved crop management, involving the recommended manual weed control and plant spacing. Use of the LCC with rice is ready for wide-scale promotion in Bangladesh.
Abbreviations: AEZ, Agro-Ecological Zone • ANOVA, analysis of variance • BRRI, Bangladesh Rice Research Institute • DAT, days after transplanting • F, farmer management • FP, farmers' practice • IFCM, improved fertilizer and crop management practice • IFM, improved fertilizer management practice • LCC, leaf color chart • LCCN, leaf color chart based N management • LSD, least significant difference • R, recommended management • RFM, recommended fertilizer management practice • RN, recommended N fertilizer management practice • SD, standard deviation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Despite considerable research to increase N fertilizer use efficiency for rice, the recovery efficiency (REN) from applied N fertilizer achieved by rice farmers is typically only about 30 to 40% (Cassman et al., 1993). This low REN is associated with large loss of fertilizer N from the soil–plant system, which can potentially lead to environmental pollution (Bijay-Singh and Yadvinder-Singh, 2003). The inefficient use of N fertilizer, particularly at high levels of yield and cropping intensity, can also result in crop lodging, increased pest and disease infestation, and reduced profitability for farmers. The development and promotion of more efficient practices for N fertilizer management in rice consequently remain a high priority for increasing profitability of rice farming while protecting the environment (Buresh et al., 2004; Dawe et al., 2004).
One approach to increasing N fertilizer use is to better match the timing and doses of N fertilizer application with the need of the crop for N. In this so-called "real-time" approach to N management the timing of N fertilizer applications is determined through periodic monitoring of crop N status (Witt et al., 2004). Two decision aids available for in situ monitoring of leaf N status in rice are the chlorophyll meter and the leaf color chart (LCC) (Balasubramanian et al., 1999; Peng et al., 1996). A chlorophyll meter can provide a quick estimate of the leaf N status, but it is relatively expensive. The LCC on the other hand is inexpensive, simple, and an easy-to-use alternative to monitor the relative greenness of the rice leaf as an indicator of crop N status (Shukla et al., 2004; Yang et al., 2003). The LCCs used in Asia are typically a durable plastic strip about 7 cm wide and 13 to 20 cm long, containing four to six panels that range in color from yellowish green to dark green. With a real-time approach to N management, farmers monitor the color of rice leaves at 7- to 10-d intervals and apply N fertilizer wherever leaves become more yellowish green than the critical color on the LCC. Real-time N management with the LCC has been under evaluation in farmers' fields in Asia since the late 1990s.
Bangladesh is a major rice-producing country with a growing population and diminishing arable land. Bangladesh must produce about 28 million tonnes of milled rice by the year 2020 to feed the growing population. This target level of production is about 25% higher than the production level for 2000, and because of limitations in arable land it must be achieved through increased yield and higher cropping intensity (Zohir et al., 2002). This increased production results in greater need of nutrients by the crop. Increased use efficiency of N fertilizer as defined by increased production of grain per unit of applied N could help ensure that yield targets are met through profitable and sustainable rice farming. In this study, we evaluate the performance of the LCC for N fertilizer management of rice and develop a recommendation for the use of the LCC in the two main rice-growing seasons in Bangladesh.
The specific objective of this study was to evaluate real-time N fertilizer management with the LCC relative to the farmers' fertilizer practice and the current recommendation made by the Bangladesh Rice Research Institute (BRRI) for N fertilizer management in intensive continuous rice-cropping systems (BRRI, 1999). Farmers in Bangladesh often do not apply the recommended rates of nutrients other than N. We therefore also evaluated the additional benefits arising from the integrated use of the LCC for N management with the current recommendation for P, K, S, and Zn fertilization (BRRI, 1999), which is promoted by the extension service. The performance of fertilizer management practices were based on grain yields, profitability in terms of added net returns as compared to the farmers' practice, and measures of N fertilizer use efficiency.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In all villages, two rice crops are grown annually in two seasons known as Aman and Boro. Aman is the wet season (June–July to November–December) in which transplanted rice is grown under partially irrigated conditions in the study area. Boro is the dry season (December–January to April–May) in which transplanted rice is grown under full irrigated conditions. Semi-dwarf, high-yielding rice varieties were grown in all villages; BR11 with 145 d duration was grown in Aman, and BRRI dhan28 with 140 d duration was grown in Boro. The number of villages varied from year to year. The experiment was conducted in one village (Chandbill) during 2000–2001, two villages (Chandbill and Baria) during 2001–2002, and five villages (Chandbill, Hizuli, Badarganj, Nolkhola, and Visawkhali) during 2002–2003. Each village averaged about 500 ha of cropland with rice grown on about 80% of the cropland.
