| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Univ. of Florida, Range Cattle Res. and Educ. Cent., 3401 Experiment Rd., Ona, FL 33865-9706
b Univ. of Florida, Southwest Res. and Educ. Cent., 2686 SR 29 N, Immokalee, FL 34142-9515
c Univ. of Florida, Dep. of Stat., P.O. Box 110339, Gainesville, FL 32611-0339
* Corresponding author (ivezenwa{at}ifas.ufl.edu)
Received for publication January 8, 2004.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AS, ammonium sulfate • DAF, days after fertilization • DM, dry matter • Nx, nitrogen application at x kg N ha–1 • OM, organic matter • Sx, sulfur application at x kg S ha–1
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Bahiagrass, the most common pasture grass in Florida with 
1 million ha, often receives only N fertilizer at 56 to 67 kg N ha–1 annually (Chambliss, 1999). With this level of management, S has not been considered to be a limitation to pasture production. Early research in Florida comparing ammonium sulfate (AS) with four other N fertilizers over 10 yr showed no difference between AS and other sources of N for bahiagrass annual yield and N uptake (Blue, 1974). However, it has been more recently demonstrated that bahiagrass yield and concentrations of S in forage increased with AS fertilization (Rechcigl et al., 1989; Rechcigl, 1991; Arthington et al., 2002).
Mitchell and Blue (1989) designed a fertilization trial to find S rates that amended deficiencies and maintained suitable bahiagrass growth for Florida pasture. In that study, N (200 or 400 kg ha–1 yr–1) and four levels of S fertilizer (0–40 kg ha–1 yr–1) were variables. They found that bahiagrass yield and N recovery were improved with S fertilization and that concentration of S in forage increased quadratically (1.3–2.2 g S kg–1) with S fertilizer.
In past research with bahiagrass, a narrow range of S rates has been applied over a relatively limited range of N rates. With increased use of AS fertilizer in pasture due to its cost advantage over ammonium nitrate, relatively large amounts of N and S can be applied together. The objective of the present study was to measure bahiagrass responses to an annually applied, factorial combination (16) of four rates of N (0–255 kg N ha–1) and S (0–285 kg S ha–1).
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
At the start in each year, plots were cut uniformly to a 2.5-cm stubble, and grass was removed on the day of fertilization. Plots were harvested three times annually in 2001 and 2002 on 35-d intervals (Table 1). In 2003, plots were harvested once at 35 DAF. The third year was added to verify the N and S responses from previous years and to obtain stolon–root samples. At each harvest, a 6-m-long by 0.46-m-wide strip of grass was cut and collected from the center of each plot with a rotary mower. Grass was weighed in the field, and 0.5-kg subsample was taken, dried (60°C for 5 d), and reweighed for DM determination. Subsamples in 2001 and 2002 were ground and analyzed for N, S, P, K, Ca, Mg, Cu, Fe, Mn, and Zn (Waters Agric. Lab., Camilla, GA; www.watersag.com; verified 12 Jan. 2005). Apparent recovery of applied N and S was calculated as annual uptake of N or S minus annual uptake of N or S by the N0S0 treatment divided by N or S applied in a respective treatment.
|
Total and sulfate S were determined in soil collected at 0- to 15-cm depth at 35 and 105 DAF in 2002 and in soil from the spodic horizon at 105 DAF (Waters Agric. Lab., Camilla, GA). Total N and organic C were determined in soil collected at 0- to 15-cm depth at 35 DAF in 2002. Six treatments were selected that would allow a partial examination of the N x S interaction and the AS response curve. These were N0S0, N85S95, N170S190, and N255S285, which correspond to 0, 400, 800, and 1200 kg AS ha–1, respectively, plus N0S285 and N255S0, which allowed additional contrast comparisons. Stubble–root–stolon samples were collected from these six treatments after forage harvest at 35 DAF in 2003. Two cores (10 cm diam. by 10 cm deep) were taken with a golf-green cup cutter from each of 24 plots. After soil was washed from the roots, two samples from each plot were composited, dried, weighed, ground, and analyzed for N and S (Waters Agric. Lab., Camilla, GA). These samples consisted of stubble (2 cm), stolons, and roots.
