Journal list menu

Volume 108, Issue 5 p. 1938-1943
Crop Economics, Production & Management
Open Access

Nitrogen Strategy and Seeding Rate Affect Rice Lodging, Yield, and Economic Returns in the Midsouthern United States

Jennifer L. Corbin

Jennifer L. Corbin

Clemson University, Clemson, SC, 29632

Search for more papers by this author
John M. Orlowski

John M. Orlowski

Delta Research and Extension Center, Mississippi State University, Stoneville, MS, 38776

Search for more papers by this author
Dustin L. Harrell

Dustin L. Harrell

Rice Research Station, Louisiana State University, Crowley, LA, 70578

Search for more papers by this author
Bobby R. Golden

Corresponding Author

Bobby R. Golden

Delta Research and Extension Center, Mississippi State University, Stoneville, MS, 38776

Corresponding author ([email protected]).

Search for more papers by this author
Larry Falconer

Larry Falconer

Delta Research and Extension Center, Mississippi State University, Stoneville, MS, 38776

Search for more papers by this author
L. Jason Krutz

L. Jason Krutz

Delta Research and Extension Center, Mississippi State University, Stoneville, MS, 38776

Search for more papers by this author
Jeffrey Gore

Jeffrey Gore

Delta Research and Extension Center, Mississippi State University, Stoneville, MS, 38776

Search for more papers by this author
Michael S. Cox

Michael S. Cox

Department of Plant and Soil Sciences, Mississippi State University, Starkville, MS, 39762

Search for more papers by this author
Timothy W. Walker

Timothy W. Walker

Horizon Ag. LLC, Memphis, TN, 38125

Search for more papers by this author
First published: 01 September 2016
Citations: 15

Available freely online through the author-supported open access option

Abstract

Seeding rate and N application strategies have been extensively researched for drill-seeded, delayed flood rice (Oryza sativa L.) production in the midsouthern United States. However, little economic analysis has been performed to determine the effects of seeding rate and N fertilization strategies on high-yielding, lodging susceptible rice cultivars. The purpose of this research was to determine the effect of multiple N fertilization strategies and seeding rates on rice lodging and grain yield and use the data to determine optimal rice management practices based on evaluation of economic returns. Studies were established in Louisiana and Mississippi during the 2010 and 2012 growing seasons. Rice was drilled at seeding rates of 161, 323, and 483 seeds m−2. Rice was fertilized with N at multiple rates and timings, representing recommended strategies as well as strategies designed to reduce lodging. Lodging and grain yield data were collected and used to determine net returns at high and low rice and N prices. Rice grain yield and net returns were maximized at a seeding rate of 323 seeds m−2. Splitting N application between early-season (V3) and late-season (R1 or R3) applications decreased lodging, but resulted in decreased grain yield compared to optimal rates of N applied at V3. The 202 kg N ha−1 applied early-season (V3) N fertilization strategy maximized both grain yield and net returns, despite relatively high levels of lodging. Optimal N fertilization during early-vegetative growth appears to be necessary to maximize yield potential in this rice production system.

Core Ideas

  • Split N application in rice limits lodging but also reduce grain yield.

  • High N rates and seeding rates increase lodging and harvest costs.

  • Appropriate early-season (preflood) N is necessary to maximize rice grain yield.

Abbreviations

  • ACH
  • adjusted harvest costs
  • DC
  • drying cost
  • DREC
  • Delta Research and Extension Center
  • FAP
  • fertilizer application cost
  • FP
  • nitrogen cost
  • FR
  • nitrogen fertilizer applied to plot
  • H
  • hauling cost
  • HC
  • harvest costs
  • NRAC
  • net returns above costs
  • P
  • rice price
  • RRS
  • Rice Research Station
  • Y
  • rice plot yield
  • One of the most important management decisions for delayed-flood rice production in the midsouthern United States is N fertilization strategy. Recommendations for N fertilizer in both Louisiana and Mississippi vary by cultivar and soil texture, but both states generally recommend the majority of N (100–180 kg N ha−1) be applied preflood with the remainder (34–67 kg N ha−1) top-dressed at midseason. Historically, rice has produced optimum grain yields when N is applied with a split application (Wells and Johnston, 1970; Reddy and Patrick, 1976; Wells and Turner, 1984; Patrick et al., 1985; Brandon and Wells, 1986; Wescott et al., 1986; Wilson et al., 1998). However, more recent research indicates that rice uses N more efficiently and produces greater grain yields when applied in a single preflood application (Bollich et al., 1994; Norman et al., 2000; Slaton et al., 2003; Bond et al., 2008). Applying high rates of N in a single application can cause unwanted agronomic effects such as greater plant height which can result in potential grain yield loss due to lodging (Gravois and Helms, 1996; Slaton et al., 2003; Bond et al., 2008; Harrell and Blanche, 2010).

