Phosphorus Removal by Sorghum-sudangrass in Northeastern USA

Phosphorus Removal by Sorghum-Sudangrass in Northeastern USA

Quirine M. Ketterings, Assistant Professor, Department of Crop and Soil Sciences, Cornell University, 817 Bradfield Hall, Ithaca NY 14853; Tom F. Kilcer, Field Crops Extension Educator, Cornell Cooperative Extension of Rensselaer County, 61 State St., Troy, NY 12180; Paul Cerosaletti, Dairy and Field Crops Extension Educator, Cornell Cooperative Extension of Delaware County, 34570 State Hwy 10, P.O. Box 184, Hamden, NY 13782; and Jerry H. Cherney, Professor, Department of Crop and Soil Sciences, Cornell University, 513 Bradfield Hall, Ithaca NY 14853

Abstract

Northeastern USA dairy producers have shown a growing interest in brown midrib sorghum-sudangrass (Sorghum bicolor (L.) Moench. × Sorghum sudanense Piper) hybrids (BMR S×S) as an environmentally sound alternative to corn (Zea mays L.) for silage. Grown in a 2-cut system with planting taking place after June 1, BMR S×S allows for the application of manure during times when manure nutrients are less likely to leach and/or runoff. For the long-term environmental sustainability of the dairy industry, manure nutrient application rates should not exceed crop removal for more years than needed to bring low fertility soils to optimum fertility. Thus, it is important to know P removal rates by this crop. Our objectives were to determine P removal for BMR S×S as impacted by yield, N and K application, and time of harvest. Seven trials were conducted in two growing seasons and three distinct agricultural areas in New York. Averaged over all trials and years, P removal through harvest (lbs of P₂O₅ per acre) was estimated as:

4.8 + [11.6 × Yield (dry matter in tons/acre)]

(r² = 0.85, P < 0.001). The mean dry matter P concentration was 0.29%. Thus, a crop of 5.6 tons of dry matter would on average remove about 70 P₂O₅ lbs/acre. However, P concentrations across all sites ranged from 0.15 to 0.53%. The variability was not explained by K rate, stand-height at harvest, or soil test P level and there were no significant differences between first cut and second cut. In the N rate studies, P concentrations were significantly greater for control plots where no N had been applied while no significant differences were detected among the different N rates. We conclude that for an accurate estimate of P removal, producers need to analyze their forage for P concentration and determine yields.

Introduction

An alternative forage crop that is economically superior to corn (Zea mais L.) silage is difficult to find if a soil resource and climate are well suited to growing corn. However, not all sites and local climates are well suited to corn. The average corn silage yield (65% moisture) in New York in the past 10 years varied from 16 tons/acre in 1998, 1999, and 2001 to 13 tons/acre in the drought year of 2002 with the average for the southern U.S. less than 10 tons/acre (8). Sorghum-sudangrass hybrids (Sorghum bicolor (L.) Moench. × Sorghum sudanense Piper) (S×S) are better adapted to drought than corn and the new brown midrib (BMR) hybrids (Fig. 1) which have high digestibility may have the potential to economically compete with corn silage in a large area in the U.S. Northeast. Sorghum-sudangrass is more tolerant of later planting than corn, allowing more flexibility in managing wet soil conditions and cropping work load. This crop also has soil conservation advantages over corn silage because of the ground cover it provides during the growing season.

Fig. 1. Brown midrib sorghum-sudangrass grown in a 2-cut system in New York.

The December 2002 release of the Final Concentrated Animal Feeding Operation (CAFO) ruling from the U.S. Environmental Protection Agency (12) made animal agriculture fully accountable to the Clean Water Act. In New York, all CAFOs are required to develop and implement a Comprehensive Nutrient Management Plan (CNMP) which meets USDA-NRCS standards and specifications by January 2009. Nutrient management planning is guided by NRCS Standard NY590 which requires that all fields be assessed for their P runoff potential using the New York Phosphorus Index (NY PI) as described in Czymmek et al. (1). If the NY PI is classified as high (between 76 and 100), P applications through manure and/or fertilizer should not exceed P removal by the crop. Similar approaches and management recommendations are used in other northeastern states (11). For many forage crops, average nutrient concentration data are available through commercial forage testing laboratories. One example is the forage database of Dairy One Forage Laboratory (2). However, because BMR S×S is a relatively new crop for the region, accurate P removal rates have yet to be determined. Furthermore, studies are needed to determine the impact of crop management practices such as time of harvest and nitrogen and potassium management on P concentrations in the forage.

