|Publication type:||Doctoral thesis|
|Title:||Analysis of flow patterns : the influence of soil compaction and soil structure on the infiltration pathways of dye tracer solutions and the quantitative evaluation of flow patterns|
|Advisors / Reviewers:||Flühler, Hannes|
|Publisher / Ed. Institution:||ETH Zürich|
|Publisher / Ed. Institution:||Zürich|
|Subjects:||Image analysis; Soil compaction; Preferential flow|
|Subject (DDC):||630: Agriculture|
|Abstract:||Infiltration patterns visualized by means of dye tracers provide information about the flow regime in a given soil under given boundary and initial conditions. The tracer distribution in the soil is affected by the soil structure and in particular the sequence of soil horizons. Soil compaction may remarkably change soil structure and the continuity of the pore network. The goal of this thesis was (i) to find out how soil compaction affects water infiltration patterns, (ii) to link soil properties, in particular its layering structure, with types of flow patterns, and (iii) to develop a method for a quantitative analysis and comparison of flow patterns. Compaction modifies the pore system, often by reducing the pore volume or by destroying the structure. As a consequence, not only pre-consolidation load, bulk density or porosity are changed, but also the transport properties of the soil. The effects of heavy vehicle traffic on the infiltration patterns of dye tracer solutions were studied at different sites under vehicles equipped with caterpillar tracks and vehicles on wheels. The influences of soil compaction made visible by the tracer infiltration images were then compared with effects found by laboratory measurements (bulk density, porosity and pre-consolidation load). The results show that in most of the cases there is a good agreement between effects found on the basis of soil mechanical measurements and flow patterns. The pre-consolidation load was found to be a good indicator to predict soil stability and the bulk density was sensitive for detecting compaction effects. The experiment showed that soil mechanical measurements may provide information about how compaction propagates with depth whereas infiltration patterns showed the changes of the water infiltration pathways caused by soil compaction. At one site, which was trafficked with a sugarbeet harvester, soil compaction led to an increased preferential flow regime. After compaction, the fine pore structures in the topsoil were locked up and no longer available for water infiltration. The water was ponding on the surface and preferentially entering open worm burrows, bypassing the main root zone. Most of the tracer experiments were conducted with Brilliant Blue as a tracer and by comparing the flow patterns of trafficked with non-trafficked plots. However, the spatially variable soil structure makes it difficult to compare the flow patterns of neighbouring plots quantitatively. Comparing the flow patterns before and after compaction at one and the same plot overcomes this basic experimental difficulty of discriminating treatment effects from between-plot variability. Therefore, an experiment was carried out with two fluorescent tracers applied onto the same plot, one before and the other after controlled compaction with a heavy vehicle. The distributions of the two dyes in the excavated soil profiles were photographed separately using a digital camera and a light source, both equipped with tracer-specific filters. The superposition of the two flow patterns shows the effect of the compaction treatment directly. This method is more sensitive than experiments using a single tracer and the quantification of the differences is easier, because the flow patterns can be compared one-to-one. However, it is technically less demanding and less costly to use the single tracer technique. This technique requires the analysis of numerous images of patchy stained profiles. To compare such images objectively is one major focus of this thesis. Therefore, we wanted to develop a quantitative method to analyze the spatial distribution of the stained areas in vertical profiles and to link the characteristics of these patterns with soil properties, in particular its structure. Since soil layers strongly affect the infiltration patterns we first divided the flow patterns into layers of similar patterns. All layers found at several sites were then partitioned into groups of similar layers by hierarchical clustering. This classification reliably distinguishes between a homogeneous infiltration and preferential flow regime, but also between zones of pronounced preferential flow and zones of lateral spreading, e.g. sand or gravel lenses. The dye coverage and the mean width of the stained structures were the most indicative characteristics for the different clusters. We found good agreement between the sequences of layers found in the flow patterns and the soil horizons.|
|Further description:||Diss., Naturwissenschaften ETH Zürich, Nr. 14658, 2002.|
|License (according to publishing contract):||Licence according to publishing contract|
|Departement:||Life Sciences and Facility Management|
|Organisational Unit:||Institute of Natural Resource Sciences (IUNR)|
|Appears in collections:||Publikationen Life Sciences und Facility Management|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.