    Another major management decision for midsouthern U.S. rice production is seeding rate. Seeding rate decisions affect a number of important plant characteristics including plant population, tiller production, and plant height which directly affects key agronomic characteristics, especially lodging and grain yield. Recommended seeding rates for Louisiana rice production vary by variety, but range from 215 to 323 seeds m−2. This allows for a final plant population density of 108 to 161 seedlings m−2 (Saichuck et al., 2008; Harrell and Blanche, 2010). Research with older inbred cultivars suggests low seeding rates do not negatively impact grain yield (Jones and Snyder, 1987; Gravois and Helms, 1992; Ottis and Talbert, 2005; Harrell and Blanche, 2010) largely due to greater levels of tillering at low seeding rates (Schnier et al., 1990; Counce et al., 1992). Research conducted with more recent rice inbred cultivars suggests that low seeding rates can have a negative impact on rice grain yields (Bond et al., 2005, 2008; Harrell and Blanche, 2010). Similar to N fertilization, increasing seeding rates can result in increased grain yield, but can also lead to undesirable agronomic characteristics such as weaker culms and increased plant height which may increase lodging (Bond et al., 2008). Lodging can create many problems at harvest, including decreased harvest efficiency, reduced grain quality, and the potential for reduced yield (Walker et al., 2008; Salassi et al., 2013).

    Since both N management and seeding rate decisions can have substantial impacts on both grain yield and lodging in rice, one objective of this research was to investigate the effect of multiple seeding rates and N fertilization strategies on lodging and grain yield of a high-yielding, lodging susceptible cultivar in drill-seeded, delayed-flood rice production. While previous studies have investigated both N strategy and seeding rates in rice, few studies have attempted to apply economic analyses to quantify the effects of agronomic decisions on grower profitability. Walker et al. (2006) used economic analysis to determine net returns for various N fertilization strategies at multiple N prices. While Walker et al. (2006) was informative, the economic analysis only took into account expenses associated with N application and differences in grain yield between treatments. While N costs and grain yield are major drivers of economic returns, other agronomic considerations such as lodging can potentially influence net returns realized by rice producers. Therefore, a second objective of this research was to use economic analysis to quantify the effect of lodging as well as grain yield in determining net returns associated with multiple N application strategies at multiple seeding rates for drill-seeded, delayed-flood rice production in the midsouthern United States.

    MATERIALS AND METHODS

    Field studies were conducted at two sites in 2010 and one site in 2012. One study site in 2010 and the site in 2012 was established at the Mississippi State University Delta Research and Extension Center (DREC) (33°25’ N, 90°54’ W) on a Tunica clay (clayey over loamy, mixed, superactive, nonacid, thermic, Vertic Epiaquert). The other study site in 2010 was established at the Louisiana State University Rice Research Station (RRS) (30°24’ N, 92°35’ W) on a Crowley silt loam (fine, smectitic, thermic Typic Albaqualf). The preceding crop was soybean [Glycine max (L.) Merr.] in all site-years.

