The effect of pH of the buffered saturating solution on the cation‐exchange capacity (CEC) of 60 Wisconsin soils was determined within the pH range 2.5 to 8.0. The relative contributions of clay and organic matter to total CEC were measured using multiple regression analyses; the independent variables were clay and organic matter contents and the dependent variables, CEC at pH 2.5, 3.5, 5.0, 6.0, 7.0, and 8.0. The average CEC of the organic matter at each pH value was 36, 73, 127, 131, 163, and 213 me. per 100 g., respectively; while that of the clay was 38, 46, 54, 56, 60, and 64 me. per 100 g. Within the limits of the standard errors of the regression coefficients, CEC of both clay and organic matter increased linearly with pH. The regression equations indicated that the mean relative contribution of organic matter to total soil CEC in this group of soils varied from 19% at pH 2.5 to 45% at pH 8.0; the mean organic matter and clay contents of the soils studied were 3.28% and 13.3% respectively. Coefficients of multiple correlation varied from 0.927 at pH 3.5 to 0.959 at pH 8.0.

Field experiments were conducted by using a tile drain monitoring facility to determine the impact of preferential flow on the transport of adsorbing and non‐adsorbing tracers. Simulated rainfall with 7.5 mm h−1 intensity and 7.5 h duration was applied to a 18‐ by 65‐m no‐till plot. After 72 min of irrigation, a pulse of Br− and rhodamine WT (water tracer) was applied through irrigation, and 4 h later, a second pulse of Cl− and rhodamine WT was applied. The breakthrough curves (BTC) of these tracers were measured by sampling the tile. The same experiments were repeated in an adjacent conventional‐till plot, except the rainfall intensity was reduced to 5 mm h−1 The results showed that both the non‐adsorbing and the adsorbing tracers applied in the same pulse arrived at the tile line at the same time and their BTC peaked at the same time. This suggested that water dynamics of preferential flow paths dominated the initial phase of the contaminant transport, regardless of the retardation properties of contaminants. The tracers from the second pulse were detected at only 13 min after application. Among the four tracer pulses in two plots, the BTC from the second pulse in the no‐till plot had the longest period in which the non‐adsorbing and adsorbing tracers had identical patterns. This indicated that the wetter the soil profile, the longer the water dynamics of preferential flow paths dominate the contaminant transport. The BTC from the second pulse applied to the two plots had identical arrival and peak times.

Field experiments were conducted on tile‐drained plots at the South East Purdue Agricultural Center in Butlerville, Indiana, to quantify contaminant transport via preferential flow paths in a silt loam soil. Breakthrough patterns of three fluorobenzoic acids (pentafluorobenzoic acid [PFBA], o‐trifluoromethylbenzoic acid [o‐TFMBA], and 2,6‐difluorobenzoic acid [2,6‐DFBA]) in a preliminary study indicated that they were transported as conservatively as is bromide (Br−). These four tracers were then sequentially applied, in an adjacent plot, during simulated precipitation (3 mm h−1 intensity, 10‐h duration). Bromide was sprayed shortly before irrigation started, while PFBA, o‐TFMBA, and 2,6‐DFBA were applied at 2, 4, and 6 h thereafter, respectively. Tile flow began increasing at around 3 h, and Br− appeared in tile drain flow ≈4 h after irrigation started, yet benzoic acids, PFBA, o‐TFMBA, and 2,6‐DFBA, were detected in the tile drainage at 102 min, 42 min, and 18 min after their applications, respectively. Tracer mass recovery from tile drainage was Br− (7.04%), PFBA (13.9%), o‐TFMBA, (18.7%), and 2,6‐DFBA (19.7%) of applied mass. The faster arrival time and greater recovery of sequentially applied tracers confirmed that water movement and contaminant transport shifts toward increasingly larger pores of the preferential flow paths as soil becomes wet during a precipitation event. The breakthrough patterns of these tracers can be used to quantify the water flux distributions of preferential paths. Because ≈90% of the chemical leached from this precipitation event occurred during the first day, it was critical to intensively monitor contaminant transport during the first 24 h after a rainfall. A soil sampling protocol based on collecting soil cores at random locations once every several days is unsuitable for determining the deep leaching under field conditions.