Responses were analyzed with a mixed linear model using the MIXED procedure of SAS (1999). The model contained the factorial set of N and S treatments as whole plots with year as the subplot and DAF as the sub-subplot. The error terms, block x N x S and block x N x S x year, which were used to test whole and subplots, respectively, were considered random effects. Yield data for the single 2003 harvest were analyzed using a MIXED model that included 35 DAF for 2001 and 2002. For interactions involving N or S, these variables were examined with contrasts to determine if responses were linear or quadratic over their levels, and equations were derived for the highest significant (P < 0.05) order. Values for R2 were calculated after removing those sources of variation not associated with the regressor variable. For interactions involving N or S x DAF, means over DAF were examined with the PDIFF (LSD) option (SAS Inst., 1999).
Soil C, N, S, stolon–root mass, and uptake of N and S by the stolon–root mass were analyzed as a randomized complete block using a general linear procedure (SAS Inst., 1999). Linear and quadratic contrasts were made to determine form over levels of AS. Contrasts were made to compare N0S0 vs. N0S285 and N255S0 vs. N255S285.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Forage Yield
There was an N x S fertilizer interaction for yield (P = 0.00001) (Fig. 1a)
. Nitrogen increased bahiagrass yield at all levels of S fertilizer. With S0 and S95, yield followed a quadratic form over levels of N, but the response became linear with S190 and S285. The yield response to S diminished as rate of N increased. For example, yield increased 2280 kg DM ha–1 between N0S0 and N255S0 but 430 kg DM ha–1 between N0S285 and N255S285.
|
Elements in Forage
Nitrogen
Concentrations of N in forage were affected by an N x S fertilizer interaction (P = 0.006) (Table 2). These concentrations increased linearly over N rates at all levels of S fertilizer. Sulfur fertilizer increased N concentration in forage only with N0. The addition of 285 kg S ha–1 with no N resulted in the same concentration of N in forage as 85 kg N ha–1.
|
|
Nitrogen and Sulfur Uptake and Recovery by Forage
Nitrogen Uptake
Annual N uptake was affected by an N x S fertilizer interaction (P < 0.0005) in which N uptake increased linearly over levels of N fertilizer at each level of S fertilizer (Table 2). Over levels of S, uptake of N diminished as rate of N increased. For example, bahiagrass uptake increased 80 kg N ha–1 between N0S0 and N0S285, but between N170S0 and N170S285, uptake increased 42 kg N ha–1. There was no response over levels of S fertilizer with N255.
Sulfur Uptake
Annual S uptake was affected by an N x S fertilizer interaction (P = 0.03) in which S uptake increased linearly over N levels at the first three levels of S fertilizer (Table 3). The increases in S uptake over N levels diminished until there was no difference in S uptake over levels of N at S285. Over levels of S, uptake of S increased at all levels of N. There was a quadratic increase in uptake of S with N0, but S uptake increased linearly at all other N levels. The rate of these increases diminished as N level increased.
Apparent Recovery
Nitrogen recovery was affected by an N x S fertilizer interaction (P < 0.0001) (Table 4). There was always a linear decrease in N recovery over levels of N within each level of S fertilizer. Over levels of S fertilizer, N recovery was quadratic with N85 while recovery was linear with N170. With N255, there were no differences in N recovery over levels of S fertilizer.
|
Stubble–Stolon–Root Mass, Nitrogen, and Sulfur Concentrations
Root–stubble data are not shown in tables but are described in text. Stubble–stolon–root mass was affected by fertilizer treatment with mass changing quadratically (P = 0.03) over fertilizer rates (N0S0 = 17, N85S95 = 25.3, N170S190 = 21.6, and N255S285 = 21.5 Mg DM ha–1). Contrasts comparing N0S0 vs. N0S285 and N255S0 vs. N255S285 were not significant, implying that N, not S, was responsible for the increase in mass.
