Agronomy Journal 94:1393-1399 (2002)
© 2002 American Society of Agronomy
PRODUCTION PAPER
Rice Response to Phosphorus Fertilizer Application Rate and Timing on Alkaline Soils in Arkansas
Nathan A. Slaton*,a,
Charles E. Wilson, Jr.b,
Richard J. Normanc,
Sixte Ntamatungirob and
Donna L. Frizzelld
a Dep. of Crop, Soil, and Environmental Sciences, University of Arkansas, 1366 West Altheimer Drive, Fayetteville, AR 72704
b Rice Research and Extension Center, University of Arkansas, P.O. Box 351, Stuttgart, AR 72160
c Dep. of Crop, Soil, and Environmental Sciences, University of Arkansas 115 Plant Sci. Bldg., Fayetteville, AR 72701
d Southeast Research and Extension Center, University of Arkansas, P.O. Box 3508, Monticello, AR 71656
* Corresponding author (nslaton{at}uark.edu)
Received for publication December 14, 2001.
 |
ABSTRACT
|
|---|
Phosphorus deficiency of rice (Oryza sativa L.) in Arkansas occurs almost exclusively on alkaline silt loam soils. Phosphorus-deficient rice has been observed where P fertilizer was applied near the time of seeding, which suggests that fertilizer P is rapidly converted to a form not available to flooded rice on alkaline soils. The primary objective of this study was to evaluate rice response to P fertilizer applied at different times. Field studies were established in six commercial rice fields. Three rates of P (9.8, 19.6, and 39.1 kg P ha-1) were applied at four different times during the growing season including preemergence (PRE), preflood (PF), 5 to 10 d postflood (POF), or at midseason (MS) and compared with an untreated control. Significant grain yield increases were measured at two of the six locations. Grain yields were maximized by application of 19.6 kg P ha-1 at the two highly responsive sites with yield increases of 24 to 41%. Application of P fertilizer PRE, PF, and POF were superior to MS applications, which were not different from the control. Phosphorus concentration in harvested grain was not affected by time or rate of P fertilizer application. The average grain P content represented 56 to 75% of the total P in the aboveground portion of rice at physiological maturity. Broadcast applications of P fertilizer to the soil surface between seeding and active tillering were equally effective at increasing rice yields and optimizing P uptake on the P deficient soils.
Abbreviations: MS, midseason • PF, preflood • POF, postflood • PRE, preemergence
|
INTRODUCTION
|
|---|
PHOSPHORUS is generally most available to plants when the soil pH is between 6.0 and 6.5. When the soil pH is <6.0, the potential for P deficiency for most crops increases. The soil P is bound in compounds, such as Fe and Al phosphates, which are essentially insoluble under aerobic or upland conditions (Sanchez and Uehara, 1980). However, rice is grown under a continuous flood for most of the season that results in anaerobic soil conditions. In anaerobic soils, the Fe phosphate compounds are reduced and are subsequently converted to a soluble, plant-available form. The increased P availability, due to Fe reduction, is normally more than adequate to supply sufficient P to rice (Shahandeh et al., 1994). On alkaline soils, Ca phosphates are the predominate pool of soil P and their solubility is not directly influenced by redox reactions. Instead, their solubility may increase after flooding when the soil pH decreases due to CO2 accumulation and Ca2+ ions are complexed by organic acids (Sanyal and De Datta, 1991). Regardless of the mechanism that increases P availability after flooding, the time required for P release and the magnitude of the soil solution P concentration varies among soils and may take several weeks to peak. If the soil solution P concentration is inadequate for optimum rice growth during the first several weeks after flooding, yield reductions may occur, regardless of the soil solution P concentration later in the growing season.