Experimental Design and Treatments
The experiment was established in farmers' fields in a randomized complete block design with one complete set of treatments in each farmer's field, which averaged 1000 to 1500 m2. Each farmer's field had similar past management and uniform soil type, and each field was considered a replication. The selected farmers' fields (replications) within a village were located within a radius of 1 to 2 km. The number of replications varied among villages and seasons and totaled 10 at one village in Aman 2000, 8 at one village in Boro 2001, 20 at two villages in Aman 2001, 13 at two villages in Boro 2002, 36 at five villages in Aman 2002, and 23 at five villages in Boro 2003.
Treatments consisted of options for managing fertilizers, weeds, and plant spacing. In all, six treatments were used. The treatments were as follows:
The number of treatments varied from season to season. The number of evaluated treatments were three (FP, RFM, and IFM) in Aman 2000, four (FP, RFM, LCCN, and IFM) in Boro 2001, five (FP, RN, LCCN, IFM, and IFCM) in Aman 2001 and Boro 2002, and four (FP, LCCN, IFM, and IFCM) in Aman 2002 and Boro 2003. A control plot without N fertilization (N0) but with P, K, S, and Zn fertilization as recommended by BRRI was included in each field in each season and year. The yields and N accumulation of grain and straw in the N0 plot were used only to calculate N fertilizer use efficiency.
In recommended fertilizer management practices, the rates of N, P, K, S, and Zn fertilizer were 80, 20, 35, 11, and 5 kg ha–1, respectively, on an elemental basis in Aman and 100, 24, 42, 11, and 5 kg ha–1, respectively, on an elemental basis in Boro. The N was applied as urea in three equal splits at 15, 30, and 50 d after transplanting (DAT). The total amounts of P as triple superphosphate, K as KCl, S as gypsum, and Zn as zinc sulfate were applied immediately before transplanting.
The LCC-based N fertilizer management in Aman 2000 and Boro 2001was based on a recommendation developed by Crop and Resource Management Network (CREMNET, 1999) The dose of N fertilizer for each application in the Aman was 20 kg N ha–1 at 15 to 28 DAT, 30 kg N ha–1 applied once at 29 to 48 DAT, and 20 kg N ha–1 between 49 DAT and flowering. The dose of N fertilizer for each application in the Boro was 30 kg N ha–1 at 15 to 28 DAT, 45 kg N ha–1 applied once at 29 to 48 DAT, and 30 kg N ha–1 between 49 DAT and flowering. In subsequent seasons, LCC-based N fertilizer management was adjusted based on results from the first two seasons. The dose of N fertilizer was consistently 25 kg N ha–1 in the Aman and 30 kg N ha–1 in Boro for each application. Leaf color measurements were made with the LCC at 10-d intervals from about 15 DAT to flowering the first four seasons by matching the color of the LCC color panels with the youngest fully extended leaf selected from 10 healthy plants in a plot. In subsequent seasons the crop growth duration for measuring leaf color with LCC was adjusted to 15 DAT to booting stage (75 DAT) in Aman and 21 DAT to booting stage (65 DAT) in Boro. The N fertilizer was applied whenever 6 or more out of 10 LCC measurements fell below the critical LCC value. The CREMNET recommendation of a critical LCC value of 4 was used for each variety in the first three seasons. Thereafter, the critical LCC value as determined from a calibration study for the varieties was adjusted to 3.5.
Crop management from land preparation to harvesting was identical in all treatments and based on the farmer's practice, except when otherwise indicated for the IFCM treatment. Three to four 30-d-old rice seedlings were transplanted between 20 July and 7 August in Aman and three to four 35-d-old rice seedlings were transplanted between 1 and 20 January in Boro across the villages and years. Rice was transplanted randomly following the farmers' practice. Weeds were controlled by hand weeding, usually one to two times during the season. The farmers' practice of pesticide application was used for all treatments placed in the farmers' field (replication). Rice was harvested between 15 and 29 November in Aman and between 20 and 30 April in Boro. In the IFCM treatment, rice was transplanted in lines at the recommended spacing of 20 by 15 cm, and weeds were controlled by hand weeding three times as recommended.