Concentrations of N (8.1–13.7 g N kg–1 with N0S0–N255S285) and S (2.0–2.8 g S kg–1 with N0S0–N255S285) increased linearly over fertilizer rates. Contrasts comparing N concentration with N0S0 vs. N0S285 and N255S0 vs. N255S285 were not significant, indicating that S fertilization had no effect on N concentration. For S concentrations, N0S0 < N0S285 (2.8 g S kg–1) and N255S0 (1.6 g S kg–1) < N255S285, indicating S fertilization did affect S concentration.
Uptake of N by the stubble–stolon–root mass increased linearly (P = 0.003) from 137 to 287 kg N ha–1 with N0S0 to N255S285, respectively. Uptake of S also increased linearly from 35 to 60 kg S ha–1 with N0S0 to N255S285, respectively. Contrasts comparing N uptake with N0S0 vs. N0S285 and N255S0 vs. N255S285 were not significant but were significant for S uptake, N0S0 < N0S285 (54 kg S ha–1) and N255S0 (35 kg S ha–1) < N255S285. This indicates that S had no effect on N uptake, but S did affect S uptake.
Elements in Soil
Soil Carbon, Nitrogen, and Sulfur
Fertilizer treatments had no effect on C (mean 37.7 g kg–1) or total N (mean 2.2 g kg–1) in the 0- to 15-cm soil at 35 DAF in 2002 (data not in table). Total and sulfate S in the 0 to 15 cm of soil at 35 and 105 DAF were also not affected by treatments or sample date (Table 5). Total S in the spodic horizon increased linearly (P = 0.05) over AS rates (41–56 mg S kg–1 soil with N0S0– N255S285, respectively). There was no difference in total S in the spodic horizon between N0S0 and N0S285 (46 mg S kg–1 soil), but N255S0 (35 mg S kg–1 soil) < N255S285 (P = 0.04).
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
It is doubtful that responses to S fertilizer were due to the direct effects of S on bahiagrass nutrition. Concentrations of S in forage were 2.3 g S kg–1 with no S fertilization (Table 3), which is well above the 1.6 g S kg–1 suggested as the critical level (Mitchell and Blue, 1989). For comparison, it is suggested that top growth of ‘Coastal’ bermudagrass (Cynodon dactylon L.) contains 1.4 to 1.9 g S kg–1 (Jordan and Bardsley, 1958). Once a critical S concentration in tissue is reached, relative yield becomes asymptotic when plotted against increasing concentration of S (Martin and Walker, 1966; Mitchell and Blue, 1989).
We suggest that the increases in bahiagrass yield, concentration of N in forage, and N uptake were essentially responses to N mineralized from soil OM. Our objective was not to study soil N mineralization, but these forage responses to N not applied as fertilizer require consideration of mineralization through literature and our limited soil N data.
Termed the "sulfate bonus," Freney and Spencer (1960) indicate that OM can be mobilized with sulfate addition, which mineralizes N. Nitrogen is released in the ratio in which it occurs in OM (Freney and Stevenson, 1966). There is a close relationship between C, N, and S in soil OM, and while there is considerable variation among individual soils, the mean C/N/S ratios of soils worldwide are very similar (Freney and Stevenson, 1966). Walker (1957) generalized that the C/N/S ratio in the A horizon of most soils was 100:8:1. The mean C/N/S ratio of nine Florida Spodosols was 150:6.4:1 (Mitchell and Blue, 1981a). The C/N/S ratio in our soil was 131:7.6:1.
The total N concentration in our surface soil averaged 2.2 g kg–1. For comparison, Blue (1979) reported 1.1 g N kg –1 in a Spodosol. The release of <2% of our soil N by mineralization as a result S fertilizer addition could account for the 80 kg N ha–1 yr–1 difference between N uptake in the N0S0 treatment (53 kg N ha–1 yr–1) and the N0S285 treatment (133 kg N ha–1 yr–1) (Table 2).
