Illinois Fertilizer
Conference Proceedings
January 24-26, 2000
| Home | 2000 Index | Search |
|
|
| Home | 2000 Index | Search |
R.L. Mulvaney, S.A. Khan, and R.G. Hoeft1
In Illinois, N fertilizer rates for corn production are based on the yield goal and corn:N price ratio, with adjustments to allow for other N inputs, such as legumes and manure. During the past decade, there has been growing interest in soil testing to estimate N availability, so as to improve the accuracy of N fertilzer recommendations. The most common approach has been soil nitrate (NO3-) testing. For several years, a preplant NO3- test (PPNT) has been used with some success west of the Missouri River. More recently, a presidedress NO3- test (PSNT) has been developed to improve the accuracy of N rate recommendations in the eastern U.S. (Magdoff et al., 1984).
In a previous FREC project, the PPNT and PSNT were evaluated for their accuracy in predicting N fertilizer applications to optimize corn yield, relative to recommendations by the conventional approach (Brown el al., 1993) The results showed that neither of the soil tests provided sufficient advantage over the conventional approach to justify their use. The results also showed no crop response to N fertilization at 33 of the 75 study sites. At 12 of the nonresponding sites, the lack of response could not be attributed to a specific cause (drought, a recent application of manure, or previous cropping to a forage legume). Moreover, examination of the yield-response data from this study (Brown, 1996) reveals that, even for the 42 responding sites, N fertilization was often not economical.
When soils are fertilized with N, a substantial proportion of the fertilizer N (usually between 10 and 20%) is incorporated (in other words, immobilized) into organic forms such as proteins and amino sugars during the synthesis of microbial biomass. As the microbes die and decay, some of the biomass N is released as ammonium (NH4+) through the process of mineralization; the remainder undergoes conversion to more stable organic N compounds, ultimately becoming a part of soil organic matter. Several studies have demonstrated that recently immobilized N (including amino acids and amino sugars) is mineralized much more rapidly than is native soil orgainc N (for example, Allen et al., 1973; Shen et al., 1989). There is growing evidence that long-term application of N fertilizers promotes the buildup of rapidly mineralizable (in other words, labile) N in soils, and may reduce crop response to further applications of fertilizer N (Motavalli et al., 1992; Stevens et al., 1996).
The organic N in soil occurs largely in association with humic colloids and clay minerals. To liberate this N through acid hydrolysis, the soil is heated with mineral acid (usually 3 or 6 M HCl) for several hours, after which N in the hydrolysate is separated into different fractions by steam distillation (Bremner, 1965; Stevenson, 1982, 1996). The major fractions include total hydrolyzable N, hydrolyzable NH3-N, (NH3 + amino sugar)-N, and amino acid-N.
Several studies have been reported that compared the distribution of organic N in different soils or among soils under different management practices (for example, Keeney and Bremner, 1964; Porter et al., 1964; Osborne, 1977). The results have generally indicated little variation in the distribution of N, regardless of soil type, cropping, or cultivation. The same sort of uniformity was observed when N-distribution analyses were performed on soil samples collected by Brown (1996) that differed in crop responsiveness to N fertilization. Subsequent studies revealed that existing methods of recovering amino acid-N and (NH3 + amino sugar)-N from soil hydrolysates are subject to serious underestimation. Therefore, new methods were developed to fractionate hydrolyzable soil N, utilizing Mason-jar diffusion techniques developed in a previous FREC project for determination of inorganic N in soils and soil extracts (Khan et al., 1997, 2000; Mulvaney et al., 1997). In the work reported herein, N-distribution analyses by the new methods were employed to identify a labile fraction of soil organic N that reduces crop responsiveness to N fertilization.
The soils used (Table 1) were surface (0-30 cm) samples from 15 of the 75 sites studied by Brown (1996). The particular samples used were selected from sites receiving normal rainfall, and included the seven sites that were least responsive to fertilizer N and the eight sites that were most responsive. The samples had been collected in late March or early April of 1990, 1991, or 1992, prior to planting corn. Following collection, the samples had been air dried and crushed to pass through a 2-mm screen. In the analyses reported in Table 1, pH was determined with a glass electrode; organic matter was estimated from organic-C analyses by the method of Mebius (1960), using a factor of 1.8 to convert organic C to organic matter, total N was measured by a semimicro-Kjeldahl procedure (Bremner, 1996); and inorganic N was determined by diffusion of 2 M KCl soil extracts, using the accelerated Mason jar techniques described by Khan et al. (1997). Analyses for organic C and total N were performed on < 0.15-mm soil. The other analyses reported were performed on < 2-mm soil. In all cases, there were three or four replicate determinations.
