The extent of agricultural drainage has created concern for its potential undesirable
effects on surface water quality. Land applications of liquid manure on tile drain
fields have the potential to transport solutes and bacteria to the drains following
precipitation or irrigation events and many times are directly sent to a surface water
body, and have been documented as a source of contamination of surface waters.
This study determined the potential for and magnitude of E. coli and solute
migration to tile drains through the soil profile. Water from subsurface drains was
analyzed for chemical and bacterial composition following tracer applications.
Two sites were selected for the study to determine transport at large (field) and
small (plot) scales. At the large-scale site, both tracers, bacteria (E. coli and Total
Coliform) and Amino-G (a conservative tracer), were used to monitor the speed of
transport from the surface to the tile drain following liquid manure applications,
tracer applications and additionally precipitation events. The concentrations of E.
coli were monitored every hour for 76 days during the spring. Both tracers,
bacteria and Amino-G, were detected in the tile drainage shortly after precipitation
events. The peak concentration of E. coli was observed to be 1.2 x 10⁶
CFU/l00mL. These elevated concentrations of E. coli might be attributed to the
characteristics of the soil, high organic matter and well-structured clay soils. Both
the rapid breakthrough of tracer to the tile drain and the peaks of tile water
temperature during precipitation events provided evidence of macropore flow.
Antecedent soil moisture and warmer temperatures appeared to provide ideal
conditions for bacteria growth.
The small-scale study site was selected for a more focused study. The purpose of
this site was to quantify more accurately the percent mass of surface applied tracer
that was transported to the tile drain, allowing mass balance calculations.
Experiments were conducted during the summer to control the rate and total
amount of irrigation. Amino-G readings were taken every 10 seconds for 125
hours of continuous irrigation. Tracer applications were conducted at runoff and
non-runoff conditions. Both types of tracer applications had Amino-G
breakthrough in less than 10 minutes after initiation of irrigation. Tracer applied at
runoff rates resulted in 4 to 17 times more total tracer mass migrating to the tile
drain than when applied at non-runoff rates. The total mass of Amino-G migrating
to the tile drain during non-runoff conditions depended on the total volume of
applied tracer, regardless of the tracer concentration. For an application of 5.6 mm
at 12 mg/L, 5.7% of the total applied tracer migrated to the tile drain, whereas for
an application of 1.9 mm at 27.7 mg/L only 2.8% of the total applied tracer
migrated to the tile drain. Tile flow response to irrigation experiments appeared to
be governed by soil moisture. Lysimeter samples were taken continuously every 4-8 hours until the 94th hour after tracer application. Tile water concentrations were
consistently greater than concentrations found in the deeper suction lysimeters at
corresponding times, providing further evidence of preferential flow. E. coli
transported through the soil and into the drains were demonstrated to be event-driven
by precipitation events and irrigation events. In addition, the characteristics
of this type of soil - the high clay content, the well-defined structure, the high level
of organic matter and rich biological activity has been known to enhance the
preferential pathways and transport processes in the soil profile, resulting in rapid
transport of surface applied solutes and effluents to tile drains. / Graduation date: 2003
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/29676 |
| Date | 21 March 2002 |
| Creators | Moreno, Daniel |
| Contributors | Dragila, Maria I. |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0027 seconds