Setback distances between septic tank systems and the shorelines of Lake Okareka, New Zealand were determined from model simulations for a worst-case scenario, using the highest hydraulic conductivity and gradient measured in the field, removal rates of the microbial indicators (Escherichia coli and F-RNA phages) determined from a column experiment, and maximum values of the design criteria for the disposal system, and assuming an absence of an unsaturated zone, a continuous discharge of the raw effluent from a failed or non-complying treatment system (both indicators at concentrations of 1×107 counts/100 ml) into the groundwater and no sorption of pathogens in the aquifer. Modelling results suggest that the minimal setback distances were 16 m to satisfy the New Zealand Recreational Water Quality Guidelines for E. coli <126 per 100 ml (Ministry for the Environment, 1999) and 48 m to meet the Drinking-Water Standards for New Zealand 2000 for enteric virus <1 per 100 l (Ministry of Health, 2000). These distances may be applicable for other lakeshores in pumice sand aquifers with groundwater velocities <7 m/day. Findings of laboratory column and batch experiments provided an insight into the microbial attenuation and transport processes in pumice sand aquifers. Bacterial removal was predominately through filtration (87–88%) and partially by die-off (12–13%), while viral removal was by both die-off (45%) and filtration (55%). In addition, microbial die-off in groundwater without aquifer material (i.e., free microbes) was much lower than die-off in groundwater with aquifer material (i.e., sorbed microbes) and contributed only 2–6% to the total removal. This implies that the setback distances estimated from die-off rates for the free microbes, determined in the laboratory without considering aquifer media and other removal processes, which are often reported in the literature, could be larger than necessary.

The transport and attenuation of cadmium (Cd) and rhodamine WT (RWT) in a pumice sand aquifer media was investigated using column experiments to study a scenario of point-source contamination. A pore-water velocity of 1.7-1.8 m/day, which is a typical field groundwater velocity in a pumice sand aquifer system, was applied to triplicate columns. A pulse of a solution containing Cd and RWT, together with the conservative tracer tritiated water ((H2O)-H-3) at pH = 7, was introduced into the columns. Experimental results showed that concentration breakthrough curves (BTCs) of (H2O)-H-3 were symmetrical and fitted well into an equilibrium model. In contrast, BTCs of Cd and RWT were asymmetrical with significant tailings and fitted well with a two-site adsorption/desorption model. The symmetric (H2O)-H-3 BTCs suggest that physical nonequilibrium was absent in the experimental system, therefore the asymmetrical BTCs of Cd and RWT were attributed to chemical non-equilibrium. Modelling results showed that, in comparison with (H2O)-H-3, Cd was apparently retarded by 101-108 times in pumice sand aquifer media (apparent adsorption coefficient 7.33-9.24 ml/g) and underwent a mass loss of 20-30% that was probably because of precipitation of CdCO3. As CdCO3 is extremely insoluble, Cd precipitation would be irreversible and therefore it would not contribute to the tailing of the Cd BTCs. The experimental results suggest that the adsorption and desorption of Cd in pumice sand aquifer media in hydrodynamic conditions was a kinetic process. Cd desorption rates were two orders-of-magnitude slower than its adsorption rates. This resulted in a prolonged mean residence time for Cd in pumice sand aquifer media, which was 10-12 days in the 18-cm-long columns under a flow velocity of 1.7-1.8 m/day. Since the mean residence time is only indicative for the arrival of the central of mass in a contaminant BTC, the time required for the total disappearance of Cd will be much longer than the mean residence time because of the significantly long tailing of the BTC. This implies that natural attenuation of Cd from a contaminated pumice sand aquifer would take a time period from decades to centuries. Batch isotherm experiments were also carried out to obtain Cd adsorption coefficients at equilibrium conditions. In comparison with the column results obtained from non-equilibrium conditions, adsorption coefficients of 20 ml/g obtained from the batch equilibrium experiments were 2-3 times higher for Cd. This finding suggests that care should be taken when using batch adsorption isotherm results to predict field problems because the delay in the appearance of contaminants in drinking water wells and springs could be overestimated. In comparison with (H2O)-H-3, RWT was retarded 5-7 times and had BTCs that were significantly more spread out. Hence the use of RWT to indicate groundwater flow in pumice sand aquifers will underestimate groundwater velocity and overestimate aquifer dispersivity. About 4-14% of RWT mass was lost during its transport, probably because of the irreversible adsorption of isomer 2 (one of two major components of RWT) dye onto the aquifer sand.