Yield-response data reported by Brown (1996) were used to quantify crop N response using the expression, (Yh-Yl)/(Fh-Fl), where Y is corn grain yield obtained with the highest (h) or lowest (l) N rate used, and F is the corresponding rate of fertilizer N applied. The results, expressed as bushels of grain yield per pound of N applied, are reported in Table 1 for each of the 15 study sites selected.
To prepare soil hydrolysates, quadruplicate samples of soil (< 0.15 mm) containing 10 mg of N were heated under reflux for 12 h after treatment with 20 mL of 6 M HCl and two drops of octyl alcohol. The hydrolysis mixture was filtered through Whatman no. 50 filter paper under vacuum. Replicate hydrolysates were combined and subsequently neutralized by addition of NaOH (Bremner, 1965; Stevenson, 1982, 1996). The hydrolysates were stored under refrigeration (5°C).
Reagent-grade alanine (an amino acid) and glucosamine hydrochloride (an amino sugar) were obtained from Sigma, St. Louis, MO. Before use, the compounds were dried over anhydrous CaSO4 in a desiccator. Aqueous solutions containing 1 g N L-1 were prepared by dissolving each compound in 25 mL of deionized water. The solutions were used within 24 h after preparation and were stored in a refrigerator (5°C) before use.
Procedures to determine total hydrolyzable N, NH3-N, (NH3 + amino sugar)-N, and amino acid-N were performed as described by Bremner (1965) and Stevenson (1982, 1996).
Apparati
Aluminum block digester. A Tecator Model 1016 Digester equipped with a Model 1012 Autostep Controller (Tecator, Höganäs, Sweden) was used.
Diffusion unit. The diffusion unit used consists of a 1-pint (473-mL) wide-mouth Mason jar equipped with a lid that has been modified to support the bottom of a 60-mm (dia.) Pyrex petri dish (Khan et al., 1997; Mulvaney et al., 1997).
Electric hot plate. A commercial griddle (West Bend Model 76212) was used. Before use, the heat control was adjusted such that the desired temperature was obtained when a thermometer was immersed in 100 mL of deionized water in a Mason jar placed in the center of the griddle.
Automatic titrator. Titrations were performed using a Metrohm Model 678 EP/KF Processor equipped with a Model 665 Dosimat (Metrohm, Herisau, Switzerland) and a microelectrode (Model MI-414P, Microelectrodes, Bedford. NH).
Reagents
Potassium sulfate-catalyst mixture. Twenty g of CuSO4•5H20 was powdered by grinding in a mortar, and was then mixed intimately with 2 g of Se and 200 g of K2SO4 (powder).
Concentrated sulfuric acid (18 M).
Dilute sulfuric acid (5 M).
Dilute sulfuric acid (0.01 M standard).
Sodium hydroxide solution (approximately 10 M).
Boric acid-indicator solution. A reagent containing 40 g of H3BO3 L-1 was prepared as described previously (Khan et al., 1997; Mulvaney et al., 1997).
Magnesium oxide. The heavy powder supplied by Fisher Scientific, Pittsburgh, PA, was used.
Ninhydrin solution. Twenty-five g of certified ninhydrin (Fisher Scientific) was dissolved in 250 mL of reagent-grade methanol.