Our objectives were to determine first and second harvest forage P concentrations and P removal by BMR S×S as impacted by N and K application rates, stand height at harvest, dry matter yield and soil test P. The BMR S×S was grown in three different New York soil and climatic regions and over two very different years including the 2002 drought year and a very good 2003 growing season.

Determining P Removal Rates for BMR S×S

Seven field trials were conducted in New York at three different locations, Delaware County (southeastern region), Columbia County (eastern region) and Tompkins County (southern region), in 2002 and 2003. The trials were part of a larger, multiple-year effort to determine stand height optimum for yield and quality and N and K requirements of the crop (Fig. 2). The Delaware trials were stand height studies conducted on Chenango gravelly silt loam (loamy-skeletal, mixed, mesic Typic Dystrochrepts). The Tompkins County trials were N and K rate studies conducted on a Mardin silt loam (coarse-loamy, mixed, mesic Typic Fragiudepts). The three trials in Columbia County were conducted on Hoosic gravelly silt loam (sandy-skeletal, mixed, mesic Typic Dystrudepts). Two of the three studies were N rate studies. The third study was a stand height study. Table 1 shows soil characteristics of each of the sites. At each location, trials were conducted as complete randomized block designs with N rate (0, 50, 100, 150, 200 and 250 lbs/acre per cut) and/or stand height at harvest (30 to 80 inches) as the treatments and four replicates per treatment. At the Tompkins County site, the main plots (N rates) were split to include a K rate (K₂O at 0, 100 or 200 lbs/acre per cut) as well. The BMR S×S was planted with a seed density of 60 to 70 lbs of seed per acre using conventional grain drills in the first two weeks of June with harvests at the end of July and the end of September. The subplots were 6 ft wide and 15 ft long, of which a 3-×-5-ft area was harvested. The Delaware County trials and the N rate studies in Columbia County were conducted on soils that had received manure in one or both of the past two years. The Tompkins County trials and the stand height study in Columbia County were conducted on soils that had not received any manure in recent years.

Fig. 2. Brown midrib sorghum-sudangrass nitrogen and potassium trial in Tompkins County, New York, in 2003.

Table 1. Soil types, soil pH, organic matter, and soil test P for each of the seven brown midrib sorghum-sudangrass nitrogen, potassium, and harvest stand-height studies conducted in New York.

Trial	County	Year	Study	Soil pH	Morgan soil test P		P addition (lbs of P₂O₅ per acre)
Trial	County	Year	Study	Soil pH	lbs/ acre	Class- ification	P addition (lbs of P₂O₅ per acre)
1	Tompkins	2002	N, K rates	6.2	5	Medium	45
2	Tompkins	2003	N, K rates	6.0	5	Medium	45
3	Columbia	2002	N rate	6.2	10	High	26
5	Columbia	2003	N rate	6.2	10	High	26
4	Columbia	2002	Height	6.2	10	High	26
6	Delaware	2002	Height	5.6†	20	High	20
7	Delaware	2002	Height	6.7	47	Very high	0

† Lime was applied to achieve a pH of 6.4.

All trials were managed as a 2-cut system. Harvest took place when the predetermined stand height was reached (height studies) or when an optimum stand height of 45 to 50 inches (5) was reached (N and K rate studies). Dry matter yields were measured and subsamples were analyzed for P according to Greweling (3). Samples were dry-ashed for 4 h at 500°C, cooled, and then dried again on a 100 to 120°C hot plate after addition of 3 ml of 6 N HCl. Ashed samples were extracted in dilute acid (1.5 N HNO₃ and 0.5 N HCl), and plant P concentrations were determined by analyzing dissolved minerals using a Thermo Jarrel Ash IRIS Advantage Inductively Coupled Plasma Radial Spectrometer (Jarrell Ash, Franklin, MA).