Three different techniques were evaluated for installing suction cups in an alluvial gravel vadose zone-a steel spike, a mechanical auger, and an air rotary drilling rig. Soil water samples were collected from suction cups monthly for three years. Nitrate and chloride concentrations in samples taken from the suction cups were similar to those measured from a linear lysimeter installed in the same field. The nitrate and chloride concentrations in the suction cup samples showed significant spatial and temporal variation consistent with leaching from random urine patches. Solute concentrations became less variable as sampling depth increased, so fewer suction cups would be required at greater depths to estimate mean concentrations of solutes leaching to groundwater. Recharge was estimated using a water balance model and combined with the concentration data to give solute fluxes or leaching losses. Averaging the fluxes from all of the suction cups indicates a fairly constant leaching loss of around 50 kg ha -1 yr -1 for nitrate-N (range = 38-58 kg ha -1 yr -1 ) and around 80kg ha -1 yr -1 for chloride (range = 79-89 kg ha -1 yr -1 ). The total cost for the construction and installation of the suction cups ranged from $290 for the 1.3 m depth spike installation to $620 for the 5 m depth suction cups installed using an air rotary drilling rig. The major difference in price between the three installation methods relates to their depth limitations and the higher cost of equipment that can install suction cups at greater depths. A significant limitation of the auger installation was the disturbance of the soil profile caused by the large diameter hole that was excavated. For the installation of suction cups in stony alluvial sub-soils, we recommend using spike installation for depths of less than 1.5m and air rotary drilling for depths of greater than 1.5 m.

A passive sampling system for use with rhodamine WT (RWT) in groundwater tracing experiments was developed to assist in the characterisation of groundwater flow paths. Amberlite XAD-7 resin was found to be suitable for adsorption of RWT, which can then be extracted using an ethanol/water mix and analysed fluorometrically. Batch and column experiments showed that XAD-7 resin has a high RWT capacity. The adsorption was slightly dependent on pH, but was always above 75% under batch conditions. The resin had a high percentage mass recovery at flow velocities around 1.5 m/day, but this decreased with increasing flow velocities. Desorption of RWT off the resin in water is dependent on the flow velocity of water and the time after the peak RWT has passed. The mass of RWT extracted from the resin bags correlated very well with both the RWT mass flux in the water and the peak concentrations observed in the monitoring wells in a field experiment. The results of resin bags were reproducible in the field with a mean coefficient of variation equal to 16%. This method has been successfully applied to two field situations with different flow velocities to indicate groundwater flow paths.

This paper investigates the effects of pore-water velocity on chemical nonequilibrium during transport of Cd, Zn, and Pb through alluvial gravel columns. Three pore-water velocities ranging from 3 to 60 m/day were applied to triplicate columns for each metal. Model results for the symmetric breakthrough curves (BTCs) of tritium (3H2O) data suggest that physical nonequilibrium components were absent in the uniformly packed columns used in these studies. As a result, values of pore-water velocity and dispersion coefficient were estimated from fitting 3H2O BTCs to an equilibrium model. The BTCs of metals display long tailing, indicating presence of chemical nonequilibrium in the system, which was further supported by the decreased metal concentrations during flow interruption. The BTCs of the metals were analysed using a two-site model, and transport parameters were derived using the CXTFIT curve-fitting program. The model results indicate that the partitioning coefficient (beta), forward rate (k1), and backward rate (k2) are positively correlated with pore-water velocity (V); while the retardation factor (R), mass transfer coefficient ((omega), and ratio of k1/k2 are inversely correlated with V. There is no apparent relationship between the fraction of exchange sites at equilibrium (f) and V. The influence of Von k2 is much greater than on R, beta, omega, and k1. A one-order-of-magnitude change in V would cause a two-order-of-magnitude change in k2 while resulting in only a one order-of-magnitude change in R, beta, omega, and k1. The forward rates for the metals are found to be two to three orders-of-magnitude greater than the corresponding backward rate. However, the difference between the two rates reduces with increasing pore-water velocity. Model results also suggest that Cd and Zn behave similarly, while Pb is much more strongly sorbed. At input concentrations of about 4 mg/l and pore-water velocities of 3-60 m/day in the groundwater within alluvial gravel, this study suggests retardation factors of 26-289 for Cd, 24-255 for Zn, and 322-6377 for Pb.