In Arkansas, rice yield responses to P fertilization have been found almost exclusively on alkaline soils or soils disturbed by land leveling. On the undisturbed, alkaline soils, soil pH has shown to be a better predictor of rice response to P fertilization than Mehlich 3 extractable P (Moore et al., 1994; Wilson et al., 2000). Vendrell and Sabbe (1990) found that Mehlich 3 extractable P did not increase 16 or 1440 h after fertilizer P was applied to an alkaline Calloway silt loam (fine-silty, mixed, thermic, Glossaquic Fragiudalfs), suggesting that this soil, which is a common soil used for rice production in Arkansas, was capable of fixing large quantities of fertilizer P in a very short time. When water-soluble P fertilizers are added to soil, meta-stable reaction products form and with time, less soluble and more stable compounds are formed (Sample et al., 1980). Triple superphosphate (200 g P kg-1) is the principal P fertilizer used for broadcast, preplant applications to soil used for rice production in Arkansas. Calcium phosphates are one of the primary initial reaction products formed after P fertilizer is applied to the soil, especially when free CaCO3 is present (Lindsay et al., 1962). The rapidity of these fertilizer reactions in the soil and the time of fertilizer application in relation to peak nutrient needs of the crop influence the availability of the fertilizer P to the crop.
Patrick et al. (1974) showed that P fertilizer drilled with the seed (i.e., banded) produced higher yields than broadcast-applied P 2 wk after seeding. Compared with P banded with seed, rice yields decreased as the time of P application was delayed to 4 and 6 wk after seeding suggesting that P was needed before tillering to produce optimum yields. Despite the reported effectiveness of P fertilizer drilled with the seed, little if any P fertilizer is applied with the seed in U.S. rice production. Most P fertilizer is broadcast onto the soil surface and incorporated or left on the soil surface when rice is drill-seeded into an untilled seedbed.
We have observed P-deficient rice on fields receiving preplant, broadcast P fertilizer, which suggests that the timing of P fertilizer application may need to be changed to maximize crop use of the P fertilizer and avoid possible P deficiency. Knowledge of how rice responds to the P fertilizer application time during the growing season is essential for the development of efficient P fertilizer recommendations. Other than the research published by Patrick et al. (1974), the effect of broadcast P fertilizer application time on P uptake and yield of rice in the USA has not been published. Our hypothesis was that application of P fertilizer to seedling rice, rather than before seedling emergence, would increase rice yields and improve P uptake on alkaline silt loams soils. The study objective was to evaluate the effect of P fertilizer application rate and time on P uptake and grain yield of flood-irrigated rice.
|
MATERIALS AND METHODS
|
|---|
Studies to evaluate the effect of P fertilizer rate and application timing on rice were conducted in six commercial production fields during 1997 and 1998 (Table 1). Silt loam soils with neutral or alkaline soil pH were desired for these studies. Field observations and previous research (Wilson et al., 2000) suggests that rice grown on alkaline soils is most likely to respond to P fertilizer application. County extension agents identified farm cooperators and fields that fit this criterion. Soybean was the previous crop in all fields except the Roberts-98 field where rice followed rice in the rotation. A 800-m2 area in each field was marked with flags so P fertilizer, broadcast-applied by the grower to the production field, would not be applied to the designated plot area. ‘Bengal’, a medium-grain rice cultivar, was drill-seeded at each location at a rate of approximately 130 kg seed ha-1. After levees were constructed by the producer, plots were established in the unfertilized area of the field. Each individual plot contained 12, 4.8 m long rows of rice with 17.8-cm row spacings. Composite soil samples were collected from the untreated control plot (NONE, 0 kg P ha-1) in each replication for routine soil analysis. The soils were oven-dried, ground, and placed through a 2-mm sieve before analysis. Soil pH was determined on a 1:2 soil/water suspension with a glass electrode. Extractable Ca, Mg, K, and P were determined by the Mehlich 3 procedure (Mehlich, 1984) and analyzed by inductively coupled atomic plasma spectroscopy (ICAP, Soltanpour et al., 1996). Selected soil properties are listed in Table 1. Based on the University of Arkansas P fertilizer recommendations, P fertilizer (approximately 20 kg P ha-1) would have been recommended for soybean at all six locations and for rice on five of the six study sites with the Wimpy-97 site being the lone exception. Direct application of P fertilizer is recommended to rice when soil pH is >6.5 and Mehlich 3–extractable P, measured by ICAP, is <25 mg P kg-1.
