Green Plan Banner

Research Program Banner

Research Report  4.3

Influence of Soil Texture andTillage-Induced Changes
on the Susceptibility of Legume-N to Leaching

Dr. B.D. Kay, and Dr. V. Rasiah
Dept. of Land Resource Science
Univ. of Guelph, Guelph, ONT N1G 2W1
COESA Report No.:  LMAP-015/94

Objectives & Expected Outputs
Executive Summary
Table of Contents
View / Download Report  [133 KB pdf]

 

Report List | Green Plan Research

Objectives and Expected Outputs
Objectives:
  1. To determine the influence of variations in soil structure, bulk density, and the volume fraction of pores (VFP) belonging to different size classes, on N-mineralization subsequent to red clover incorporation and;
  2. To develop equations to predict the rates of N-mineralization on different soils.
Expected Outputs: To improve the understanding of the influence of legumes, used as winter cover crops or underseeded in row-crops, on soil N dynamics, and their place in developing sustainable crop production systems
Type: Contribution Agreement, University
Spending Profile: 93-94: $53.8 K
Status: Available March 1995

 

Executive Summary

A legume underseeded in cereal represents an important source of nitrogen N, for the subsequent crop in the rotation and can provide a complementary improvement in soil structure. The extent to which the N benefit is realized depends on the rate of mineralization of the legume N in relation to the N requirements of the succeeding crop. The rate of N mineralization may vary with inherent soil conditions as well as properties such as bulk density that change as tillage practices are changed.

The objectives of this study were:

  1. to determine the influence of variations in soil texture, bulk density, and the volume fraction of pores, VFP, belonging to different size classes, on N-mineralization subsequent to red clover incorporation and;

  2. to develop equations to predict the rates of N-mineralization on different soils.

Soil samples for the study were collected from Mr. Don Lobb's farm, near Clinton, in Huron County. This site was one of the T-2000 sites investigated during the Ontario Land Stewardship program and is one of the longest running field scale side-by-side comparison of zero and conventional tillage in Ontario. The comparison is maintained as transects about 500 m in length which traverse soils with clay contents ranging from 7 to 40%. The site was under corn production in 1991, 1992, and 1993.

Soil samples were collected from the conventionally tilled transect at 7 locations. Samples were collected from the top 15 cm depth, air-dried, sieved and the material < 4.00 mm characterized. Bulk soil properties (clay, silt, sand, organic matter, total-N and CaCO3 contents, pH and cation exchange capacity) were determined. Water retention characteristics were measured at two bulk densities that reflected conditions encountered under conventional and zero till conditions at this site. The bulk density for the low density condition ranged from 1.264 to 1.327 g cm-3 (depending on the texture) and that of the high density from 1.362 to 1.451 g cm-3.

Red clover (only the shoot biomass) was added at a rate equivalent to 6000 kg ha-1 (on dry weight basis) and was assumed to be incorporated in the top 15 cm depth. The total-N in the shoot biomass was 2.36% and the amount of legume-N added was about 140 kg ha-1 (75 mg legume-N kg-1 soil). The legume residue was incubated with soil at 23 1C and a soil water potential of -15 kPa.

The water retention and bulk density data indicated that the air-filled porosity, AFP, at -kPa for the low density condition tanged from 0.161 to 0.316% and that of the high density from 0.041 to 0.217%.

The nitrate-N data from the red clover added treatment was fitted to a two pool first order model, whereas the control could only be fitted to a one pool model. Each best fit was significant at P 0.05 and the R2 for the best fits ranged from 0.71 to 0.98. The size of the potentially mineralizable resistant N-pool in the red clover added treatment ranged from 28.16 to 68.12 mg kg-1 whereas that of the labile pool, N1, ranged from 10.32 to 27.11 mg kg-1. The corresponding range in the rate constant for the resistant pool, kr, was from 0.0038 to 0.0065 d-1 and that of the labile pool, k1 was from 0.0081 0.0596 d-1. The size of the N-pool in the control, No, ranged from 11.81 to 25.41 mg kg-1 and the corresponding range for the rate constant, ko, was from 0.0179 to 0.1313 d-1.

.Simple correlation analyses indicated the values of k1 were significantly influenced by the largest number of bulk soil properties, followed by N1 and Nr. The pedotransfer function, PTF, developed to predict k1 indicated that 93% of the variability in k1 was accounted for by bulk density, C:N ratio, silt and CaCO3 contents. The PTF to predict N1 indicated that only 88% of the variability in k1 was accounted for by bulk density and clay and CaCO3 contents. Bulk density, CEC, and total-N accounted for 85% of the variability in Nr. Fifty eight percent of the variability in kr was accounted for by bulk density, CEC, and organic matter contents. The PTF's developed can be used to predict values for N1,, k1 and kr, thereby enabling modelling of the temporal variation in legume-N mineralization.

Many of the bulk soil properties had a significant influence on the VFP belonging to different size classes. An assessment of the influence of VFP on N-mineralization parameters indicated that VFP had a significant influence on N1, k1, and Nr, but not on kr. The stepwise variable selection analysis indicated that 78% of the variability in N1 was accounted for by VFP with effective diameters in the size range 5 to 10 m and those 1.5 m. Values of k1 were also influenced by size classes 1.5 and 1.5 to 3 m. Values of Nr were less strongly influenced by pore characteristics.

An assessment of the combined influence of bulk soil properties and pore characteristics on N1 and k1 indicated that these two parameters are influenced only by pore characteristics whereas Nr was influenced both by pore characteristics and a bulk soil property, ie pH. On the other hand, values of kr and ko were influenced only by bulk soil properties.

A comparison of the R2 values of the PTF's obtained for N1, Nr, k1, and kr as functions of VFP and bulk soil properties with corresponding R2 values of the PTF's obtained using the information only on bulk soil properties indicated the R2 values for latter PTF's were greater than the former. Thus, for consistency, it is recommended that PTF's obtained using the information only on bulk soil properties be used to predict values for N-mineralization parameters. Further, the information on bulk soil properties is readily available in soil survey data. Analyses of pore characteristics are more useful in understanding the mechanisms whereby bulk soil properties influence mineralization dynamics.

The amount of N mineralized at any time in the red clover added treatment could be predicted from the bulk soil properties using the two pool model but could not be predicted using simple or multiple correlations between the amount of N mineralized and soil properties. The two pool model and associated pedotransfer functions are, therefore, necessary to provide predictions on mineralization dynamics. These models will be particularly useful in assessing the risk of legume-N leaching to water bodies across a range of soil textures.

 

TABLE OF CONTENTS

EXECUTIVE SUMMARY

1.0 INTRODUCTION

2.0 MATERIALS AND METHODS

2.1   Site Description
2.2   Soil Sample Collection and Processing
2.3   Soil Characterization
2.4   Bulk density adjustments
2.5   Characteristics of the red clover material
2.6   Establishment of water retention curves
2.7   Computation of volume fraction of pores
2.8   The relation between bulk soil properties
2.9   Incubation Experiments
2.10 Statistical Analyses

3.0 RESULTS AND DISCUSSION

3.1   N-mineralization
3.2   Fitting N-mineralization data to first order rate equations
3.3   Estimates of N-mineralization parameters
3.4   Interrelation between the N-mineralization parameters
3.5   Influence of soil properties on N-mineralization parameters
3.6   Influence of pore size characteristics on N mineralization

4.0 CONCLUSIONS

5.0 REFERENCES

 


 

Report List

Green Plan Research Home | Canada-Ontario GREEN PLAN

 

Last Updated: May 17, 2011 11:00:07 AM