Using New Theory and Experimental Methods to Understand the Relative Controls of Storage, Antecedent Conditions and Precipitation Intensity on Transit Time Distributions through a Sloping Soil Lysimeter

Abstract:

Transit time distributions (TTD) are an efficient way of characterizing transport through the complex flow dynamics of a hydrologic system, and can serve as a basis for spatially-integrated solute transport modeling. Recently there has been progress in the development of a theory of time-variable TTDs that captures the effect of temporal variability in the timing of fluxes as well as changes in flow pathways. Furthermore, a new formulation of this theory allows the essential transport properties of a system to be parameterized by a physically meaningful time-variable probability distribution, the Ω function. This distribution determines how the age distribution of water in storage is sampled by the outflow. The form of the Ω function varies if the flow pathways change, but is not determined by the timing of fluxes (unlike the TTD). In this study, we use this theory to characterize transport by transient flows through a homogeneously packed 1 m³ sloping soil lysimeter. The transit time distribution associated with each of four irrigation periods (repeated daily for 24 days) are compared to examine the significance of changes in the Ω function due to variations in total storage, antecedent conditions, and precipitation intensity. We observe both the time-variable TTD and the Ω function experimentally by applying the PERTH method (Harman and Kim, 2014, GRL, 41, 1567-1575). The method allows us to observe multiple overlapping time-variable TTD in controlled experiments using only two conservative tracers. We hypothesize that both the TTD and the Ω function will vary in time, even in this small scale, because water will take different flow pathways depending on the initial state of the lysimeter and irrigation intensity. However, based on primarily modeling, we conjecture that major variability in the Ω function will be limited to a period during and immediately after each irrigation. We anticipate the Ω function is almost time-invariant (or scales simply with total storage) during the recession period because flow pathways are stable during this period. This is one of the first experimental studies of this type, and the results offer insights into solute transport in transient, variably-saturated systems.