View this table:
[in this window]
[in a new window]
|
Table 1. Selected soil characteristics from sites utilized for P fertilizer studies conducted in 1997 and 1998 in Arkansas.
|
|
Phosphorus fertilizer (triple superphosphate, 200 g P kg-1) was applied at rates of 0, 9.8, 19.6, and 39.1 kg P ha-1. Phosphorus fertilizer was broadcast-applied before seedling emergence (PRE), preflood (PF) at the 4- to 5-leaf stage, 5 to 10 d postflood (POF), or at midseason (MS) when the main stems' topmost internode had elongated 0.5 to 1.5 cm. The first two P application timings, PE and PF, were made to dry or moist soils. The POF and MS fertilizer applications were made into the floodwater. The PRE, PF, POF, and MS applications were approximately 5 to 10, 25 to 30, 35 to 40, and 65 to 70 d after seeding, respectively. Because the studies were conducted in grower fields, preplant incorporated P applications were not included. The PRE application adequately represents the fertilization practices of a large portion of the rice land area in Arkansas grown using reduced spring tillage with P and K fertilizers broadcast applied to the soil surface before or after seeding. With the exception of P fertilization, the plots were managed by the cooperating producer according to management practices utilized for the remainder of the field. Management with respect to fertilization, irrigation, and pest control was similar to guidelines recommended by the Cooperative Extension Service for the dry-seeded, delayed flood rice cultural system (Slaton, 2001). Potassium and Zn fertilizers were applied to the plot area as recommended by routine soil analysis.
At 3 wk after 50% heading, all of the aboveground plant tissue was harvested from a 1-m section of the second inside row for determination of total dry matter, straw and panicle tissue P concentrations, and total P uptake. Tissues were oven-dried at 60°C until a constant moisture was reached. After drying, samples were weighed and the panicles were removed from the stems at the top node. The individual component weights for panicles and rice straw (leaf and stalk tissue) were weighed, ground to pass a 0.85-mm sieve, and digested with concentrated HNO3 and 30% H2O2 (Jones and Case, 1990). The digests were analyzed for P by ICAP (Soltanpour et al., 1996). At maturity, grain yields were measured by harvesting 2.6 m2 of the four center rows with a small-plot combine. Grain yields were adjusted to a uniform moisture content of 120 g kg-1. Total P uptake by rice straw and grain were calculated using the straw or grain P concentrations and the straw or harvested grain weights. The efficiency of P fertilizer uptake was calculated by difference.
Each experiment was arranged in a randomized complete block, 3 (P application rate) x 4 (time of application) factorial design and was compared with an untreated control. Each treatment was replicated four times. Locations were analyzed separately. Analysis of variance procedures were conducted with the PROC GLM procedure in SAS (SAS Inst., Cary, NC). Mean separations were performed by Fisher's protected least significant difference method at a significance level of 0.05.
|
RESULTS AND DISCUSSION
|
|---|
Grain Yields
The two-way interaction between P application rate and time was not statistically significant (P 
0.05) for rice grain yield at any location. The main treatment effects of P fertilizer rate (Table 2) (averaged across P application time) and P fertilizer application time (Table 3) (averaged across P application rate) significantly affected grain yields at some locations. At the Davis-97 and Davis-98 locations, all P fertilizer rates significantly increased rice yields above the control (Table 2). Application of 19.6 kg P ha-1 maximized grain yields, producing 24 to 41% more grain than the control, at these two highly responsive sites. Grain yields at the Roberts-98 (P = 0.132) and Wimpy-97 (P = 0.345) locations were not significantly affected by P fertilizer rate, but did show a trend for yields to increase when P fertilizer was applied. At the Brooks-97 location, grain yields tended to decrease when P fertilizer was applied, but only the 19.6 kg P ha-1 rate had significantly lower yields than the control (Table 2). At the Wimpy-98 site, application of 19.6 kg P ha-1 produced the lowest numerical yield, but was significantly lower than only 39.1 kg P ha-1. Place et al. (1970) and Wilson et al. (1997) also reported that application of P fertilizer at seeding significantly decreased rice grain yields.
View this table:
[in this window]
[in a new window]
|
Table 2. Influence of P fertilizer application rate, averaged across P application times, on rice grain yields at six locations during 1997 and 1998 in Arkansas.
|
|
View this table:
[in this window]
[in a new window]
|
Table 3. Influence of P fertilizer application time, averaged across P fertilizer application rates, on rice grain yields at six locations during 1997 and 1998 in Arkansas.
|
|
Phosphorus fertilizer applied PRE, PF, and POF produced significantly higher yields than the control at Davis-97 and Davis-98 (Table 3). At Davis-97, P applied at MS did not significantly increase yields higher than the control and resulted in yields significantly lower than when P was applied PRE and POF. Similarly, MS application of P at Wimpy-98 produced significantly lower yields than P applied PF and POF, but was not significantly different from NONE and PRE applications. The Brooks-97 (P = 0.791), Wimpy-97 (P = 0.360), and Roberts-98 (P = 0.070) locations showed no consistent or significant trends for grain yield response to P application time.
