Passive capillary wick samplers (PCAPs) are primarily used to sample water from the vadose zone. PCAPs use fiberglass wicks to form a hanging water column that exerts suction on the surrounding soil. Although PCAPs have been used to estimate soil water flux, the accuracy with which PCAPs can estimate flux comes into question due to over/undersampling caused by this applied flux. I used numerical models to explore the effects of a PCAP on flow through the vadose zone. Specifically, I used a two-dimensional axisymmetric flow model of a PCAP embedded in a medium based on HYDRUS. Both steady-state and transient conditions were simulated through the application of various precipitation rates and periods across several soil textures. In this study, I examine soil hydraulic properties, across the soil texture triangle, subject to a range of precipitation events. Results show that the PCAP does over/underestimate water flux. The degree of error is quantified by defining a capture efficiency, which is the ratio of the flux into the plate and the flux that would occur at the same depth with no PCAP present. Higher fluxes and longer time periods resulted in increased convergence of flux into the PCAP, while lower fluxes and shorter durations resulted in divergence of flux from the PCAP. The goal of the study is to understand the behavior of PCAPs under different conditions and to use that knowledge to interpret field measurements in the Jemez River Basin Critical Zone Observatory.

Dye tracer and time-lapse ground-penetrating radar (GPR) were used to image preferential flow paths in the shallow, unsaturated zone on hillslopes in two adjacent watersheds within the Susquehanna-Shale Hills Critical Zone Observatory (CZO). At each site we injected about 50 L of water mixed with brilliant blue dye (4 g/L) into a trench cut perpendicular to the slope (~1.0 m long by ~0.20 m wide by ~0.20 m deep) to create a line of infiltration. GPR (800 MHz antennae with constant offset) was used to monitor the movement of the dye tracer downslope on a 1.0 m x 2.0 m grid with a 0.05 m line spacing. The site was then excavated and the stained pathways photographed to document the dye movement. We saw a considerable difference in the pattern of shallow preferential flow between the two sites despite similar soil characteristics and slope position. Both sites showed dye penetrating down to saprolite (~0.40 m); however, lateral flow migration between the two sites was different. At the Missed Grouse field site, the lateral migration was ~0.55 m as an evenly dispersed plume, but at distance of 0.70 m a finger of dye was observed. At the Shale Hills field site, the total lateral flow was ~0.40 m, dye was barely visible until the excavation reached ~0.10 m, and there was more evidence of distinct fingering in the vertical direction. Based on laboratory and field experiments as well as processing of the radargrams, the following conclusions were drawn: 1) time-lapse GPR successfully delineated the extent of lateral flow, but the GPR resolution was insufficient to detect small fingers of dye; 2) there was not a distinct GPR reflection at the regolith-saprock boundary, but this interface could be estimated from the extent of signal attenuation; 3) the preliminary soil moisture conditions may explain differences in the extent of infiltration at the two sites; 4) rapid infiltration into the underlying saprock limited the extent of shallow lateral flow at both sites and can be seen using the mass balance calculation and the lateral extent of dye within the radargrams; and 5) variations in flow patterns were observed between sites with similar settings at Susquehanna-Shale Hills CZO. / Geology