Water and soil redistribution in a cultivated Saskatchewan landscape
The concept of topographically controlled moisture redistribution underlies the conceptual model of soil development used by most researchers in the semi-arid northern Great Plains. The first part of this study explores the variability in the process of soil moisture redistribution, and its spatial distribution as it may be linked to topographic or pedogenic attributes. Recent advances in tillage research have shown that within cultivated landscapes tillage-induced redistribution of soils is the dominant erosion mechanism. The second part of this study attempts to validate the new model of tillage redistribution under Saskatchewan conditions. The redistribution of soil moisture was monitored on three occasions between June 1997 and October 1998 using the redistribution of a chloride tracer as a surrogate for moisture redistribution. The results of the chloride tracer provided clear evidence for the direction and magnitude of water flow during the study period. The results confirm the distinction in the moisture redistribution process between depression-centred soils and upland soils. The primary control on the movement of the tracer at depression-centred soils was spring flooding in early 1998. Subsequent redistribution of the remaining tracer was dominated by discharge phenomena. Subsurface flow in upland soils is clearly anisotropic, with observed differences in the degree of lateral and vertical redistribution of the chloride mass between landform elements and between soil profile classes. Two tillage experiments were conducted to evaluate the tillage process under Saskatchewan field conditions. The results of the tillage experiments were compared with the medium to long-term soil redistribution history as derived from cesium-137 redistribution. The cesium-137 results clearly show that the field-scale pattern of erosion at this Saskatchewan site corresponds to the characteristic pattern for tillage-induced erosion. Experimentally determined erosion values largely match or exceed those derived from the cesium-137 redistribution.Cesium-137 derived erosion rates for divergent shoulders, divergent backslopes and convergent backslopes were 42.3, 32.4, and 10.8 Mg ha-1 y-l, respectively. Experimentally determined values for net tillage erosion in these landform elements using a surface applied granular tracer and the median slope gradient for each category were 44, 74, and 86 Mg ha-1 y-l, respectively. Net erosion values determined using an aqueous tracer applied to the soil surface had values of 16, 28, and 32 Mg ha-1 y-l for divergent shoulders, divergent backslopes and convergent backslopes respectively. The comparatively low net erosion value for the divergent shoulder landform elements in the second experiment reflects the lack of a slope curvature factor in the erosion calculation. These results confirm that the redistribution of water and the redistribution of soil materials by tillage are the basic controls on the distribution of soil taxa in this landscape.
Doctor of Philosophy (Ph.D.)