No abstract

Soybeans [Glycine max (L.) Merr.] have not responded favorably to no‐till culture on some Mississippi soils. Positional unavailability of surface‐applied fertilizers may be one reason for the poor response. A 2 X 2 X 3 factorial field experiment with five replications in a randomized complete block split‐split plot design was used to measure effects of tillage (conventional vs. no‐tillage), P and K placement (broadcast vs. injected), and P and K rates (0 and 0, 13 and 37, 40 and 112 lb/acre P and K, respectively) on soybean growth and yield from 1983 to 1985 on three soils. The soils were Okolona silty clay (fine, mont‐morillonitic, thermic Typic Chromudert) at Brooksville, Memphis silt loam (fine‐silty, siliceous, thermic Typic Hapludalf) at Raymond, and Prentiss very fine sandy loam (coarse‐loamy, siliceous, thermic Glossic Fragiudult) at Newton. All soils were initially low in P and K. Seed yields in general were lowest at Brooksville, and ranged from 16.4 to 22.4 bu/acre with a 3‐yr average of 19.4 bu/acre. Yields were much better at the two other sites, except Raymond in 1983 when seed yield averaged 16.9 bu/acre. The Raymond site had the best overall average yield of 29.2 bu/acre, while the 2‐yr average at Newton was 28.2 bu/acre. No‐tillage reduced growth and yields on the Okolona and Memphis soils. On the Prentiss soil, only growth was reduced by no‐tillage. Under no‐till conditions,injection of P and K produced greater yields than broadcast on the Okolona and Memphis soils but not on the Prentiss soil. There was little response to rate of P and K rates on either soil. The low yields on Okolona silty clay appeared to be related to variables not considered in this study.

The C2H2‐block method was used to measure denitrification in a maize (Zea mays L.) field over the 85‐d period between sidedressing with N fertilizer and harvest. Measurement locations were changed weekly in order to avoid problems associated with microbial utilization of C2H2 and inhibition of nitrification. Nitrous oxide (N2O) fluxes were measured both with and without C2H2 addition to soil. Without C2H2 addition to soil, 0.3 kg N ha−1 was evolved from both unfertilized and Ca(NO3)2 treated soil, whereas 2.5 kg N ha−1 was evolved from urea treated soil. This difference was attributed to formation of N2O during nitrification of NH+4 liberated from urea. With C2H2 addition to soil, 1.6, 2.0, and 2.4 kg N ha−1 were evolved as the result of denitrification in the zero N, Ca(NO3)2, and urea treated sites, respectively. Total gaseous N loss from a particular N treatment was approximated by adding together the N2O‐N loss with and without C2H2 addition to soil. The gaseous N loss associated with fertilizer addition was <3% of that applied. The ratio of N2/N2O evolved without C2H2 addition to soil was estimated to be in the range of 4:1 to 7:1 for the zero N and Ca(NO3)2 treatments and about 1:1 for the urea treatment. Denitrification was responsible for at least 80% of the measured gaseous N loss from the zero N and Ca(NO3)2 treatments. About equal amounts of gaseous N loss from the urea treatment occurred via denitrification and during nitrification of liberated NH+4.

Past models of denitrification are reviewed. None of these models include a description of anaerobic volume and transient solute diffusion to and away from zones of N reduction. In this paper, models describing denitrification in the anaerobic zone of a saturated homogeneous soil are presented. The models consider the reduction sequence: nitrate (NO−3) → nitrite (NO−2) → nitrous oxide (N2O) → dinitrogen (N2). Anaerobic volume was described by an approximate solution to a moving boundary problem. Transient diffusion of NO−3, NO−2, N2O, and N2 was allowed throughout the saturated soil, while reduction of the nitrogenous species occurred only in the anaerobic region. Four types of reaction terms were considered, resulting in a zero‐order model, a Michaelis‐Menten model with differential rates of reduction, a Michaelis‐Menten model with NO−3 and NO−2 inhibition of N2O reduction, and a competitive inhibition model. The use of an operational Vmax which reflects electron donor availability was proposed.

No abstract