Procedures
Total hydrolyzable N. Five mL, of soil hydrolysate was pipetted into a 50-mL Pyrex digestion tube and treated with 0.5 g of K2SO4-catalyst mixture and 2 mL of 18 M H2SO4 The tube was transferred to an A1 block digester and heated for 1.5 h at 150°C, then for 1 h at 250°C, and finally for 3 h at 350°C. After cooling, the digest was diluted with approximately 2 mL of deionized water from a wash bottle, homogenized by vortex mixing, and decanted into a Mason jar. The latter operation was repeated three times to ensure complete transfer of N in the digest to the Mason jar. The digest was then neutralized with 10 mL of 10 M NaOH, which was dispensed onto the side of the jar to neutralize. any acidity deposited during transfer of the digest. Within 30 s, the jar was sealed by attaching a lid equipped with a petri dish containing 5 mL of H3BO3-indicator solution, and the jar was swirled to mix the contents and then transferred to a hot plate maintained at 50 ± 2°C. After 4 h, the jar was removed from the hot plate, and NH4+-N in the H3BO3 solution was determined by titration with 0.01 M H2SO4.
Hydrolyzable ammonia-N. Ten mL of soil hydrolysate in a Mason jar was treated with 0.05 g of MgO using a calibrated spoon. The jar was swirled to mix the contents, then sealed by attaching a lid with a Petri dish containing 5 mL of H3BO3-indicator solution. Diffusion was performed for 2 h at 50°C on a hot plate, followed by titrimetric determination of NH4+-N.
(Ammonia + amino sugar)-N. To 10 mL of soil hydrolysate in a Mason jar was added 2 mL of 10 M NaOH. After swirling thejar to mix the contents, a lid having a petri dish with 5 mL of H3BO3 solution was attached, and the jar was heated on a hot plate (50°C) for 5 h. The amount of NH4+-N collected was determined as described previously.
Amino acid-N. After completing diffusion of (NH3 + amino sugar)-N, 2.5 mL of 5 M H2SO4 was added to the jar, followed by 1 mL of ninhydrin solution. The jar was swirled to mix the contents, and after being covered with an unmodified lid to minimize the loss of water, was heated for 90 min at 65 to 70°C on a hot plate. A few minutes were allowed for the jar to cool, after which the contents were treated with 1 mL of 10 M NaOH and mixed by swirling. The jar was sealed by attaching a lid with 5 mL of H3BO3 solution in a petri dish, then heated at 50°C for 2 h on a hot plate. The H3BO3 solution was titrated as described previously.
The present project was motivated by the lack of N response observed at many of the sites used by Brown (1996) to compare different N recommendation systems for corn, and by the widespread occurrence of these nonresponsive sites throughout Illinois. Besides the proven yield approach described in the Illinois Agronomy Handbook, N recommendations were made on the basis of soil NO3- tests before (PPNT) or after (PSNT) planting. Surprisingly, neither test was effective in detecting nonresponsive sites. This finding can probably be attributed to the dynamic nature of N-cycle processes in soil, which depend greatly on weather conditions. Most of these processes affect soil NO3- concentrations, and therefore a one-time test for NO3- is likely to be of little value for predicting crop N availability throughout the growing season.
The mineralizable N in soil occurs largely as labile constituents of microbial biomass, such as proteins and amino sugars. As compared to soil NO3-, lower variability would be expected in the concentrations of these constituents, since they are involved in fewer N-cycle processes. This suggests that chemical fractionation of soil organic N, and particularly analyses of amino acid-N and amino sugar-N, could have potential value in detecting sites that are nonresponsive to N fertilization.
The N in soil hydrolysates can be fractionated using the steam-distillation methods originated by Bremner (1965), so these methods were initially utilized in comparing the distribution of N among sites found by Brown (1996) to differ widely in crop N responsiveness. No consistent differences were detected between responsive and nonresponsive sites in the concentration of amino acid-N or amino sugar-N. This finding can be attributed to inherent defects in methodology, based on subsequent recovery tests showing serious underestimation and extensive variability when steam distillations were performed on purified samples of alanine and glucosamine.
Given the difficulties that vitiated the use of steam distillation for fractionation of soil hydrolyzable N, a substantial effort was made to develop new methods for this purpose by adapting Mason-jar diffusion methodology for inorganic-N analysis of soils and soil extracts (Khan et al., 1997, 2000; Mulvaney et al., 1997). This endeavor proved quite successful and ultimately led to the diffusion methods described for determination of total hydrolyzable N, hydrolyzable NH3-N, (NH3 + amino sugar)-N, and amino acid-N. The reliability of these methods was confirmed through specificity tests using purified samples of amino acids, amino sugars, purines, pyrimidines, and amides, and by checking the recovery of 15N-labeled (NH4)2SO4 or glycine added to soil hydrolysates (data not reported). Unlike steam distillation, recovery by diffusion was quantitative when analyses for amino acid-N were performed on different amountsof alanine, and when analyses for (NH3 + amino sugar)-N were performed on different amounts of glucosamine (Table 2).