Patrick et al. (1974) suggested that P fertilizer had to be applied at seeding to produce optimum yields, but our data shows P fertilizer can be applied until beginning tillering without yield loss. When P nutrition is limiting, MS applications of P fertilizer may nominally increase grain yields, but are too late for production of maximum rice yields. Similar rice yield responses have been documented for MS N fertilizer applications when PF N was either mismanaged or inadequate (Wilson et al., 1998). It is not known from these data if applications of P fertilizer during later stages of reproductive growth (i.e., booting or heading) are beneficial to grain yields under P limiting situations or possibly detrimental when P availability is sufficient.
Although a significant yield increase at the Davis-98 site was observed when P was applied POF compared with PRE, yield data from these studies does not provide conclusive evidence to recommend P fertilizer applications be made closer to the time of flooding in the direct-seeded, delayed flood, rice production system on soils that are classified as highly responsive to P fertilization. We have observed P deficiency of rice where P was applied near seeding or the previous fall on soils similar to those represented in this study; however, our yield data does not provide conclusive evidence that P applied before seedling emergence is ineffective. However, the data conclusively shows that P fertilizer can be applied several weeks after seeding without reducing rice grain yield. Alternative times (i.e., other than before seeding) to apply P fertilizer to rice would not be an inconvenience to growers since nearly all of the seasonal N fertilizer requirement is applied by aircraft. When only N and P are needed to optimize yields, both fertilizers could be blended and applied in a single application immediately before flooding.
Phosphorus Concentration in Rice Tissues at Physiological Maturity
Phosphorus concentration in the harvested grain was not affected by the time or rate of P fertilizer application (data not shown). The mean grain P concentrations were 2.81 g P kg-1 at Brooks-97, 3.29 g P kg-1 at Davis-97, 2.93 g P kg-1 at Wimpy-97, 2.55 g P kg-1 at Davis-98, and 3.06 g P kg-1 at Wimpy-98 and are similar to published rice grain P concentrations (Nelson, 1980). Grain samples from the Roberts-98 location were lost before analysis. The concentration of P in the grain at 3 wk after 50% heading or physiological maturity was about twice that of the rice straw (Tables 4 and 5).
View this table:
[in this window]
[in a new window]
|
Table 4. Influence of P fertilizer application rate, averaged across P fertilizer application times, on rice straw P concentration 3 wk after 50% heading at six locations in 1997 and 1998 in Arkansas.
|
|
View this table:
[in this window]
[in a new window]
|
Table 5. Influence of P fertilizer application time, averaged across P fertilizer application rates, on rice straw P concentration 3 wk after 50% heading at six locations in 1997 and 1998 in Arkansas.
|
|
Straw P concentration was only influenced by the main effects of P fertilizer application time and rate at about one-half of the locations (Tables 4 and 5). At the Davis-98, Roberts-98, and Wimpy-98 locations, straw P concentration tended to increase or significantly increased as P fertilizer rate increased (Table 4). Fertilizer P was either taken up by the plant or stimulated uptake of native soil P at these three locations.
Straw P concentration increased as the time of application was delayed at the Davis-97 location and, to some degree, at the Wimpy-97 location, but no significant response was measured at the Brooks-97 and Roberts-98 locations (Table 5). Straw P concentration did not follow a consistent trend for time of application at the Wimpy-98 site. Where significant differences in tissue P concentration occurred, P concentration tended to be numerically higher when the P fertilizer was applied during rapid vegetative growth (i.e., PF, POF, and MS).
The numerical differences in straw P concentration among the locations may be attributed to differences in yields, slight variations in sampling time, the relative availability of native soil P, or a combination of some or all of these factors. The Davis-97 location showed the greatest grain yield response to P fertilization and tended to have the lowest straw P concentrations of the six locations. In contrast, Brooks-97 showed no positive yield response to P fertilization and also had the highest straw P concentration of the untreated controls. The relationship between the untreated control straw P concentration and the relative yield of the untreated control for the six locations was not significant (P = 0.176, data not shown). However, our data clearly shows that the critical straw concentration of 0.60 g P kg-1 at maturity reported by Dobermann and Fairhurst (2000) is not appropriate for rice grown in Arkansas.