Plant and Soil Sampling and Analysis
Initial soil samples were collected in each farmer's field (replication) before fertilization in the Aman in each year. The samples were collected from the 0- to 15-cm depth from 6 to 12 locations in each field using a 5-cm diameter auger. The soil from a field was mixed thoroughly, and a composite sample was air-dried and sieved to pass a 2-mm sieve. Soil was analyzed for pH, Kjeldahl N, cation exchange capacity, and particle size in the Analytical Service Laboratory (ASL) of IRRI, Philippines. Properties of soil for the experimental fields are shown in Table 1. The soils were silty clay loam at Chandbill and Baria, silty clay at Nolkhola, and silt loam at Hizuli, Badarganj, and Visawkhali.
|
Data Analysis
The agronomic efficiency (AEN), apparent recovery efficiency (REN), and partial factor productivity (PFPN) from applied N fertilizer were calculated as follows (Cassman et al., 1998):
 |
 |
 |
Analysis of variance (ANOVA) was performed on yield and N efficiency parameters to determine the effects of villages, treatments, and their interactions using the PROC MIXED procedure of the SAS/STAT software (SAS Inst., 1999). Least significant difference (LSD) at the 0.05 level of probability was used to evaluate the differences among treatment means. In addition, P values were used to indicate the significance of differences among treatment means. Descriptive statistics such as means and standard deviation were used to determine the variability of parameters.
Treatments were evaluated based on added net return relative to the farmers' practice (FP), which is the difference between added gross return and added cost for a treatment as compared with FP. Added gross return equaled [(Yield of treatment – Yield of FP) x Price of yield]. Added cost equaled the sum of costs for differences in labor [(Labor for transplanting, weeding, measuring leaf color with the LCC and fertilizer application of the treatment – Labor for transplanting, weeding and fertilizer application of FP) x Wage rate] and for additional fertilizer (Fertilizer costs of treatment – Fertilizer costs of FP). The prices of different fertilizers, rough rice, and labor wage in different villages and years ranged from (in US$) $0.20 to $0.22 kg–1 N, $1.02 to $1.19 kg–1 P, $0.24 to $0.31 kg–1 K, $0.27 kg–1 S, $2.6 to $2.8 kg–1 Zn, $0.10 to $0.11 kg–1 rough rice, and $1.11 to $1.36 person d–1 (US$1 = Bangladesh Taka 58.8).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
In Boro, 50 to 100% of farmers applied basal N, all farmers (100%) applied N at early tillering and late tillering to panicle initiation (Table 3). The average N application for farmers applying N was 27 to 41 kg N ha–1 for basal, 60 to 86 kg N ha–1 for early tillering, and 54 to 79 kg N ha–1 for late tillering to panicle initiation. Farmers, across villages in both seasons, generally applied all N by 47 DAT and much of the total was applied by 30 DAT. This could result in excess N at early growth stages and N deficit at later growth stages for the varieties, which grew for 110 to 120 d from transplanting to harvest.
All farmers applied P fertilizer in both Aman (Table 2) and Boro (Table 3). The average P application across villages and years ranged from 17 to 30 kg P ha–1 in Aman (Table 2) and 21 to 39 kg P ha–1 in Boro (Table 3). These rates of P application are near to or slightly above the recommendation of 20 kg P ha–1 for Aman and 24 kg P ha–1 for Boro.
All farmers applied K fertilizer in Aman, except at Baria (Table 2); and 75 to 100% of farmers applied K fertilizer in Boro (Table 3). The average K application across villages and years for farmers applying K ranged from 23 to 48 kg K ha–1 in Aman (Table 2) and 24 to 59 kg K ha–1 in Boro (Table 3). These average K applications were close to currently recommended rates of 35 kg K ha–1 for Aman and 42 kg K ha–1 for Boro, but 20 to 25% of farmers at some villages did not apply K in Boro. The farmers' use of K might be associated with a recent campaign of the extension service to promote the benefits of K in reducing insect and disease infestation. Use of K was, however, more variable than P use across villages and years. Farmers at Chandbill in Boro 2001 and at Badarganj in Boro 2003, for example, applied more K (57–59 kg ha–1) than the currently recommended rate (42 kg K ha–1), while farmers of Visawkhali applied only 24 kg K ha–1 in Boro 2003.
