The Chemical and Aqueous Transport (CAT) meter [Tryon et al., 2001] is designed to quantify both inflow and outflow rates on the order of 0.01 cm/yr to 100 m/yr. At high outflow rates, a time series record of the outflow fluid chemistry may also be obtained. These instruments have been in use since 1998 and have been very successful in monitoring long term fluid flow in both seep and non-seep environments. The CAT meter uses the dilution of a chemical tracer to measure flow through the outlet tubing exiting the top of a collection chamber. The pump contains two osmotic membranes that separate the chambers containing pure water from the saline side that is held at saturation levels by an excess of NaCl. Due to the constant gradient, distilled water is drawn from the fresh water chamber through the osmotic membrane into the saline chamber at a rate that is constant for a given temperature. The saline output side of the pump system is rigged to inject the tracer while the distilled input side of the two pumps are connected to separate sample coils into which they draw fluid from either side of the tracer injection point. Each sample coil is initially filled with deionized water. Having two sample coils allows both inflow and outflow to be measured. A unique pattern of chemical tracer distribution is recorded in the sample coils allowing a serial record of the flow rates to be determined. Upon recovery of the instruments the sample coils are subsampled at appropriate intervals and analyzed using a Perkin-Elmer Optima 3000XL ICP-OES. Both tracer concentration and major ion concentration (Na, Ca, Mg, S, K, Sr, B, Li) are determined simultaneously. A subset of these instruments are equipped with an auxillary osmotic pump connected to copper coils and high pressure valves so that they can be returned to the surface at ambient pressure, maintaining the gas composition of the fluids for analysis.

As explained in Tryon et al., 2001, diffusion in the sample coils is negligible. Typical sample sizes are 25-75 cm of tubing, many times the characteristic diffusion length for typical seawater ions at ocean bottom temperatures. Our data has shown that we typically achieve resolutions of ~0.5% of the deployment time in the latest portions of the record and ~2% in the oldest portion for deployments of a year. At this time we have no quantitative data with regards to long-term He diffusion in the copper coils, however, since the diffusion coefficient (D) of He is ~7x10-5, and of typical seawater ions is ~1-3x10-5, the characteristic diffusion length (µD1/2) for He should only be about twice that of the other species we analyse for (45 cm for a year at 25°C and much less at ocean bottom temperatures). Our copper coils are a smaller diameter than the plastic coils and thus our samples are typically twice the length, offsetting any difference in diffusion length.