A Simple Soil Test for Detecting Sites that are Nonresponsive to Nitrogen Fertilization

of low correlations with the production of mineral N and crop N uptake. Recent work indicates that accumulation of amino sugar N in soil Soil testing for NO3 is currently considered the best reduces the yield response of corn (Zea mays L.) to N fertilization, and that nonresponsive sites are detectable by determination of amino option for identifying sites where N fertilization will be sugar N in soil hydrolysates. Unfortunately, the hydrolysis process is ineffective in producing a yield response by corn (Bundy too complicated and time-consuming for use in routine soil testing. and Meisinger, 1994). Two soil NO3 tests have been A much simpler technique was developed to estimate amino sugar N developed that differ in the time and depth of sampling. without the need for acid hydrolysis. In this test, 1 g of air-dried soil With the preplant NO3 test (PPNT), profile samples are is treated with 10 mL of 2 M NaOH in a 473-mL (1-pint) wide-mouth collected in the early spring to a depth of 60 or 90 cm, Mason jar, and the sample is heated for 5 h at 48 to 50 C on a hot to account for carryover of mineral N from previous plate to liberate (NH4 amino sugar)-N as gaseous NH3. The NH3 cropping (e.g., Bundy and Malone, 1988; Roth and Fox, is collected in H3BO3–indicator solution, and subsequently determined 1990; Schmitt and Randall, 1994). With the presidedress by acidimetric titration. Recovery ranged from 97 to 102% when NO3 test (PSNT), soil sampling is done to a depth of analyses were performed after treating samples with 15N-labeled (NH4 )2SO4 or glucosamine, but did not exceed 6.5% with labeled 30 cm in late spring, so that soil N mineralization can glycine and was undetectable with labeled NO3 or NO2. Comparative be taken into account and supplemented, if necessary, studies using 12 nonresponsive and 13 responsive soils showed a very by sidedressing (e.g., Magdoff et al., 1984; Fox et al., high correlation between soil-test N and hydrolyzable amino sugar N 1989; Blackmer et al., 1989; Meisinger et al., 1992; Bundy (r 0.90***). Test values were significantly higher (P 0.001) for and Andraski, 1993). The PSNT has been recommended nonresponsive (237–435 mg N kg 1) than for responsive (72–223 mg more widely than the PPNT in the eastern USA, but N kg 1) soils. The soil test described has important economic implicausage has been limited by the need to collect soil samples tions for production agriculture, and also should be of value for conduring the growing season, and by the fact that N fertiltrolling NO3 pollution of ground and surface water. ization must be postponed until after testing and can be ineffective if adverse weather conditions delay sidedressing. Besides logistical problems, an inherent limitaS testing is used routinely to guide agricultural tion with the PPNT and PSNT arises from the extensive applications of limestone, P, and K, whereas N apspatial and temporal variability in soil NO3 concentraplications for corn production are often based on a yield tions, which depend on numerous N-cycle processes, goal, with adjustments to allow for other N inputs, such including mineralization, immobilization, nitrification, as legumes and manure. A yield-based recommendation denitrification, leaching, and plant uptake. Consequently, may have merit on a long-term basis, but under or overa one-time test for soil NO3 is apt to be of little value for fertilization is apt to occur in any given growing season predicting crop N availability throughout the growing since soil N availability is not taken into account. Insuffiseason, particularly in a humid region where these procient application of N can have serious economic consecesses occur extensively. quences for the farmer, whereas excessive fertilization Ideally, a soil test for N would estimate a labile orincreases the risk of environmental pollution. ganic fraction that supplies the plant through mineralEstimation of plant-available N is complicated enorization. Such an approach would have the major advanmously by the dynamic nature of soil N, owing largely tage over NO3 testing that soil test levels would depend to the effects of temperature and moisture supply on on fewer N-cycle processes and should, therefore, be N-cycle processes. Numerous biological and chemical less variable. This would make the time of soil sampling methods have been proposed as an index of soil N availmuch less critical than with NO3 testing, so that soil N ability (Bremner, 1965; Keeney, 1982; Stanford, 1982; availability could potentially be predicted on the basis Bundy and Meisinger, 1994), but none has been adopted of a one-time test prior to the growing season. widely for soil testing. Biological methods are necessarNumerous attempts have been made to identify a ily time-consuming because of the need for incubation, labile pool of soil organic N through chemical fractionand the results represent the net effect of mineralizaation of the N in soil hydrolysates (e.g., Keeney and tion-immobilization turnover rather than gross mineralBremner, 1964; Porter et al., 1964; Khan, 1971; Smith ization. Chemical methods of estimating potentially and Young, 1975; Meints and Peterson, 1977), but with mineralizable soil N have been based on an empirical little tangible progress. The stagnation can be attributed, approach, and their use has been very limited because at least in part, to serious defects in methodology that vitiated analyses for amino sugar N and amino acid N. S.A. Khan and R.L. Mulvaney, Dep. of Natural Resources and EnviThese defects were identified and eliminated through a ron. Sci.; and R.G. Hoeft, Dep. of Crop Sci., Univ. of Illinois, 1102 substantial effort that ultimately led to simple diffusion S. Goodwin Ave., Urbana, IL 61801. This study was a part of Project ILLU-65-0326, Illinois Agric. Exp. Stn. Received 20 Feb. 2001. *CorreAbbreviations: LSD, least significant difference; PPNT, preplant nisponding author ([email protected]). trate test; PSNT, presidedress nitrate test; *** significant at the 0.001 probability level. Published in Soil Sci. Soc. Am. J. 65:1751–1760 (2001).