A field site was established at Beltsville, MD, in 1986 to assess the effect of conventional and no-till cultural practices on the movement of pesticides into shallow groundwater. Groundwater samples taken from unconflned (<1.5 m deep) and confined (<3 m deep) monitoring wells in 1986-1988 were analyzed for atrazine [6-chloro-A'-ethyl-A'-(l-methylethyl)-l,3,5-triazine-2,4-diamine], deethylatrazine [6-chloro-./V-( 1 -methylethyl)-!,3,5-triazine-2,4-diamine], alach-. lor [2-chloro-A'-(2,6-diethylphenyl)-A'-(methoxymethyl)acetamide], cyanazine [2-([4-chloro-6-(ethylamino)-l,3,5-triazine-2-yl]amino)-2methylpropanenitrile], and carbofuran (2,3-dihydro-2,2-dimethyl-7benzofuranyl methylcarbamate). Atrazine was found in groundwater all year, while cyanazine, alachlor, and carbofuran were present only for a short period (<3 mo) after pesticide application. Fairly constant background levels of <0.5 /tg L ' atrazine were found under fields treated before 1986, while levels under continuously treated fields were <2.0 /ig L" 1 for 22 of 25 samplings. Pesticide residues in unconfined groundwater were usually higher (ca. 2 to 4X) than in confined groundwater. Rainfall timing relative to pesticide application was critically important to pesticide leaching. A prolonged rain immediately after the 1988 application resulted in peak atrazine and cyanazine levels of ca. 200 fig L~' in unconfined and ca. 30 to 40 Mg L ' in confined groundwater, which resulted in short-term levels ca. 2 to 50 X greater under no-till than conventional till plots. Results of this study suggest that preferential transport occurred.

Both physical and biological processes have temporal and Structural pores associated with macropore-type preferential flow spatial patterns of formation and destruction cycles pathways can accelerate chemical transport in unsaturated soils, thereby potentially causing groundwater contamination. To predict chemical (Gupta et al., 2002). Tillage practices and compaction, transport through these pathways, classical deterministic models de-for example, often destroy the continuity of large strucpend on soil hydraulic conductivity, which effectively lumps flow contural pores (Isensee et al., 1990). The coefficient of varitributions from all individual pathways. We contend, however, that quanation of soil hydraulic conductivity, which is often dictifying the pore spectrum of preferential pathways, without lumping the tated by the soil structural pores, ranged from 100 to contributions of individual pores, is the appropriate method for simu-400% (Libardi et al., 1980; Warrick and Nielsen, 1980). lating convective chemical transport through macropore-type prefer-This suggests that the spatial variability of soil structural ential pathways. In this study, we conducted field-scale experiments pores measured by using core-or block-sized samples is by using an improved tile drain monitoring protocol to measure the very large. As a result, the size spectrum of large strucmass flux breakthrough patterns of conservative tracers. The tails of tural pores measured at several random locations by these patterns suggested that the impact of preferential pathways on contaminant transport can be conceptualized as that occurring through small sample sizes in a field may not represent the speccylindrical capillary tubes. We then proposed a distribution function trum of the entire field. Temporal extrapolation of meabracketed by sharp cutoff points to represent the pore spectrum of surements may be similarly invalid. For these reasons, these tubes. Finally, we used the measured tracer breakthrough curves it is difficult to directly measure the field-scale spectrum (BTCs) as data sources to find the parameters of the proposed funcof the large structural pores, yet this property is among tion. Our results, based on the best fitting, showed that the preferential the most important soil properties when dealing with pathways are naturally clustered into domains; preferential pathways issues related to water quality. with a wide range of pore radii could become active simultaneously Soil characteristic curves and hydraulic conductivity when infiltration rate increases. Because the derived pore spectra sicurves were used in past research from the 1950s and multaneously satisfy both water movement and solute transport, pore 1960s [summarized by Hillel (1980, p. 183-185) and Jury spectra can be used to (i) calculate soil hydraulic conductivity of preferential pathways in deterministic approaches, and (ii) construct multi-et al. (1991, p. 89-94)] to quantify the soil pore-size ple probability density functions (PDFs) for the transfer function ap-d...