    Clearfield (BASF, Ludwigshafen, Germany) long grain rice variety CL151 (Reg. no. CV-133, PI 654463) was seeded in all site-years. CL151 was chosen because it is a popular, high-yielding variety that is highly susceptible to lodging. Seeds were individually packaged on a per plot basis to achieve the desired seeding rates of 161, 323, and 483 seeds m−2. Plots were seeded with a small-plot grain drill (Great Plains Manufacturing Inc., Salina, KS) equipped with a Hege Belt Cone (Hege, Wintersteiger Inc., Salt Lake City, UT) calibrated to evenly distribute seed over a known distance. Individual plots were 4.6 m in length and consisted of eight rows spaced 0.2 m apart. Seeding occurred on 14 March at RRS, 28 April at DREC in 2010, and 9 April at DREC in 2012. A permanent flood was established at the V5 growth stage and maintained until R8 which corresponded to approximately 2 wk prior to harvest (Counce et al., 2000).

    The trials were established in a randomized complete block arrangement. Treatments were a factorial arrangement of three previously mentioned seeding rates and 10 N fertilization strategies. Nitrogen fertilization strategies consisted of multiple rates of N applied at the V3, R1, and R3 growth stages (Table 1). The N strategies chosen represented both rice fertilization programs recommended by university extension guidelines (i.e., 100% of N applied at V3), as well as programs that producers could employ to reduce lodging, while still maintaining yield potential (i.e., split N applications).

    Table 1. Growth stages of N application and total N rate for rice fertilization strategies for studies in Mississippi and Louisiana during the 2010 and 2012 growing seasons.
    N strategy Total N
    V3 R1 R3
    kg N ha−1
    101 101
    101 50 151
    101 50 151
    101 50 50 201
    151 151
    151 50 201
    151 50 201
    151 50 50 251
    201 201
    251 251

    Preflood N was applied with a custom-manufactured, self-propelled fertilizer distributer equipped with a belt cone (Hege, Wintersteiger, Inc., Salt Lake City, UT). The R1 and R3 applications of N were hand-broadcast into each plot into the flood water. Seeds were treated with thiamethoxam {3-[(2-chloro-1,3-thiazol-5-yl)methyl]-5-methyl-N-nitro-1,3,5-oxadiazinan-4-imine} at 1.4 g kg−1seed to protect against rice water weevil (Lissorhoptrus oryzophilus), which can limit yield potential, especially in low density plots. Plots were kept weed-free by applying recommended labeled rates of glyphosate [N-(phosphonomethyl)glycine] and clomazone {2-[(2-chlorophenyl)methyl]-4,4-dimethyl-3-isoxazolidinone} 1 d after planting, imazethapyr {5-ethyl-2-[(RS)-4-isopropyl-4-methyl-5-oxo-2-imidazolin-2-yl]nicotinic acid} at V2 and V4, and quinclorac (3,7-dichloro-8-quinolinecarboxylic acid) at V4 (Buehring, 2008; Saichuck et al., 2008). Diseases and late-season insect pests did not exceed threshold populations, and thus were not chemically controlled.

    Immediately prior to harvest, all plots were visually rated for percent lodging and lodging severity. Lodging percentage was evaluated as a percentage of the plot lodged and lodging severity was assessed on a scale of 1 to 5, where 1 indicated erect and 5 indicated horizontal and matted to the ground. Plots were harvested when grain moisture for the latest maturing plots were below 220 g kg−1 with a small plot combine (Wintersteiger, Inc., Salt Lake City, UT) equipped with a grain gauge (Harvest Master, Juniper Systems, Inc., Logan UT). Yields were standardized to 120 g kg−1 moisture.

    Economic Analysis

    Harvest costs (HC) were calculated for each plot. Harvest costs were estimated using the Rice Combine Harvest Cost Estimation: Performance Rates and Harvest Costs decision support aid (Salassi and Deliberto, 2010). All harvest costs were based on an estimated 607 harvested ha yr−1 using a 355 horsepower combine with fuel consumption of 69.3 L of diesel h−1 and a field efficiency of 75% using an 8.25 m header. The initial purchase price of the combine with header was estimated at US$407,000, repair and maintenance costs were estimated at 25% of the purchase price over a useful life of 10 yr with annual use at 250 h yr−1. The salvage value of the combine and header was estimated at 30% of the original purchase price, and amortized at an annual rate of 4.5%. The diesel price was assumed to be $0.87 L−1 and labor cost was estimated that $12.50 h−1 (Falconer et al., 2013).