Phosphorus Content of Rice Tissues at Physiological Maturity
Phosphorus fertilization significantly affected the P content of rice straw only at the Davis-98 location, where there was a significant interaction between application time and rate (Table 6). Straw P content at the Davis-98 location tended to increase as P fertilizer rate increased when applied PRE, but not at the other application times, which were statistically and numerically similar. The straw P contents of 39.1 kg P ha-1 applied PRE, 39.1 kg P ha-1 applied PF, and 19.6 kg P ha-1 applied POF were significantly greater than the control. The average straw P content at the other locations was 18.1 kg P ha-1 at Brooks-97, 8.8 kg P ha-1 at Davis-97, 12.3 kg P ha-1 at Wimpy-97, 10.7 kg P ha-1 at Roberts-98, and 10.5 kg P ha-1 at Wimpy-98.
View this table:
[in this window]
[in a new window]
|
Table 6. Effect of the P fertilizer application time and rate interaction on total straw P content at physiological maturity at the Davis-98 location in Arkansas.
|
|
The P contents of harvested rice grain were not affected by P rate, time of application, or the two-way interaction at any location, except at Davis-97, where the main effect of time of P application was significant (data not shown). Total P removed by harvested rice grain followed the same general order as that observed for the effect of P application time on grain yields (Table 3). The highest grain yields at Davis-97 were achieved when P was applied at PRE, PF, and POF and also generally removed more P than yields produced when P fertilizer was applied at MS or when NONE was applied. Phosphorus fertilizer applications made PRE (26.5 kg P ha-1) and POF (27.1 kg P ha-1) contained significantly greater grain P contents than MS (20.5 kg P ha-1) P applications and the untreated control (19.7 kg P ha-1). The grain P content of PF P applications (23.7 kg P ha-1) was intermediate and not significantly different from the other P application times.
Grain P content ranged from 23.9 to 28.7 kg P ha-1 across locations when maximum grain yields were achieved, except at Davis-98, which had 21.8 kg P ha-1 associated with the highest grain yield. In contrast, straw P content ranged from 8.3 to 19.7 kg P ha-1 across locations when maximum grain yields were produced. Closer examination of the straw P contents reveals that the two locations, Brooks-97 and Davis-98, with the highest grain yields (i.e., maximum grain yields approximately 9000 kg ha-1) had the highest straw P contents of around 19 kg P ha-1. Conversely, the other locations with maximum grain yields of 6500 to 8000 kg ha-1 had much lower straw P contents of only 8.0 to 12.9 kg P ha-1. This suggests that straw P content at maturity or the partitioning of P between grain and straw may be related to the grain yield of rice and probably indicates a yield-limiting factor. If another yield-limiting factor was present, it may or may not be associated with soil P availability, but certainly affected total P uptake by the rice plant.
The Davis-97 and Wimpy-97 locations were the only sites with a significant effect of P application time on total P uptake (Table 7). The PRE, PF, and POF P application times that had the most influence on straw P concentration, total dry matter, and grain yield also had the greatest influence on total P uptake at the Davis-97 and Wimpy-97 locations. Although no significant effect of P application time on total P uptake by rice was found at the other locations, P applied at these three times at the Davis-98 and Wimpy-97 locations usually resulted in significant or numerically higher total P uptakes than when P was applied at MS or not applied.
View this table:
[in this window]
[in a new window]
|
Table 7. Influence of P fertilizer application time, averaged across P fertilizer application rates, on total P uptake by rice (i.e., grain + straw) at physiological maturity in 1997 and 1998 in Arkansas.
|
|
Total P uptake by rice across all locations ranged from 27.7 to 48.4 kg P ha-1 (Table 7). The two locations, Brooks-97 and Davis-98, with maximum grain yields >9000 kg ha-1 had the greatest total P uptakes (>40 kg P ha-1) of all the locations. The other locations with maximum grain yields of 6500 to 8000 kg ha-1 had total P uptakes generally in the range of 30 to 35 kg P ha-1. Fageria et al. (1997) and Terman and Allen (1970) determined that grain P content represented 60 to 75% of the total aboveground P content of rice at maturity. The average grain P contents in our studies were 60 to 75% of the total P accumulated by rice at the locations with maximum grain yields of 6500 to 8000 kg ha-1. However, the two locations with grain yields >9000 kg ha-1 had <60% of the total P taken up residing in the grain.