Studies were analyzed by site and by cut as well as averaged over the entire growing season (i.e., P concentrations averaged over the two cuts and P removal for full-season yields). The Columbia County N rate and all height studies were analyzed using PROC MIXED in SAS (10) with block effects and interactions including blocks as random effects and N rate or dry matter yield for each time of harvest as fixed effects. The N × K studies in Tompkins County were analyzed as a split-plot design using PROC MIXED with N application rates as the main treatment and K application as the subplots. Mean differences were considered significant if P < 0.05. For the individual stand height studies, regression analyses were performed using PROC REG in SAS with total P removal (total yield) or P concentration as dependent variables and dry matter yield per acre as the independent variable. PROC REG was also used to determine the relationship between P concentrations and dry matter yields across all trials.

Impacts on Phosphorus Removal by BMR S×S

The original treatment (stand height at harvest, N or K rate) or harvest number (first or second cut harvest) did not significantly impact P concentrations in the forage with the exception of plots which did not receive fertilizer N; in the trials in Tompkins County the average P concentration was 0.33% without N addition versus 0.28 to 0.30% P with N application. In the Columbia County trials where manure had been applied, the average P concentrations were 0.30% (2002) and 0.37% (2003) versus 0.25 to 0.27% (2002) and 0.27 to 0.34% (2003) when N had been applied. This decrease in P concentration of the forage with N application is most likely a dilution effect caused by the substantial (1.3 to 2.0 fold) yield increase upon the addition of N at just 50 lbs/acre per cut. Although values can vary considerably depending on plant species, plant age, and concentration of other mineral elements, a P concentration of 0.20% or greater is considered sufficient for adequate production (7).

There were no consistent trends in P concentration of the forage when N had been applied. Average P concentrations over the two cuts in each location ranged from 0.24%, at one of the 2002 Delaware County sites where the soil tested very high in P but where no additional P fertilizer was used, to 0.31% for the 2003 trials (Table 2). Phosphorus concentrations from individual plots ranged 0.15% to 0.53% with an overall average of 0.29% P (dry matter basis) if the plots without N application were included and 0.28% when N limited plots were excluded. This is greater than the 0.21% reported for S×S by Lang (6), similar to the 0.33% P reported by Johnston and Usherwood (4), but considerably lower than the 0.44% reported by Penn State College of Agricultural Sciences Cooperative Extension (9). Differences in P concentration among trial sites in New York in 2002 and 2003 were not explained by soil test P level. The forage P concentration was less than 0.20% in two instances (Delaware County stand height trials) only. In both cases, harvest took place at stand heights of 54 and 55 inches and dry matter yields of 2.9 to 3.3 tons/acre. Regression parameters for each of the trials individually are reported in Table 2.

Table 2. Phosphorus concentrations (± 1 standard deviation) and ranges in brown midrib sorghum-sudangrass grown in seven trials in New York as well as regression parameters (± 1 standard error) for the prediction of P₂O₅ removed with harvest (lbs of P₂O₅ per acre) as impacted by dry matter yield (tons/acre).

Trial	Soil P	Intercept	Slope	Model fit		Mean P	Min	Max
Trial	Soil P	Intercept	Slope	P value	r²	(%)
1	Medium	3.1±1.1	11.4±0.4	<0.0001	0.92	0.27±0.04	0.20	0.44
2	Medium	5.0±1.6	12.6±0.5	<0.0001	0.90	0.31±0.04	0.23	0.42
3	High	18.8±3.3	7.3±0.9	<0.0001	0.80	0.27±0.03	0.21	0.35
4	High	24.7±12.6	9.2±2.2	<0.0001	0.49	0.31±0.08	0.22	0.53
5	High	-1.9±7.5	14.1±1.9	0.0006	0.85	0.29±0.04	0.22	0.37
6	High	12.7±3.3	8.5±0.7	<0.0001	0.91	0.26±0.04	0.15	0.33
7	Very High	3.4±6.0	10.1±1.3	<0.0001	0.82	0.24±0.03	0.19	0.33

Phosphorus removal was linearly related to dry matter yield in both cuts. Across all sites and trial years, P removal by the crop was estimated as:

P removal (lbs of P₂O₅ per acre) = 4.8 + [11.6 × yield (dry matter tons/acre)]

(r² = 0.85, P < 0.001). Thus, 85% of the variability in P removal rates across all trials was explained by yield alone (Fig. 3). According to this equation, a BMR S×S crop with 5.6 tons of dry matter (16 tons at 35% dry matter) would remove about 70 lbs of P₂O₅ per acre. This is 8 lbs of P₂O₅ per acre more than the 62 lbs of P₂O₅ per acre estimated for a similar corn silage yield using an average corn silage P concentration of 0.24% of dry matter (2). However, actual P removal rates for BMR S×S ranged from 50 to 90 lbs of P₂O₅ per acre. The additional variability in P concentrations and removal for the individual trials shown in Fig. 3 was not explained by N (beyond a 50-lbs/acre application of N) or K application rate, stand height at harvest, first or second cut, or soil test P, although at one of 2002 Delaware County sites (site 6), P concentrations tended to decrease slightly with stand-height at harvest (r² = 0.53, P < 0.001). Additional research is needed to identify location differences that may have resulted in varying P concentrations in harvest.