    Since lodging affects the speed at which combines can harvest the rice crop, harvest costs were adjusted using the percent lodging and lodging severity determined for each plot. Combine field speeds were estimated for each lodging severity by conferring with experienced commercial rice producers near the study sites. For example, combine speed for rice with a lodging severity of 1 was estimated at 4.83 km h−1, while combine field speed for rice with a lodging severity of 5 was estimated at 1.61 km h−1, respectively (Table 2).

    Table 2. Combine speed and estimated harvest costs rice at multiple lodging severities.
    Lodging severity Combine field speed Harvest costs
    km ha−1 $ ha−1
    1 4.83 $109.96
    2 4.02 $118.29
    3 3.22 $130.77
    4 2.41 $151.57
    5 1.61 $193.16
    Harvest costs for each lodging score were then adjusted for each experimental unit using the following equation:
    urn:x-wiley:00021962:agj2agronj2016030128:equation:agj2agronj2016030128-math-0001(1)
    where AHCi = the harvest cost per hectare for each plot adjusted for percent lodging and lodging severity, %Lodgedi = the percentage of the plot lodged, HC1 = the estimated cost per hectare to harvest rice with no lodging (lodging severity score of 1) and HCLodgingSeverityScorei = the estimated harvest cost per hectare based on the lodging severity score for the plot.
    The calculated adjusted harvest cost (AHC) values for each plot were used to construct partial budgets for each plot to estimate net returns above cost (NRAC) for seed, fertilizer, harvest, hauling, and drying costs. The NRAC for each experimental unit was calculated using the following equation:
    urn:x-wiley:00021962:agj2agronj2016030128:equation:agj2agronj2016030128-math-0002(2)

    where NRACi = the net return above seed, fertilizer, harvest, hauling and drying costs, Yi = rice plot yield (kg ha−1), P = rice price ($ kg−1), FRi = nitrogen fertilizer applied to the plot (kg ha−1), FP = nitrogen cost ($ kg−1), FAP = fertilizer application cost ($ kg−1), H = hauling cost ($ kg−1), DC = drying costs ($ kg−1), and AHCi = the harvest cost ($ ha−1) for the plot adjusted for percent lodging and lodging severity.

    To test the sensitivity of the model to changes in rice and N prices, NRAC was calculated for two scenarios. The first scenario was a low rice sale price and a low N price, while the second scenario was based on a high rice sale price and high N price. The low and high average annual prices reported for rice in Mississippi between 2010 and 2012 were used to determine the low and high rice sale prices (USDA NASS, 2012). Similarly, the low and high N prices were based on reported prices for urea-N between 2010 and 2012 (Falconer et al., 2013). Low and high reported rice prices were $0.23 and $0.326 kg−1 respectively, with low and high N prices of $0.78 and $1.36 kg−1, respectively. Fertilizer application costs were based on a rate of $0.31 kg−1 of N applied. Seed rice price was held constant at $2.20 kg−1. Hauling and drying charges were assessed at $0.015 and $0.02 kg−1, respectively.

    Data were subjected to ANOVA using the PROC MIXED procedure in SAS (SAS Institute, Cary, NC). Seeding rate, N strategy, and the seeding rate × N strategy interaction were considered fixed effects. Year, location(year), replication(location × year) were considered random effects so that broader inferences could be drawn regarding the effect N strategies and seeding rates for rice production across the midsouthern United States (Carmer et al., 1989). Analysis of variance was conducted for rice grain yield, percent lodging, lodging severity, and NRAC at the two price scenarios. For all analyses, the level of significance was set at 5%. Least square means were calculated and mean separation was conducted using PDMIX800 macro in SAS (Saxton, 1998).

