Methods comparison and spatial analysis of field parameter quantification of unsaturated flow at an existing wastewater infiltration system : Mines Park, Golden, Colorado

Subsurface water reclamation systems (SWRS) in mountain watersheds commonly experience issues of water sustainability, and are generally designed through judgment and experience rather than through rigorous quantitative methods. I have made a detailed vadose-zone characterization of several saturated and unsaturated flow parameters in an excavated trench representative of the size of a typical domestic drainage field at a wastewater reclamation test site (WRTS) at Mines Park, Golden, CO with the objective to improve the pre-design techniques associated with SWRS. Most wastewater infiltration is an unsaturated water flow process; therefore, measurement of soil-hydraulic parameters, especially saturated and unsaturated hydraulic conductivity (Ksat and Kunsat), is required. Toward this end, I compared the consistency of estimating or measuring Ksat and Kunsat through field, laboratory, and empirical methods. Field data collection included Guelph Permeameter and Tension-disc Infiltrometer measurements in the excavated trench, and Slug Test measurements at nearby monitoring wells. Representative soil samples were collected for in-lab experimentation at the UT Bureau of Economic Geology, including the van Genuchten-Mualem transformations of capillary-pressure curves determined from hanging-column measurements. Additional estimates of Ksat and Kunsat were determined through nine empirical functions and the Rosetta neural network found within Hydrus-1D software. Spatial variations in Ksat and Kunsat determined from the various methods are illustrated by curves that plot Ksat or Kunsat according to location. Statistical analyses demonstrate such a large degree of spatial variability that no single measurement adequately represents the values of these parameters, whether along the excavated trench or between the upper and lower tiers of the trench. The following main conclusions are offered: (1) spatial variations in Ksat and Kunsat occur along a trench and with depth, in which Kunsat exhibits much larger variability with depth than does Ksat; (2) of the empirical functions, results of the Hazen (1893), Harleman et al. (1963), and Beyer (1964) methods compare most favorably to both field methods (Guelph Permeameter, Tension-disc Infiltrometer), providing data that agree with the field-measured values within a factor of two; (3) the Guelph Permeameter field method is the least time consuming and less complicated, but the most arduous to conduct of all field methods; (4) Rosetta estimates with either sand/silt/clay input or soil classification input data provide Kunsat values most similar to the Hanging Column values using the van Genuchten-Mualem relationship; and (5) a single measurement did not appear to represent the hydraulic properties of an entire area occupied by a typical wastewater infiltration trench. I found variation both horizontally along the length of each tier and vertically between Tier 1 (2 feet below ground surface) and Tier 2 (4 feet bgs). This research advances the use of rigorous quantitative measurements and methods in understanding the design of SWRS