Soil Sulfur
While finer-textured soils can store relatively large amounts (>100 mg kg–1) of sulfate (Jordan and Bardsley, 1958; Harward and Reisenauer, 1966), Florida's sandy Spodosols contain little sulfate. Our 2.5 mg kg–1 of sulfate in the upper 15 cm of soil (Table 5) was very similar to other low sulfate values for Florida Spodosols, such as 2 to 3 mg kg–1 (Neller, 1959; Mitchell and Blue, 1989).
Most of the S contained in soil is organic, and sulfate comes almost entirely from microbial conversions of these organic forms (Freney and Stevenson, 1966). Our soil contained 301 mg total S kg–1 soil, which is the equivalent of 675 kg S ha–1. While our total S concentration was greater than other reports for Florida Spodosols, such as 88 mg total S kg–1 soil (Mitchell and Blue, 1989) or 160 mg total S kg–1 (Monteiro and Blue,1990), all values indicate a substantial S sink in the organic fraction.
Greater S mineralization in soils occurs with lower C/S ratios (Barrow, 1960a; Bettany et al., 1974). Barrow (1960b) studied mineralization of S from various ground plant materials incubated with soil and showed that mineralized inorganic S accumulated in soil with C/S ratios < 200, but with C/S ratios > 400, S was immobilized in soil OM. With a soil C/S ratio of 131 in our soil, sulfate could be mineralized from OM and made available for bahiagrass growth. Considering that the minimum concentration of S in our bahiagrass forage was relatively high at 2.3 g kg–1 with no S fertilizer, it would seem that mineralization of S was taking place in our pasture.
The amount of S needed by grasses increases with level of N fertilization (McNaught and Christoffels, 1961). Mitchell and Blue (1989) showed that bahiagrass needed 30 kg S ha–1 yr–1 with 200 kg N ha–1 yr–1 while 45 kg S ha–1 yr–1 was needed with 400 kg N ha–1 yr–1. The University of Florida recommends 56 to 67 kg N ha–1 for grazing on bahiagrass pastures in central and south Florida (Chambliss, 1999). For older pastures with relatively high soil OM and total S concentrations, S fertilization may not be necessary if concentrations in tissue do not fall below 1.6 g kg–1.
On the other extreme, annual application of 56 to 67 kg N ha–1 as AS to bahiagrass pasture would also provide 64 to 77 kg S ha–1. High concentrations of S in bahiagrass forage (4.8–5.1 g S kg–1) with use of 67 kg N ha–1 from AS fertilizer resulted in cattle with lower liver Cu concentrations in Florida than contemporaries grazing unfertilized bahiagrass or bahiagrass fertilized with ammonium nitrate (Arthington et al., 2002). When dietary S is excessive (>4.0 g S kg–1 of diet DM; NRC, 1996), it combines with Mo to form a thiomolybdate complex that binds Cu and renders it unavailable to ruminants (Mason, 1990; Suttle, 1991). Based on our research, it is difficult to see how such high S concentrations could have resulted. We applied 3.7 times more S in the present study than in the Arthington et al. (2002) study, yet maximum S concentrations in forage at 35 DAF did not exceed 3.8 g S kg–1. Obviously, S fertilization was not alone in determining S concentrations in bahiagrass as soil, environment, and unknown factors must play a role.
Apparent Recovery of Nitrogen and Sulfur
Over the range of N fertilizer rates with S0 (Table 4), recovery of applied N was in agreement with N recovery reported by others. Blue (1970)(1974) found recovery of N by bahiagrass forage on a Spodosol ranged from 400 to 500 g N kg–1 of applied N during the first four experimental years, but it increased to about 700 g N kg–1 of applied N during the subsequent 6 yr. Blue and Graetz (1977) reported N recovery in bahiagrass forage at 640 to 790 g N kg–1 of applied N. Sveda et al. (1992) reported recovery of applied N in bahiagrass forage averaged 520 g N kg–1 of applied N. However, our N recovery rates with N85 exceeded N applied at S285. Forage N uptake increased 80 kg ha–1 between N0S0 and N0S285 (Table 2), which may have been mineralized from OM. Assuming there was mineralization taking place at N85, uptake could exceed recovery.
