SR9084 - Quantifying Preferential Flow and Recharge at the Field Scale: The First Step in Characterizing the Nature and Timing of Pathogen Transport to Groundwater

Researcher:

Dr. Gary Parkin, Dept. of Land Resource Science, University of Guelph

Objectives:

1. To measure the relative importance of preferential and gradual contaminant transport pathways through the unsaturated zone to the water table below tile-drained farm fields under conventional and no-till management practices on an annual basis.
   

2. To measure the amount of tile flow and groundwater recharge between tile drains under conventional and no-till management practices.
   

3. To determine sampler placement location to best measure average field-scale recharge on a sloping landscape.

Expected Benefits:

1. Will help to address the question of which season does preferential flow (and hence bacteria movement to groundwater) occur most often during the year.
   

Summary of Research Results:

In experiment I continuous measurements of soil water storage in no-till (NT) and "conventional" tillage by fall ploughing (CT) were used to detect bypass flow of infiltrating soil water below 40 cm depth. The data suggests that preferential flow is more common in the CT as opposed to NT plots especially during the summer season. Although this finding is contrary to the presumption that NT results in more macropores and hence more preferential flow, it agrees with the findings of another study conducted in Huron County in 1992 that also concluded that ploughed soil contains more macropores.

Experiment II has demonstrated the feasibility of a technique traditionally used to estimate the contribution of groundwater and event water to stream flow during a rainstorm for use on tile drains. Results indicate that for a rain storm in August 2003, the amount of event water (water that enters the tile drain without mixing with pre-existing soil water or groundwater), represented a maximum of 17% of the total tile flow suggesting that preferential flow to the tile drain during this rain event accounted for a maximum of 17% of the total tile flow.

Experiment III addressed objective 2 using a computer model (DRAINMOD) to estimate the amounts of tile drainage and deep drainage (groundwater recharge) beneath NT and CT plots at ERF. The only difference between the NT and CT model inputs was that the maximum water storage was increased by about 50 mm for the NT plots to reflect the measured increase in storage. The surplus water is split between tile drainage and deep drainage (groundwater recharge) by DRAINMOD as no runoff was predicted to occur in years 2001-2003. The model estimated slightly higher values of tile and deep drainage for CT versus NT.

Experiment IV addressed objective 3 and also contributed to addressing objective 1 using two geophysical techniques and pan lysimeter data. The pan lysimeters installed at different locations along a sloping field did not show a clear trend in drainage amount with slope position. Our original hypothesis that the average amount of drainage would occur at the mid-slope position is not acceptable according to data from this study. The geophysical methods have demonstrated their potential value in determining zones of higher drainage.

Recommendations based on this study include:

1. Physical methods (resistivity and EM 31) could be used to detect zones of relatively high drainage for sampler placement for drainage or leaching studies.
   

2. To measure the average amount of drainage or leaching along a sloping landscape, samplers should be installed at a regular interval (with samplers as closely spaced as funds allow) along the entire length of the landscape due to significant spatial variability.