Fertilizer P uptake, calculated by difference across P rates using total P uptakes in Table 7, showed that the two responsive sites, Davis-97 and Davis-98, had the highest uptakes of fertilizer P (data not shown). Fertilizer P uptake was relatively constant when P fertilizer was applied PRE, PF, and POF, but tended to decline when applied at MS. In contrast, P fertilization actually reduced grain yield and total dry matter at maturity (data not shown) at the Brooks-97 site and showed a net negative uptake of fertilizer P or native soil P. The average uptake of applied P fertilizer, averaged across P application rates and times, was -9.9% for Brooks-97, 20.1% at Davis-97, 8.6% at Davis-98, 8.6% at Wimpy-97, and 1.5% at Wimpy-98. The recovery of fertilizer P tended to decline when P was applied at MS. The average recovery of applied P fertilizer found in our studies was generally comparable to P fertilizer recovery efficiencies stated by De Datta (1981) and Terman and Allen (1970).
|
GENERAL DISCUSSION
|
|---|
Four of the soils in our studies did not show significant grain yield or P uptake increases to P fertilization indicating that the native soil P availability was adequate for the duration of the season. Routine soil testing (Table 1) failed to show clear relationships between relative grain yield and total P uptake and Mehlich 3–extractable P, soil pH, other Mehlich 3–extractable nutrients, or combinations of these factors on these six soils. Specific reasons explaining why some neutral to alkaline silt loams respond to P fertilization and others do not cannot be provided from this data. However, the availability of P is certainly related to the relative concentrations of Ca, P, or both in the soil solution (Ponnamperuma, 1972; Sah and Mikkelsen, 1986). The role of Ca on P availability is further substantiated by Ntamatungiro et al. (1999), who observed rice tissue P concentrations decreased and tissue Ca concentrations increased during active tillering (about 2 wk after flooding) as lime rate increased on a DeWitt silt loam. Additionally, tissue P concentration increases during the first 2 wk after flooding as P fertilizer rate increases on both acid and alkaline soils in Arkansas (unpublished data, 2001). However, rice tissue P concentrations are generally equal among P fertilizer rates as soon as 5 wk after flooding, suggesting that P in the soil solution is high enough to satisfy the plant uptake rate and the effect of P fertilizer rate is diminished (Slaton et al., 2000).
The trends in rice tissue P concentration from early vegetative to reproductive growth suggest that P availability is limiting only during early vegetative growth on these soils. Thus, the availability of soil P and the need for P fertilization may best be evaluated by measuring rice tissue P concentration between the five-leaf stage and the midtillering growth stage (approximately 14 d after flooding). Although only about 30% of the total plant P at maturity is absorbed by midseason (panicle initiation) (Fageria et al., 1997), adequate availability of P during the first 5 wk, and especially during the first 2 wk, after flooding rice grown in the direct-seeded, delayed flood cultural system is critical to establish high yield potential for rice. Measurement of total P uptake at physiological maturity does not accurately describe the availability of P during vegetative growth or account for the eventual increase in P availability after prolonged flooding. However, it appears that small increases in P uptake, presumably during the first 2 to 3 wk after flooding seedling rice, has a significant effect on rice yield potential.
Further research is needed to more accurately characterize the effect of P application rate and time on seasonal rice uptake of P and rice grain yields on soils where rice responds to P fertilization. Monitoring soil solution P or plant P uptake continuously during the season rather than only at maturity would probably improve our understanding of P availability, the efficiency of P fertilizer uptake as affected by time and rate of application, and possibly lead to a better means of identifying the soils that require P fertilization to maximize rice yields.