Table 3 compares the results obtained by distillation and diffusion in estimating hydrolyzable NH3-N, amino acid-N, and amino sugar-N for five of the soils collected by Brown (1996). Whereas good agreement was generally observed for NH3-N, analyses of amino acid-N and amino sugar-N were always lower by distillation than by diffusion. The difference was particularly striking for amino sugars, in which case analyses by diffusion were usually at least threefold higher than those by distillation. This is the same trend that was observed for the recovery tests reported in Table 2 and can no doubt be attributed to serious underestimation when steam distillation is employed for determination of amino acid-N and (NH3 + amino sugar)-N.
Based on the quantitative recoveries achieved for chemical standards (Table 2) and the high concentrations of amino acid-N and amino sugar-N obtained for soil hydrolysates (Table 3) by the diffusion methods described, there was good reason to expect a difference in the distribution of hydrolyzable N when these methods were applied to a set of soil samples collected from sites that differed in crop N responsiveness. Table 4 shows that this expectation was not in vain. A distinct difference was observed between responsive and nonresponsive soils in their content of amino sugar-N, but not in their content of total hydrolyzable N, hydrolyzable NH3-N, or amino acid-N. Examination of Table 4 reveals that the concentration of amino sugar-N was always higher for nonresponsive soils as compared to responsive soils. The lowest value for any nonresponsive soil was nearly twice as large as the highest value for any responsive soil, and on average, the difference was more than fivefold.
The clear distinction between responsive and nonresponsive soils on the basis of their content of amino sugar-N (Table 4) is truly remarkable, given the fact that both groups include soils that range widely in their properties and were collected throughout Illinois (Table 1). Equally important, a dramatic difference was observed between these groups in fertilizer-N responsiveness. This is evident from the fertilizer-response data reported in Table 1, which were almost always negative for nonresponsive sites and always positive for responsive sites, in which case grain yield increased by 0.3 to 0.6 bushel per pound of fertilizer N.
Further evidence of a close relationship between soil amino sugar content and nonresponsiveness to N fertilization is provided by Table 5, which shows the r-values obtained when simple correlations were performed to relate the fertilizer-response data in Table 1 to different soil N fractions. Among the parameters tested, the highest statistical significance was obtained for amino sugar-N. The correlation was negative, as would be expected for any potentially available form of soil N. Significant correlations were also obtained for NO3--N, (NH4+ + N03-)-N, total hydrolyzable-N, and amino acid-N, but not for organic matter, total soil N, exchangeable NH4+-N, or hydrolyzable NH3-N. The lack of a significant relationship for organic matter is noteworthy but not particularly surprising in view of the fact that a responsive soil (no. 12) had the highest organic matter content.
A study was conducted to ascertain whether a specific soil N fraction can be identified that reduces the yield response of corn to N fertilization, even with favorable growing conditions. Initial attempts to distinguish between responsive and nonresponsive sites were unsuccessful, owing to inherent defects in existing steam-distillation methods for fractionating hydrolyzable forms of soil N. Simple diffusion techniques were subsequently developed that quantitatively recover amino acid-N and amino sugar-N in soil hydrolysates. When these techniques were applied to soil samples from sites that differed dramatically in crop N response, a much higher content of amino sugar-N was detected for seven nonresponsive sites than for eight responsive sites. A very highly significant negative correlation was obtained in relating soil amino sugar content to fertilizer-N response, which suggests that this form of organic N is readily mineralized during the growing season. There is good reason to expect that a routine soil test could be developed to detect sites that will not respond to N fertilization.
Table 1. Chemical Properties of soil from N-response sites
Table 3. Comparison of distillation and diffusion for N-distribution analysis of soil hydrolysates.
Table 4. Distribution of hydrolyzable N for soil from N-response sites.
Table 5. Correlations between N-fertilizer response and soil chemical properties
1 R.L. Mulvaney is a Professor and S.A. Khan is a Research Associate, Dept. of Natural Resources and Environmental Sciences, Univ. of Illinois. R.G. Hoeft is a Professor, Dept. of Crop Sciences, Univ. of Illinois.