The amount of S recovered by bahiagrass forage accounted for very little of the S from fertilizer and did not exceed 136 g S kg–1 of S fertilizer applied (Table 4). At most, annual uptake of S by forage did not exceed 30 kg ha–1 with S285 (Table 3). If uptake by the stubble–root–stolon mass is considered, an additional maximum of 60 kg S ha–1 was taken up by bahiagrass. Since a relatively small amount of applied S was taken up by bahiagrass, and S fertilization neither increased total or sulfate S in soil, it is assumed that much S was lost. In the Mitchell and Blue (1989) study, 40 kg S ha–1 was applied in each of 6 yr with no effect on sulfate concentrations in the soil.
Accounting for the fate of S not taken up by bahiagrass is complicated by the texture and drainage of our soils. Sulfate is weakly held in arable soil and leaches quickly (Chang and Thomas, 1963). In lysimeter studies on a sandy loam, 900 g kg–1 of S35–labeled gypsum leached from the soil in a single growing season with 483 mm of rain (McKell and Williams, 1960). Jones et al. (1968) found concentrations of S were greatest in the first percolate of the rainy season, and 800 to 900 g kg–1 of S applied as gypsum to a sandy soil was accounted for within 1 yr. Although our soil was aerobic, relatively dry, with rapid drainage when S was applied, we received a 3-yr average of 406 mm in the first 35 d after S fertilizer application (Table 1). At 35 DAF, concentrations of S in the soil were not affected by S fertilizer level because sulfate had leached from the upper 15 cm of soil.
Although we found more sulfate in the spodic horizon (14 mg kg–1) than in the upper 15 cm of soil (Table 5), that does not imply S leaching, but it is possible. Surface-applied P and K fertilizer reportedly moved to the spodic horizon within 28 DAF (Dantzman and McCaleb, 1969). Retention of sulfate is related to the amount of clay in a soil, and since the spodic horizon has a relatively high clay content, high concentrations of sulfate are naturally found there (Neller, 1959). Soils with kaolin clay and oxides of Fe and Al, which are characteristic of the spodic horizon, attract and hold sulfate (Berg and Thomas, 1959). However, in spite of the greater concentrations, Mitchell and Blue (1981b) found S in the spodic horizon contributed little to plant growth.
Soil drainage is a second issue that complicates S accountability. Within 35 DAF, the water table rose to the surface with excessive rain (Table 1), and conditions became anaerobic. Sulfate is not stable in anaerobic environments and is rapidly converted by bacteria to sulfide (Starkey, 1966). A small amount of soil sulfide can be lost by volatilization as H2S (Barrow, 1961), but the majority was probably lost when water receded.
Lateral movement of N and S is possible. Lateral movement of applied K was detected at 0.9 m, but not at 2.7 m, from the edge of a fertilized area after 1 yr (Dantzman and McCaleb, 1969). Our sampling points were 2.3 m from the center of a plot to the edge of an adjacent plot. There were abrupt visual differences between fertilized plots (tall, green) and unfertilized border plots (short, yellow). It is difficult to believe there was much mixing of fertilizer elements due to lateral movement.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. C. Newman, S. Agyin-Birikorang, M. B. Adjei, J. M. Scholberg, M. L. Silveira, J. M. B. Vendramini, J. E. Rechcigl, L. E. Sollenberger, and M. Chrysostome Enhancing Phosphorus Phytoremedation Potential of Two Warm-Season Perennial Grasses with Nitrogen Fertilization Agron. J., November 1, 2009; 101(6): 1345 - 1351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. C. Newman, S. Agyin-Birikorang, M. B. Adjei, J. M. Scholberg, M. L. Silveira, J. M. B. Vendramini, J. E. Rechcigl, and L. E. Sollenberger Nitrogen Fertilization Effect on Phosphorus Remediation Potential of Three Perennial Warm-Season Forages Agron. J., August 31, 2009; 101(5): 1243 - 1248. [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 | |||