Many farmers did not apply S and Zn, and in some villages no farmers applied S and Zn (Tables 2 and 3). The average S application across villages and years for farmers applying S ranged from 4 to 16 kg S ha–1 in Aman (Table 2) and 8 to 28 kg S ha–1 in Boro (Table 3). The average Zn application across villages and years for farmers applying Zn ranged from 3 to 10 kg Zn ha–1 in Aman (Table 2) and 3 to 7 kg Zn ha–1 in Boro (Table 3). Many farmers did not apply S and Zn, but those who applied S and Zn used the amounts near to currently recommended rates of 11 kg S ha–1 and 5 kg Zn ha–1.
Grain Yield
Nutrient and crop management options significantly affected grain yields in the three Aman and three Boro seasons (Table 4). In Boro 2001, the use of the LCC for N management (LCCN) without any other change in farmers' fertilizer or crop management practices significantly increased yield as compared with the FP (4.8 vs. 4.2 Mg ha–1). In both Aman 2000 and Boro 2001, the use of recommended rates of N, P, K, S, and Zn (RFM) significantly increased yield as compared with the FP (4.8 vs. 3.7 Mg ha–1 in Aman and 5.0 vs. 4.2 Mg ha–1 in Boro). Use of the LCC (IFM) rather than the recommended N application for N management (RFM) further increased yield by 0.4 Mg ha–1 in Aman 2000 and 0.6 Mg ha–1 in Boro 2001 (Table 4). Results from the first two seasons indicated considerable opportunity to increase farmers' yield levels through improved N management with the LCC.
|
The integration of LCC-managed N with recommended P, K, S, and Zn fertilization either alone (IFM) or in combination with recommended weed control and crop spacing (IFCM) slightly increased yields in Aman 2001 and Boro 2002. The highest yields in Aman 2001 (4.7 Mg ha–1) and Boro 2002 (6.5 Mg ha–1) were obtained in IFCM. These yields were significantly higher than with LCCN, indicating that modification of the farmers practice to combine the use of the LCC for N management with recommended use of other nutrients and improved crop management further increased yields.
In Aman 2002 and Boro 2003, additional villages were included in the evaluation of nutrient and crop management options. The interaction effect of villages and treatment was significant in Aman 2002 (Table 4). Modification of the farmers' practice to manage N with the LCC, without changing any other practice (LCCN), significantly increased yield as compared with the FP in all cases in the Aman 2002 and Boro 2003, except for the one village of Nolkhola in Aman 2002 (Table 4). The integration of LCC-managed N with recommended P, K, S, and Zn fertilization (IFM) significantly increased yield as compared with only use of the LCC (LCCN) at three of the five villages in Aman 2002 including Nolkhola. Farmers did not apply S and Zn at Nolkhola in Aman 2002 (Table 2), and the significantly higher yield with IFM than LCCN as compared with FP suggests that the limitation of a nutrient other than N, such as S or Zn, must be overcome at this village before farmers can derive the benefits of N management with the LCC. In all other villages and seasons, farmers could obtain a significant yield increase by using the LCC for N management without any change in management of other nutrients. The use of recommended manual weed control and plant spacing (IFCM) consistently increased yield relative to IFM in Aman 2002 and Boro 2003, except for Nolkhola in Aman 2002 (Table 4). Irrespective of season and village, IFCM produced the highest yields.