Allen, A.L., F.J. Stevenson, and L.T. Kurtz. 1973. Chemical distribution of residual fertilizer nitrogen in soil as revealed by nitrogen-15 studies. Journal of Environmental Quality, 2:120-124.
Bremner, J.M. 1965. Organic forms of nitrogen. In: Methods of Soil Analysis. Part 2 (C.A. Black et al., ed.). Agron. Monogr. 9. American Society of Agronomy, Madison, WI. pp. 1238-1255.
Bremner, J.M. 1996. Nitrogen –Total. In: Methods of Soil Analysis. Part. 3. Chemical Methods (D.L. Sparks et al., ed.). SSSA Book Set. 5. Soil Science Society of America, Madison, WI. pp. 1085-1121.
Brown, H.M. 1996. Evaluation of nitrogen availability indices. Ph.D. thesis, University of Illinois, Urbana-Champaign, IL.
Brown, H.M., R.G. Hoeft, and E.D. Nafziger. 1993. Evaluation of three N recommendation systems for corn yield and residual soil nitrate. In: 1993 Illinois Fertilizer Conference Proceedings (R.G. Hoeft, ed.). pp. 43-49.
Keeney, D.R., and J.M. Bremner. 1964. Effect of cultivation on the nitrogen distribution in soils. Soil Science Society of America Proceedings, 28:653-656.
Khan, S.A., R.L. Mulvaney, and C.S. Mulvaney. 1997. Accelerated diffusion methods for inorganic-nitrogen analysis of soil extracts and water. Soil Science Society of America Journal, 61:936-942.
Khan, S.A., R.L. Mulvaney, and R.G. Hoeft. 2000. Direct-diffusion methods for inorganic-nitrogen analysis of soil. Soil Science Society of America Journal (in press).
Magdoff, F.R., D. Ross, and J. Amadon. 1984. A soil test for nitrogen availability to corn. Soil Science Society of America Journal, 48:1301-1304.
Mebius, L.J. 1960. A rapid method for the determination of organic carbon in soil. Analytica Chimica Acta, 22:120-124.
Motavalli, P.P., L.G. Bundy, W.W. Andrasaki, and A.E. Peterson. 1992. Residual effects of long-term nitrogen fertilization on nitrogen availability to corn. Journal of Production Agriculture, 5:363-368.
Mulvaney, R.L., S.A. Khan, W.B. Stevens, and C.S. Mulvaney. 1997. Improved diffusion methods for determination of inorganic nitrogen in soil extracts and water. Biology and Fertility of Soils, 24:413-420.
Osborne, G.1. 1977. Chemical fractionation of soil nitrogen in six soils from southern New South Wales. Australian Journal of Soil Research, 15:159-165.
Porter, L.K., B.A. Stewart, and H.J. Haas. 1964. Effects of long-time cropping on hydrolyzable organic nitrogen fractions in some Great Plains soils. Soil Science Society of America Proceedings, 28:368-370.
Shen, S.M., P.B.S. Hart, D.S. Powlson, and D.S. Jenkinson. 1989. The nitrogen cycle in the Broadbalk wheat experiment: 15N-labeled fertilizer residues in the soil and in the soil microbial biomass. Soil Biology & Biochemistry, 21:529-533.
Stevens, W.B., R.G. Hoeft, and R.L. Mulvaney. 1996. Effect of N fertilization on accumulation and release of readily-mineralizable organic N. In: 1996 Illinois Fertilizer Conference Proceedings (R.G. Hoeft, ed.). pp. 91-104.
Stevenson, F.J. 1982. Nitrogen – Organic forms. In: Methods of Soil Analysis. Part 2 (A.L. Page et al., ed.). Agron. Monogr. 9 (2nd ed.). American Society of Agronomy and Soil Science Society of America, Madison, WI. pp. 625-641.
Stevenson, F.J. 1996. Nitrogen – Organic forms. In: Methods of Soil Analysis. Part. 3. Chemical Methods (D.L. Sparks et al., ed.). SSSA Book Set. 5. Soil Science Society of America, Madison, WI. pp. 1185-1200.
