TECHNICAL NOTES: AN AUTOMATED SAMPLING SYSTEM FOR LARGE SOIL COLUMN TRANSPORT STUDIES B. L. Woodbury, S. D Comfort, W. L. Powers ABSTRACT. An automated column effluent sampling system for large unsaturated soil columns was constructed and evaluated. The automated collection system consist of a programmable three-solenoid valve system that maintains a constant matric potential at the bottom of the soil column and periodically switches column effluent from a waste receptacle to a sampling vial. We compared the automated system to a traditional vacuum chamber-fraction collector system that continuously sampled all column effluent. Comparisons were made by measuring ^H20 breakthrough curves (BTCs) generated by the two collection systems using the same intact soil columns. No significant differences were observed in ^H20 BTCs between collection systems. These results indicate that the automated collection system is effective for large column transport studies and provides additional advantages by maintaining effluent resolution (sample volume), yet reduces sample size (numbers) and labor required for sample collection. Keywords. Fate and transport, Solute transport, Miscible displacement. D etermining the potential for pesticides to contaminate groundwater requires an understanding of solute transport mechanisms that occur in the field and the ability to accurately represent these processes in transport models. While some biological and chemical processes influencing solute transport can be studied in the laboratory using repacked soil columns, care must be taken when extrapolating these results to the field. Solute transport models based on the convection-dispersion equation require accurate estimates of dispersion coefficients, which can only be realistically obtained by measuring solute movement in the field or through intact soil columns. Numerous researchers have conducted field-scale transport and fate studies (i.e.. Butters et al., 1989; Comfort et al., 1993; Flury, 1996), but these studies are usually characterized as expensive and labor intensive with results often complicated by climatic conditions. Several transport studies have used undisturbed soil cores as an alternative to field studies (White et al., 1986; Seyfried and Rao, 1987; Singh and Kanwar, 1991; Gaber et al., 1995), but experiments performed with large soil columns (i.e., 15 cm diameter or larger) have procedural challenges. Depending on the adsorption properties of the solute, several pore volumes may need to be added and collected (via a fraction collector, porous cup or wick sampler) before breakthrough curves are observed. Using the traditional approach of Article was submitted for publication in April 1996; reviewed and approved for publication by the Soil and Water Div. of ASAE in August 1996. Contribution from the Agricultural Research Division, University of Nebraska, Lincoln, Nebr. Journal Paper No. 11468 . The authors are Bryan L. Woodbury, Graduate Student, Dept. of Civil Engineering, University of Nebraska, Lincoln; Steve D. Comfort, Assistant Professor of Soil Chemistry, University of Nebraska, Lincoln; and William L. Powers, Professor of Soil Physics, Dept. of Agronomy, University of Nebraska, Lincoln. Corresponding author: William Powers, Dept. of Agronomy, University of Nebraska, Lincoln, NE 68583- 0915; e-mail: <bpowers@unlinfo.unl.edu> placing soil columns atop a vacuum chamber and collecting effluent with a fraction collector can result in numerous samples and frequent interruptions in matric potential (vacuum) while sample vials are being replaced. In this study, we constructed and evaluated an automated sampling system (automated system) designed for large soil columns. This system maintains a constant matric potential at the bottom of the soil core, periodically diverts column effluent from a waste receptacle to a sampling vial, and eliminates the need for a vacuum chamber. Using the same intact soil columns, the automated system was tested against the more traditional vacuum chamber-fraction collector system (vacuum chamber system) by comparing the solute breakthrough curves (BTCs) generated by the two collection systems. MATERIALS AND METHODS SOIL COLUMN PREPARATION Collection systems were compared using intact soil columns (30-cm length; 15-cm diameter) of a Sharpsburg silty clay loam (Typic Argiudoll) obtained from the University of Nebraska Research and Development Center, near Mead, Nebraska. The soil columns were acquired by attaching a beveled steel bit to the end of polyvinyl chloride (PVC) pipe and driving it into the soil surface (Ap and Btl horizons) with a Giddings probe truck. Columns showing any observable signs of compaction (surface level inside the column vs. outside column) were discarded. Uncompacted soil cores were brought to the laboratory and prepared for solute transport studies. This was accomplished by cutting soil cores into 30-cm segments with a miter-box guided saw. The soil cores were secured at both ends with customized acrylic caps (table 1). Materials, vendors, and approximate unit costs for each item purchased for construction of soil columns and collection systems are provided (table 1). The bottom end caps supported a porous stainless steel plate (bubbling pressure « 24.5 kPa, table 1). A silicone sealant was placed VOL. 39(6):2163-2166 Transactions of the AS AE © 1996 American Society of Agricultural Engineers 0001-2351 / 96 / 3906-2163 2163