Fig. 3. Phosphorus removal by brown midrib sorghum-sudangrass versus dry matter yield for 7 field trials conducted in 2002-2003 in 3 regions of New York State.

Conclusions and Recommendations

Differences in yields across 3 locations and two years of field trials explained 85% of the variability in P removal rates. The average P concentration across all sites was 0.28% P. A crop of 5.6 tons of dry matter would remove P₂O₅ at 70 lbs/acre. This is the P₂O₅ equivalent of 5000 gallons of liquid manure assuming an average P concentration in the manure of 13.6 lbs P₂O₅ per 1000 gallons, the average manure composition of 503 samples analyzed in 2003 by Dairy One (Paul Sirois, personal communication). However, S×S P concentrations of individual plots ranged from 0.15% to 0.53% indicating that producers need to analyze their forage for P concentration and determine yields to obtain accurate values for P removal by this crop in addition to sampling manure to determine manure application rates for environmentally-sound nutrient management planning. Additional studies are needed to determine what causes the site-to-site differences observed in this study.

Acknowledgments

This research was funded with grants from Garrison and Townsend Inc. and the Phosphate and Potash Institute, and in-kind donations of Honeywell Inc., Cornell Cooperative Extension of Rensselaer County, and Cornell Cooperative Extension of Delaware County.

Literature Cited

1. Czymmek, K. J., Ketterings, Q. M., Geohring, L. D., and Albrecht, G. L. 2003. New York State phosphorus runoff index. Dept. Crop Soil Sci. Ext. Bull. E03-13.

2. Dairy One. 2004. Field composition library: Corn silage, accumulated crop years: 05/01/2000 through 04/30/2003. Online. Dairy One Forage Laboratory. Ithaca, New York.

3. Greweling, T. 1976. Dry ashing. Cornell Univ. Agric. Exp. Stat. Res. Bull. 6:1-35.

4. Johnston, A. M., and Usherwood, N. R. 2002. Crop nutrient needs. Pages 13-21 in: Plant Nutrient Use in North American Agriculture, PPI/PPIC/FAR Tech. Bull. 2002-1.

5. Kilcer, T. F., Ketterings, Q. M., Cherney, J. H., Cerosaletti, P., and Barney, P. Optimum stand height for forage brown midrib sorghum × sudangrass in Northeastern USA. J. Agron. Plant Sci. In press.

6. Lang, B. 2001. Sudan/sorghum forage management. Iowa State University Extension Fact Sheet BL-50.

7. Marschner, H. 1995. Mineral nutrition of higher plants. Academic Press Inc. San Diego, CA.

8. New York Agricultural Statistics Services. 2003. 2002 New York Agricultural Statistics. Online. USDA-NASS. Albany, NY.

9. Penn State College of Agricultural Sciences Cooperative Extension. 2004. Soil testing, recommendations, phosphorus and potassium recommendations. Online. The Agronomy Guide: Part 1, Sect. 2, Soil fertility management. Penn State Coll. Agric. Sci., Dept. Crop Soil Sci. University Park, PA.

10. SAS Institute Inc., 1999: SAS/STAT User’s Guide. Release 8.00. SAS Institute Inc., Cary, NC, USA.

11. Sharpley, A. N., Weld, J. L., Beegle, D. B., Kleinman, P. J. A., Gburek, W. J., Moore, P. A., Jr., and Mullins, G. 2003. Development of phosphorus indices for nutrient management planning strategies in the United States. J. Soil Water Conserv. 58:137-152.

12. U.S. Environmental Protection Agency. 2002. Concentrated Animal Feeding Operations (CAFO)-Final Rule. Online. Nat. Pollutant Discharge Elimination Syst. (NPDES).