Skip to main content
SLU publication database (SLUpub)

Other publication2013

Occurrence of power law tailings in inert tracer breakthrough curves through undisturbed soil

Koestel, Johannes

Abstract

Breakthrough curves from inert solute transport experiments in soils often exhibit power law tailings. Understanding how onset, duration and inclination of log-log linear slopes in breakthrough curve recessions are related to flow field properties and boundary conditions will improve predictions of contaminant displacement in the subsurface. The study of Willmann et al. (2008) [1] investigates exactly this for aquifers. However, flow field properties are difficult to measure, at least at large scales for which predictions of contaminant displacement are needed. It is therefore of interest to examine if soil properties and site factors can be used as proxy variables to directly infer to the occurrence and characteristics of log-log linear slopes in breakthrough curve tailings. In this study I investigated 82 breakthrough curves from inert tracer experiments conducted on undisturbed soil columns under steady state irrigation. The breakthrough curves were collated from the peer-reviewed literature.

A preliminary evaluation of the dataset yielded that half of the investigated breakthrough curves exhibited log-log linear slopes. The inclination of the slopes ranged from -1.3 to -5. As also reported by Willmann et al. (2008), I found that the inclination of the log-log linear slope was correlated with breakthrough curve connectivity indicators like small relative 5%-arrival times (p0.05, Figure 1; see [2] and [3]). Note that p0.05 is the inverse of what is denoted as CT1 in Knudby and Carrera (2005) [2]. Figure 1: The duration and inclination of log-log linear slopes in breakthrough curve tailings are shown together with the corresponding relative 5%-arrival time, p0.05. Reduced inclinations of log-log linear slopes were promoted by initially dry conditions, near saturated conditions as well as efforts to reduce air entrapment. The use of anionic tracers was related with steeper log-log linear slopes. The onset of the log-log linear slopes was not correlated with any soil property or experimental condition. The duration of the log-log linear slope was decreased for soil samples with a large bulk density and for experiments with a seepage face at the bottom of the column. For the future, I plan to increase the number of experiments in the database and to deepen the statistical evaluation.

[1] Willmann, M., Carrera, J. and Sanchez-Vila, X., Transport up-scaling in heterogeneous aquifers: What physical parameters control memory functions?, Water Resources Research, Dec. 44(12), - (2008). [2] Knudby C. and Carrera J., On the relationship between indicators of geostatistical, flow and transport connectivity, Advances in Water Resources, 28(4), 405-421 (2005). [3] Koestel J., Moeys J. and Jarvis N., Meta-analysis of the effects of soil properties, site factors and experimental conditions on solute transport, Hydrol Earth Syst Sci., 16(6), 1647-1665 (2012).