|
SUMMARY
|
|---|
Phosphorus fertilization increased rice grain yields above the untreated control at two of the six locations. Applications of 19.6 kg P ha-1 made PRE, PF, and POF maximized yields at the two highly responsive locations and are within the range of current P fertilizer application rate recommendations (Slaton, 2001). Phosphorus fertilizer applied at MS or the onset of reproductive growth was too late to maximize grain yields, but produced yields numerically and sometimes significantly greater than the unfertilized control at the two responsive sites. The two highly responsive sites had the greatest average fertilizer P recovery, which ranged from 9 to 20% of the applied P. In Arkansas, harvested rice grain removes 20 to 30 kg P ha-1, which generally represents 56 to 75% of the total aboveground P uptake and is roughly equal to the amount of P fertilizer required to maximize grain yields. Soil chemical properties, determined by routine soil analysis, from these six study sites failed to show consistent relationships with grain yield or total P uptake that could be used to predict the soils that require P fertilization when cropped to rice.
The results of this study indicate that broadcast applications of P fertilizer to the soil surface between seeding and active tillering were equally effective at increasing rice yields and optimizing P uptake on P deficient soils. This is in contrast to the findings of Patrick et al. (1974), who suggested P must be applied at seeding to maximize rice yields. Application of P fertilizer between seeding and flooding in the dry-seeded, delayed-flood management system allows growers more flexible fertilization options that can potentially reduce commercial fertilizer application costs when only N and P are required to maximize rice yields. When P fertilizer is not applied during this period and P deficiency occurs, nominal rice yield increases may occur when P is applied by panicle initiation, but maximum yields will not be produced.
|
ACKNOWLEDGMENTS
|
|---|
This work was funded by the Arkansas Rice Research and Promotion Board, the Potash and Phosphate Institute, and the Arkansas Fertilizer Tonnage Fees. The authors extend special thanks to the county extension agents and the cooperating rice producers for their invaluable assistance.
|
NOTES
|
|---|
Contribution of the University of Arkansas Agric. Exp. Stn. Manuscript No. 01097.
|
REFERENCES
|
|---|
- De Datta, S.K. 1981. Principles and practices of rice production. John Wiley & Sons, New York.
- Dobermann, A., and T. Fairhurst. 2000. Rice: Nutrient disorders and nutrient management. Potash and Phosphate Inst., Potash and Phosphate Inst. of Canada, IRRI, Singapore and Los Banos.
- Fageria, N.K., A.B. Santos, and V.C. Baligar. 1997. Phosphorus soil test calibration for lowland rice on an Inceptisol. Agron. J. 89:737–742.[Abstract/Free Full Text]
- Jones, J.B., and V.W. Case. 1990. Sampling, handling, and analyzing plant tissue samples. p. 389–428. In R.L. Westerman (ed.) Soil testing and plant analysis. 3rd ed. SSSA Book Series 3. SSSA, Madison, WI.
- Lindsay, W.L., A.W. Frazier, and H.F. Stephenson. 1962. Identification of reaction products from phosphate fertilizers in soils. Soil Sci. Soc. Am. Proc. 26:446–452.
- Mehlich, A. 1984. Mehlich 3 soil test extractant: A modification of Mehlich 2 extractant. Commun. Soil Sci. Plant Anal. 15:1409–1416.
- Moore, P.A., Jr., Y. Sun, D.M. Miller, C.A. Beyrouty, and R.J. Norman. 1994. Phosphorus deficiency in rice on high pH soils. p. 160–167. In W.E. Sabbe (ed.) Arkansas soil fertility studies 1993. Res. Ser. 436. Arkansas Agric. Exp. Stn., Fayetteville, AR.
- Nelson, L.E. 1980. Phosphorus nutrition of cotton, peanuts, rice, sugarcane, and tobacco. p. 693–736. In F.E. Khasawneh et al. (ed.) The role of phosphorus in agriculture. ASA, CSSA, and SSSA, Madison, WI.
- Ntamatungiro, S., N.A. Slaton, C.E. Wilson, J.F. Robinson, L. Ashlock, M. Daniels, and T. Windham. 1999. Effects of lime, phosphorus, and zinc application on rice and soybean production. p. 268–276. In R.J. Norman and T.H. Johnston (ed.) B.R. Wells Rice Res. Ser. 1998. Res. Ser. 468. Arkansas Agric. Exp. Stn., Fayetteville, AR.