    RESULTS AND DISCUSSION

    Percent lodging was affected by an N strategy × seeding rate interaction (Table 3). Percent lodging was greatest when rice was seeded at 323 and 483 seeds m−2 and fertilized with 251 kg N ha−1 applied V3. Within each seeding rate the 251 kg N ha−1 V3 N strategy resulted in the greatest percent lodging (Table 4). The 201 kg N ha−1 applied V3 N strategy resulted in limited (2%) lodging when rice was seeded at 161 seeds m−2, and relatively high (33%) percent lodging when rice was seeded at 483 seeds m−2. Lodging was not observed for N strategies where the total N was 101 or 151 kg N ha−1 at any seeding rate. When total N rate was 201 or 251 kg ha−1, splitting the N application reduced lodging. For example, when rice was seeded at 323 seeds m−2 percent lodging was 45% when 251 kg N ha−1 was applied V3, but only 3% when 151 kg N ha−1 was applied V3 and 50 kg N ha−1 was applied at both R1 and R3. Similarly, at the 483 seeds m−2 seeding rate 201 kg N ha−1 applied V3 resulted in 33% lodging, but when the N was split between 151 kg N ha−1 V3 and 50 kg N ha−1 at R1, lodging was reduced to 7%.

    Table 3. Model significance for the ANOVA for the main effects of seeding rate and N strategy and the seeding rate × N strategy interaction on percent lodging, lodging severity, grain yield, net returns above costs (NRAC)-Low, and NRAC-High for studies in Louisiana and Mississippi during the 2010 and 2012 growing seasons.
    Effect Percent lodging Lodging severity Grain yield NRAC-Low NRAC-High
    P > F
    Seeding rate 0.0003 0.00159 <0.0001 <0.001 <0.001
    N strategy <0.0001 <0.0001 <0.0001 <0.001 <0.001
    N strategy × Seeding rate 0.0001 0.0374 ns ns ns
    • Lodging severity 1 to 5 scale: 1 = no lodging, 5 = plants laying flat.
    • ns, not significant at the P ≤ 0.05 level.
    Table 4. Percent lodging and lodging severity for the rice at three seeding rates and 10 N strategies for studies in Louisiana and Mississippi during the 2010 and 2012 growing seasons.
    N strategy Seeding rate Seeding rate
    V3 R1 R3 161 323 483 161 323 483
    kg N ha−1 % lodging lodging severity (1–5)
    101 0e 0e 0e 1.0c 1.0c 1.0c
    101 50 0e 0e 0e 1.0c 1.0c 1.0c
    101 50 0e 0e 0e 1.0c 1.0c 1.0c
    101 50 50 0e 0e 0e 1.0c 1.0c 1.0c
    151 0e 0e 0e 1.0c 1.0c 1.0c
    151 50 0e 0e 0e 1.0c 1.0c 1.0c
    151 50 0e 2e 5e 1.0c 1.1c 1.1c
    151 50 50 1e 3e 7de 1.1c 1.1c 1.1c
    201 2e 15d 33c 1.1c 1.4c 2.0ab
    251 25bc 45a 51a 1.7b 2.4a 2.4a
    • Values followed by the same letter are not statistically different at P ≤ 0.05.
    • Lodging severity 1 to 5 scale, (1 = no lodging, 5 = plants laying flat).

    Lodging severity was also affected by the seeding rate × N strategy interaction (Table 3). Similar to the percent lodging data, lodging severity was greatest when 251 kg N ha−1 V3 was applied to rice seeded at 323 and 483 seeds m−2 (Table 4). At seeding rates of 161 and 323 seeds m−2 lodging severity was greater at the 251 kg N ha−1 V3 strategy than at all other N strategies. However, when seeded at 483 seeds m−2 rice lodging severity was similar at both 201 kg N ha−1 V3 and 251 kg N ha−1 at V3 N strategies. As with the percent lodging data, lodging severity was decreased when 201 and 251 kg N ha−1 rates were split between V3 and R1 and/or R3 N applications, suggesting that splitting N between multiple growth stages is a strategy that can minimize lodging in susceptible rice cultivars.