When examined across all villages and seasons, the modification of the farmers' practice with only use of the LCC to manage N increased average grain yield by 0.1 to 0.7 Mg ha–1. This increase was significant (P < 0.1) in all cases except Nolkhola, where yield was apparently limited more by a nutrient other than N than by inefficient farmers' management of N fertilizer (Table 5). In about half the combinations of villages and seasons, grain yield was further increased by 0.3 Mg ha–1 when use of the LCC was combined with recommended P, K, S, and Zn fertilization. This increment in yield increase presumably resulted from increased use of K, S, or Zn rather than P because farmers' rates of P application were comparable to or even greater than the recommended P rate. The combined use of the LCC for N management with recommended P, K, S, and Zn fertilization and recommended manual weed control and plant spacing significantly increased average yield by 0.6 to 2.0 Mg ha–1 as compared with FP (Table 5). The frequent higher yields with IFCM than LCCN (Table 4) indicate a benefit from integrating the recommended use of other nutrients and improved crop management with the use of the LCC for N management.
|
|
In Aman 2001, the use of the LCC (LCCN, IFM, and IFCM) significantly increased grain yield as compared with FP (Table 4), but this increased grain yield did not lead to significant increases in AEN, REN, or PFPN (Table 6). This failure to increase N fertilizer use efficiency in Aman 2001 was due to the large amount of N applied with the LCC (150 kg ha–1). The relative low AEN (13–15 kg grain increase kg–1 N applied) and REN (0.35–0.41 kg N taken up kg–1 N applied) with the LCC in Aman 2001 indicate scope for further increasing N fertilizer use efficiency by reducing the rate of N application with the LCC without decreasing grain yield. The dose of N fertilizer applied whenever leaf color fell below a critical LCC value was already relatively low (25 kg N ha–1) in the Aman. Therefore, the most appropriate approach to reducing N application with the LCC was to use a lower (less green) critical LCC value. Other research simultaneously conducted in 2000–2001 on the calibration of the LCC for common rice varieties in Bangladesh (Alam et al., unpublished data, 2002) indicated 3.5 was more appropriate than 4 as the critical LCC value. Starting from Boro 2002, the critical LCC value was consequently reduced from 4 to 3.5 and starting from Aman 2002 the last time for measuring leaf color with the LCC was changed from the flowering to booting stage.
In all cases when the critical LCC value was adjusted to 3.5 (Boro 2002, Aman 2002, and Boro 2003), the AEN, REN, and PFPN were significantly higher with use of the LCC (LCCN, IFM, and IFCM) than with FP (Table 6). The AEN, REN, and PFPN only slightly increased when use of the LCC (LCCN) was further combined with recommended P, K, S, and Zn fertilization (IFM) and then with improved weed control and plant spacing (IFCM). The slight increase in N fertilizer use efficiency with IFM and IFCM as compared with LCCN resulted from higher grain yield because the rate of N fertilizer use was similar for all the LCC treatments (LCCN, IFM, and IFCM). This finding is consistent with the general recognition that optimal crop and fertilizer management are favorable for achieving high N fertilizer use efficiency (Peng and Cassman, 1998). Other studies have comparably shown that use of the LCC for N management combined with improved P and K fertilizer management often significantly increases yield and N fertilizer use efficiency as compared with the farmers' fertilizer practice in Asian rice fields (Dobermann et al., 2002, 2004; Gines et al., 2004).
A distinct finding from our research is the relatively consistent significant increases in grain yield (Table 4) and N fertilizer use efficiency (Table 6) achieved in farmers' fields when the only change in the farmers' management of rice was use of the LCC for managing N fertilizer. The gains in yield and N use efficiency were achieved through improved timing for topdressing urea fertilizer, whereby the applications of N more effectively matched the real-time need of the rice crop for additional N. Adjustments in the LCC practice to better match the growing conditions and varieties in Bangladesh, particularly the adjustment of the critical LCC value to 3.5 in the last three seasons, further increased the N fertilizer use efficiency—largely because of a reduction in N fertilizer rates while maintaining high rice yield.
Other evaluations of the LCC in intensive, irrigated rice production systems of Asia have often reported savings in N fertilizer use, which can result in increased N fertilizer use efficiency (Bijay-Singh et al., 2002). This increase in N fertilizer use efficiency has, however, seldom led to such widespread increases in grain yield for transplanted rice as reported in our study, when the current farmers' practices for managing other nutrients and the crop are retained. This huge potential for both increased yield and increased N fertilizer use efficiency in Bangladesh arises because farmers applied excessive N fertilizer, which was typically not timed to match the crop need for added N. Farmers' applications of N before transplanting and during early tillering generally exceeded crop need for N (Tables 2 and 3). In addition, farmers already used sufficient P fertilizer and many farmers were using K as a result of recent promotion through the extension service. Hence, N was typically the nutrient most limiting yield, although there appears to be increasing risk of S or Zn limitations at selected locations.