- Patrick, W.H., Jr., F.J. Peterson, and F.E. Wilson. 1974. Response of lowland rice to time and method of application of phosphate. Agron. J. 66:459–460.[Abstract/Free Full Text]
- Place, G.A., B.R. Wells, J.L. Sims, and V.L. Hall. 1970. Effects of nitrogen and phosphorus fertilization on growth, yield and cooking qualities of rice. Agron. J. 62:239–243.[Abstract/Free Full Text]
- Ponnamperuma, F.N. 1972. The chemistry of submerged soils. Adv. Agron. 24:29–96.
- Sah, R.N., and D.S. Mikkelsen. 1986. Transformations of inorganic phosphorus during flooding and draining cycles of soil. Soil Sci. Soc. Am. J. 50:62–67.[Abstract/Free Full Text]
- Sample, E.C., R.J. Roper, and G.J. Racz. 1980. Reactions of phosphate fertilizers in soils. p. 263–310. In F.E. Khasawneh et al. (ed.) The role of phosphorus in agriculture. ASA, CSSA, and SSSA, Madison, WI.
- Sanchez, P.A., and G. Uehara. 1980. Management considerations for acid soils with high phosphorus fixation capacity. p. 471–514. In F.E. Khasawneh et al. (ed.) The role of phosphorus in agriculture. ASA, CSSA, and SSSA, Madison, WI.
- Sanyal, S.K., and S.K. De Datta. 1991. Chemistry of phosphorus transformations in soil. Adv. Soil Sci. 16:1–120.
- Shahandeh, H., L.R. Hossner, and F.T. Turner. 1994. Phosphorus relationships in flooded rice soils with low extractable phosphorus. Soil Sci. Soc. Am. J. 58:1184–1189.[Abstract/Free Full Text]
- Slaton, N.A. 2001. Rice production handbook. Misc. Publ. 192. Arkansas Coop. Ext. Serv., Univ. of Arkansas, Little Rock, AR.
- Slaton, N.A., C.E. Wilson, Jr., S. Ntamatungiro, R.J. Norman, and D.L. Boothe. 2000. Effects of previous crop and phosphorus fertilization rate on rice. p. 294–303. In R.J. Norman and C.A. Beyrouty (ed.) B.R. Wells Rice Res. Ser. 1999. Res. Ser. 476. Arkansas Agric. Exp. Stn., Fayetteville, AR.
- Soltanpour, P.N., G.W. Johnson, S.M. Workman, J.B. Jones, Jr., and R.O. Miller. 1996. Inductively coupled plasma emission spectrometry and inductively coupled plasma–mass spectroscopy. p. 91–140. In D.L. Sparks (ed.) Methods of soil analysis. Part 3. SSSA Book Series 5. SSSA, Madison, WI.
- Terman, G.L., and S.E. Allen. 1970. Fertilizer and soil P uptake by paddy rice, as affected by soil P level, source and date of application. J. Agric. Sci. 75:547–552.
- Vendrell, P.F., and W.E. Sabbe. 1990. Evaluation of a fast method for determination of P and K fixation. Commun. Soil Sci. Plant Anal. 21:1277–1297.
- Wilson, C.E., Jr., P.K. Bollich, and R.J. Norman. 1998. Nitrogen application timing effects on nitrogen efficiency of dry-seeded rice. Soil Sci. Soc. Am. J. 62:959–964.[Abstract/Free Full Text]
- Wilson, C.E., Jr., N.A. Slaton, S. Ntamatungiro, D.L. Frizzell, R.J. Norman, and W.B. Koen. 2000. Soil pH and Mehlich 3 extractable phosphorus as indices of phosphorus availability to paddy rice. p. 123–124. In Proc. Rice Tech. Working Group, 28th, Biloxi, MS. 27 Feb.–1 Mar., Louisiana State Univ. Agric. Center and Exp. Stn., Rice Res. Stn., Crowley, LA.
- Wilson, C.E., Jr., N.A. Slaton, S. Ntamatungiro, R.J. Norman, D.L. Frizzell, and B.R. Wells. 1997. Phosphorus fertilizer management on alkaline soils. p. 142–146. In R.J. Norman and T.H. Johnston (ed.) B.R. Wells Rice Res. Studies 1996. Res. Ser. 456. Arkansas Agric. Exp. Stn., Fayetteville, AR.
TOP
ABSTRACT
INTRODUCTION