    While splitting N applications between V3 and R1 and/or R3 growth stages was successful at decreasing percent lodging and lodging severity, grain yield was reduced compared to N strategies where all N was applied at V3 (Table 5). When pooled over seeding rates, the 201 and 251 kg N ha−1 applied V3 resulted in the greatest grain yields (Table 6), indicating that early N application is critical to maximize yield potential of delayed-flood rice production systems characteristic of midsouthern United States. The 201 kg N ha−1 N rate applied V3 produced greater grain yield than other N strategies that had the same total N rate, but was split between multiple growth stages. For example, the 201 kg N ha−1 rate yielded 4% more than the 151 kg N ha−1 followed by 50 kg N ha−1 at R1 and the 151 kg N ha−1 followed by 50 kg N ha−1 at R3 (11,479 vs. 11,064 and 11,011 kg ha−1, respectively). The 201 kg N ha−1 applied at V3 N strategy also yielded 9% more than the 101 kg N ha−1 applied at V3 followed by 50 kg N ha−1 at both R1 and R3 (11479 vs. 10426 kg ha−1). Interestingly, 151 kg N ha−1 applied V3 yielded more than 101 kg N ha−1 applied V3 followed by 50 kg N ha−1 at both R1 and R3 (10,864 vs. 10,426 kg ha−1), further reinforcing the importance of adequate early-season N to rice grain yield determination. These results are similar to those of Bollich et al. (1994) that also found that rice grain yield was greatest with a single V3 application of N and was decreased with two-way and three-way split applications. Similarly, Mengel and Wilson (1988) reported that N applied at V3 resulted in increased vegetative growth, N uptake, and grain yield compared to top-dressed N applications. More recent studies have also concluded that adequate V3 N rate is necessary for maximum rice yield (Slaton et al., 2003, 2004; Walker et al., 2006).

    Table 5. Grain yield and net returns above costs (NRAC) for low rice price-low N price and high rice price-high N price scenarios at three seeding rates and 10 N strategies for studies in Louisiana and Mississippi during the 2010 and 2012 growing seasons.
    N strategy Grain yield Low rice price-Low N price High rice price-High N price
    Seeding rate Seeding rate
    V3 R1 R3 Total N 161 323 483 Avg. 161 323 483 Avg.
    kg N ha−1 $ ha−1
    101 101 9,612a 1524 1526 1462 1504e 2305 2347 2293 2315e
    101 50 151 10,041a 1513 1536 1452 1500e 2286 2358 2273 2305e
    101 50 - 151 10,375b 1598 1631 1568 1599cd 2409 2496 2444 2449cd
    101 50 50 201 10,426b 1557 1592 1514 1554de 2346 2435 2360 2380 de
    151 151 10,864c 1723 1712 1651 1695b 2592 2615 2565 2590b
    151 50 201 11,011cd 1696 1707 1606 1669b 2548 2603 2494 2548b
    151 50 - 201 11,064cd 1642 1716 1682 1679b 2470 2616 2605 2563b
    151 50 50 251 11,166 de 1649 1692 1593 1644bc 2476 2578 2472 2508bc
    201 201 11,479 f 1793 1778 1706 1758a 2690 2707 2643 2679a
    251 251 11,431ef 1750 1751 1560 1687b 2625 2670 2429 2574b
    Avg. 1664a 1664a 1579b 2474b 2542a 2457b
    • Values followed by the same letter are not statistically different at P ≤ 0.05.
    Table 6. Rice grain yield for the main effect of N strategy for studies in Louisiana and Mississippi during the 2010 and 2012 growing seasons.
    Seeding rate Grain yield
    seeds m−2 kg ha−1
    161 10,427b
    323 10,907a
    483 10,908a
    • Values followed by the same letter are not statistically different at P ≤ 0.05.

    Rice grain yield was also affected by the main effects of seeding rate (Table 6). When pooled over N strategies rice seeded at 323 and 483 seeds m−2 yielded 5% greater than rice that was seeded at 161 seeds m−2 (10,907 and 10,908 vs. 10,427 kg ha−1, respectively) (Table 5). Bond et al. (2005) reported that rice response to seeding rate varied by cultivar, but optimum rice grain yield could be achieved in Louisiana and Mississippi with a seeding rate of 323 seeds m−2. Similarly, Harrell and Blanche (2010) suggested that grain yield was optimized for one rice cultivar at 323 seeds m−2, while yield was optimized for another cultivar at seeding rates above 323 seeds m−2 when produced in a similar drill-seeded, delayed flood production system.

