Limnol. Oceanogr., 44(7), 1999, 1802–1809

We have used new gel pore-water profilers and conventional seepage meters to determine the advective flux of water from an underwater saline seep into Lake Kinneret. The gel probes sampled pore waters from medium to coarse sands that could not be sampled by conventional coring methods. The anions Cl, Br, and SO4 were constant at levels just above those for the lake for 3–5 cm into the sediment due to wave action or other turbulent mixing processes. There was then a sharp increase in concentration to values of approximately 8,000 mg Cl liter21, 370 mg SO4 liter21, and 120 mg Br liter21 at a depth of ;8.5 cm. Using an advection–diffusion model, the linear interstitial advection velocity (LIV) of the groundwater into the lake was calculated to vary between 140 and 275 cm yr21. The LIV values from conventional seepage flux meters at the same site were 30 and 164 cm yr21. Differences between the LIV measurements of these two methods may be due to a number of possible factors, including groundwater flux heterogeneity. Groundwater seepage directly into lakes can be an important process both as it affects the overall water budget and in many situations the water quality of the lake (see Boyle [1994] and references therein). The flux of saline water from seepage into a lake can be estimated from detailed pore-water chemical profiles using an advection–diffusion model (Munk 1966; Berner 1972; Lee et al. 1980; Cornett et al. 1989). A requirement of such studies is that it is possible to determine a pore-water profile of sufficient resolution to resolve different solutions of the model. In this study, we have used a recently developed technique employing gel sampler probes (Krom et al. 1994; Mortimer et al. 1998). Previous attempts to sample pore waters at these sublittoral sites have not been successful because conventional sediment corers, whether deployed from a boat or by a diver, tend to bounce off the sandy substrate or the corers do not hold the sediment. We have measured high-resolution porewater profiles of chloride, bromide, and sulfate at three locations in the area of the saline seeps offshore from the Tiberias hot springs and then modeled the resulting profiles to estimate the flux of saline water into Lake Kinneret. The seepage rate of groundwater into a lake can also be measured directly using seepage meters. Several designs have been proposed (e.g., Lee 1977; Cherkauer and McBride 1988; Boyle 1994). It has been shown that the most reliable and cost-effective seepage meter design employs an inverted container with a seepage bag mounted on its top to measure water displacement over a specified area of the lake bottom (Lee 1977; Lee and Cherry 1978; Boyle 1994). In the Kinneret study we have used a modified seepage meter designed by Boyle (1994) to measure the groundwater seepage rate directly. The seepage rates measured by the seepage meter were compared with flux rates estimated from modeling the pore-water profiles measured using the gel probes. The primary purpose of this research was to make an initial assessment of the compatibility, reproducibility, and installation logistics of these two methods with the aim of providing better estimates of littoral groundwater discharge into lakes. Lake Kinneret serves as a major source (;30%) of drinking and irrigation water to Israel. The principal source of freshwater into Lake Kinneret is the inflowing Jordan River with a chlorinity of 17 mg Cl liter21. There are a number of onshore saline springs that discharged into the lake and were diverted in 1964 into a channel. As a result, the salinity of the lake decreased from 365 mg Cl liter21 in 1964 to its present level of 225 6 30 mg Cl liter21 (Nishri et al. 1999). The present salinity balance indicates that there are substantial internal salinity sources to the lake estimated to be 90,000 tones of chloride per year (Smith et al. 1989). A number of point sources of saline groundwater have been identified on the lake bottom (e.g., Tabgha and Fulya areas). However, there have been no studies that quantify the flux of saline water from these subaqueous springs. Recently, an area of saline seeps has been located offshore of the Tiberias Hot Springs (THS) (Manwaring 1996); springs that have been used for therapeutic and recreational purposes since at least Roman times. It was an aim of this study to determine the flux of saline water from these potentially important seepage areas into the lake. Protocol—Pore waters at three sites were sampled using gel probes. The sites were situated at 30 m (Sta. 30), 35 m (Sta. 35), and 37 m (Sta. 37) offshore from the THS in water depths of 4.5, 5.0, and 5.5 m, respectively. The sampling was carried out on 20–21 November 1996. Seepage flux meters were deployed at the first two of these sites, 30 m and 35 m from the 23rd of October 1996 for 1 week. In addition, a sample of water from the THS was collected for chemical comparisons with the pore-water samples. Gel sampling of the pore waters was performed using the procedure described by Krom et al. (1994). Probe lengths used in this investigation were 40 cm; but it was only possible to insert the probes ;20 cm into the sand because a hard layer was encountered at this depth. The polyacrylamide gels were prepared in Leeds, UK, transported to Israel in watertight plastic containers with Milli-Q water, and assembled a few days before use. The probes were placed directly into the sediment by divers, left to equilibrate overnight, and recovered the next day. The gel was divided immediately after recovery into narrow sections (;0.5 cm), put into tightly closed microcentrifuge tubes, and weighed. Analysis was performed within 1 week using a Dionex ion chromatograph with an autosampler attached. Replicate analysis of known standards was 1.3% (Cl), 2.9% (Br), and 4.6% (SO4; relative standard deviation,