The rates of N fertilization with the recommended N management practice were insufficient to achieve high yield. The recommended practice for N management increased N fertilizer use efficiency because of the relatively low N rates (Table 6), but it failed to consistently increase yield relative to the farmers practice; and yields were consistently lower with the recommended N management practice than with the LCC (Table 4). Highest yields were typically obtained at N rates higher than recommended but lower than the farmers' current practice.
Financial Analysis
Added costs and added net returns for LCC treatments relative to the FP were used to assess the profitability for use of the LCC either alone (LCCN); in combination with recommended P, K, S, and Zn (IFM); or in combination with both the recommended P, K, S, and Zn and improved manual weed control and plant spacing (IFCM) (Table 7). Modification of the farmers' practice to only include use of the LCC for N management did not increase added cost, except in Aman 2001; but as explained earlier the LCC practice in Aman 2001 was not yet optimal. Results for Boro 2002, Aman 2002, and Boro 2003 correspond to an improved LCC practice that has since been recommended for widespread use in Bangladesh. A negative added cost with use of only the LCC (LCCN) indicates that the saving in cost of urea fertilizer was greater than added labor cost associated with leaf color measurements with the LCC and application of N. The average added net returns with use of only the LCC was always positive, ranging (in US$) from $41 to $65 ha–1 across seasons and years. In each of the last three seasons, when the LCC practice had been improved, the added net return was $18 ha–1 in at least 75% of the farmers' fields (Table 7). Virtually all the positive net return arises from increased grain yield rather than savings in N fertilizer, even though the rates of N fertilizer were reduced with the LCC.
|
Use of improved manual weed control and plant spacing increased cost because of added labor requirements. Average added cost for this treatment (IFCM) ranged from $18–47 ha–1 (Table 7). The incremental added net return from improved weed control and plant spacing was, however, small in all seasons except Aman 2002 ($38 ha–1 in Aman 2002; $–2–8 ha–1 in other seasons) because of added costs and relatively small incremental increase in grain yield. The relatively high incremental added net return in Aman 2002 arose from relatively larger increases in yield with improved manual weed control and plant spacing in this season (Table 5).
The large ranges in added net returns (Table 7) suggest substantial variation in the benefit from use of the LCC. In the last three seasons when LCC use was optimal, >75% of the farmers obtained attractive positive net benefits from use of the LCC for N management without any other change in their fertilizer or crop management. A very small percentage of the farmers, however, obtained a negative added net return with use of the LCC. This negative added net return was always associated with lower yield and N use in the LCCN treatment as compared with FP. The few cases of lower yield and profit with the LCC likely resulted from poor crop management by the farmer leading to greater N limitations with use of the LCC than with FP, which received more N fertilizer.
The positive added net returns with use of the LCC in the vast majority of farmers' fields confirmed the LCC is ready for wide scale evaluation and promotion in Bangladesh. Our results suggest the LCC can be promoted for N management with no additional modification of the farmers' current fertilizer and crop management practices at locations where P and K fertilizer use is near the recommended rate and weed and crop management are not major constraints. The promotion of the LCC with recommended P, K, S, and Zn fertilization is best targeted to locations where farmers use insufficient K or where S or Zn limit yield.
The use of improved manual weed control and plant spacing together with the LCC for N management often resulted in an incremental increase in grain yield. However, the relative contributions of manual weed control and plant spacing to grain yield and N use efficiency cannot be determined from our study because the two factors were combined into one treatment. Adoption of improved manual weed control or plant spacing depends on availability of labor, wage rate, and opportunity costs of farmers' time. The opportunity cost of labor was relatively low in our study area. Such labor-demanding improved practices could, however, become less attractive in the future as wages and opportunity cost of labor increase with progress in economic development (Dawe et al., 2004).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Stevens, A. Wrather, M. Rhine, E. Vories, and D. Dunn Predicting Rice Yield Response to Midseason Nitrogen with Plant Area Measurements Agron. J., February 26, 2008; 100(2): 387 - 392. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