    Economic Analysis

    Across all plots, 8.5% had some lodging and incurred a harvest cost higher than the baseline. Both the main effects of seeding rate and N strategy affected NRAC for the low rice price-low N price scenario, with no interaction (Table 3). The 483 seeds m−2 seeding rate resulted in lower NRAC values compared to both the 161 seeds m−2 seeding rate and the 323 seeds m−2 seeding rate ($1579 vs. $1664 and 1664 ha−1, respectively) (Table 5). These results also contrast markedly with the grain yield results in which the 323 and 483 seeds m−2 seeding rates increased grain yield compared to the 161 seeds m−2 seeding rate. Increased levels of lodging observed for rice planted at 483 seeds m−2 resulted in increased harvest costs, eliminating the economic benefit associated with higher grain yield. When analyzed solely for grain yield the 201 kg N ha−1 V3 strategy had similar yields as 251 kg N ha−1. However, rice fertilized with the 201 kg N ha−1 V3 had higher a greater NRAC than all other N fertilization strategies, including the 251 kg N ha−1 ($1785 vs. $1687 ha−1), due to greater AHC and N costs associated with the 251 kg N ha−1 V3 N strategy. The lack of NRAC differences between the 161 and 323 seed m−2 seeding rates indicates that when rice prices are low, rice producers might consider reducing seeding rates. However, producers should be cognizant of potential problems associated with low seeding rates such as potentially increased weed and insect pest pressure.

    Similar, to the low rice price-low N price scenario, both the main effects of seeding rate and N strategy affected NRAC at the high rice price- high N price scenario with no interaction (Table 3). However, unlike the low rice price-low N price scenario the 323 seeds m−2 seeding rate resulted in greater NRAC than both the 161 seeds m−2 and 483 seeds m−2 seeding rates ($2542 vs. $2474 and $2458 ha−1, respectively) under the high rice price-high N price scenario (Table 5), suggesting that when rice prices are elevated, growers should drill rice at 323 seeds m−2 to ensure maximum economic returns. These results support those of Bond et al. (2005) that optimal rice yields can be achieved with a seeding rate of 323 seeds m−2 in Louisiana and Mississippi. Although the 251 and 201 kg N ha−1 N applied at V3 N strategies resulted in similar grain yield, the 201 kg N ha−1 at V3 strategy had greater NRAC than the 251 kg N ha−1, due to greater harvest costs due to lodging associated with the 251 kg N ha−1 at V3 N strategy. While the splitting N application across multiple growth stages decreased percent lodging and lodging severity compared with the 201 kg N ha−1 at V3 N strategy, the increased harvest efficiency was not enough to overcome the grain yield penalty associated with suboptimal (<201 kg N ha−1 at V3) early-season N rate.

    CONCLUSION

    Yield and economic analysis indicated that the optimal N strategy for this highly lodging susceptible rice cultivar was 201 kg N ha−1 applied at V3 across a range of rice and N prices. Splitting N application between multiple growth stages did help to minimize lodging, but resulted in decreased grain yield and net returns comparted to optimal N rate (201 kg N ha−1) applied early in the growing season (V3). These results suggest that early season N rate is critical to maximize yield potential. However, it appears that an application of N at the R1 or R3 growth stage can help recover some yield potential lost due to suboptimal early-season N.

    This research supported the findings of other studies across the midsouthern United States that yield of modern rice cultivars is maximized at 323 seeds m−2, for drill-seeded, delayed flood rice, when planting within the optimum planting window. However, seeding rates may need to be adjusted for factors such as early or late planting and differences in soil conditions at planting. Economic analysis indicated that in situations where rice and N prices are low, growers can consider reducing seeding rates to maximize net returns, although other agronomic consequences of decreasing seeding rates should be considered.

    This study is one of the first studies to attempt to include the effects of both grain yield and lodging on net producer returns in rice. However, it should be noted that this research was purposely conducted on a single, high-yielding cultivar, CL151, which is highly susceptible to lodging. Yield and net returns will certainly vary for other cultivars based on both their yield potential and lodging susceptibility.