Limnol. Oceanogr., 44(6), 1999, 1575–1582

This note describes how a submarine, the F.A. Forel, carrying a vertical array of high-resolution temperature sensors, was used along with conventional measurements from a lowered conductivity-temperature-depth probe (CTD) to make novel measurements of the temperature field in Lake Geneva during summertime conditions of stable stratification and during winter convection. The submarine speed was about 0.5 m s21. In addition to the temperatures, the pressure, orientation, and tilt were recorded at frequencies of at least 10 Hz. Observations were made on a vertical scale of 0.1 to 2.5 m and on a horizontal scale from 0.5 m to 1 km. Examples of the data are presented. During the summer, evidence was found of internal waves and of extensive layers of low vertical temperature gradient, with vertical and horizontal scales of 0.5 m and 0.5 km, respectively; within this gradient, the temperature changed monotonically in the horizontal. During periods favoring convection, in the winter, when air temperatures were about 78C below the surface-water temperature, convectively unstable regions, typically of 5-m horizontal scale, were observed in the mixed layer. These appeared to be convective plumes. These winter measurements also included observations of a layer of cold water that was adjacent to the sloping boundary of the lake. This was identified as being a plume of dense cold water with thickness on the order of 10 m, which was driven by surface cooling, and consequent more rapid temperature decrease, in the shallow nearshore water. On meeting the thermocline at a depth of about 100 m, this plume spread horizontally and formed an intrusion some 30 m thick. A few accounts describing the use of submarines in making physical measurements in the ocean have been published. These include that of Osborn and Lueck (1985), who made measurements of turbulence in the ocean from the USS Dolphin submarine, that of Osborn et al. (1992), who also made acoustic measurements of subsurface bubble clouds from the USS Dolphin, that of Gargett (1982), who measured turbulence from a Pisces submersible, and those of Wadhams (1978) and Wadhams et al. (1979), who made measurements from a submarine under ice. There are, so far as we know, no accounts of comparable observations made from manned submarines in lakes. Here we describe measurements of the thermal structure of the upper layers of Lake Geneva made from a small submarine, F.A. Forel; these measurements illustrate the value of submarine observations in lakes. The objective of these submarine dives was to investigate fine-scale spatial distribution and variability of the temperature structure in the lake. During summer, measurements were made in the thermocline over the sloping sides of the lake as part of an investigation into the effect of internal waves in boundary mixing. During winter, we observed the structure of cold convective plumes that form in the near-surface layer and those that descend the sloping boundaries as gravity currents as a consequence of cooling in shallow water around the lake boundaries. Observations were made at 0.1 to 2.5 m on a vertical scale and at 0.05 m to 1.0 km on horizontal scales, these being the observations most readily made from the submarine. While the submarine does not provide a unique method of measuring the temperature structure described here, it is a fairly stable platform to use to make measurements that would be more difficult to make from the surface (e.g., by towed arrays). In addition, the submarine provides direct observations of the motions made visible by suspended sediment or passive organisms, and it is a useful facility to use to examine bottom-mounted or moored instruments in situ in order to ensure that they are properly and freely deployed. The lake, the submarine, and the sensors—The observations described here were made in the vicinity of Ouchy (468329N, 68139E), on the northern shore of Lake Geneva, where the shoreline lies in a 1108 direction. The lake here is about 15 km wide and, after a region some 100 m wide close to the shore (where the depth increases gradually to about 4 m), the sides slope down at some 108 to a maximum lake depth of 309 m (see Fig. 4a later). Depths of 100 m are reached at about 0.7 km from shore. The slopes are composed of soft, fine sediment and are irregular—incised by channels a few meters in depth that run mainly downslope— and are sometimes observed to have steep (.308) sides. The F.A. Forel, sketched in Fig. 1, is 2.2 m wide and 7.55 m in overall length and is 2.25 m high from the skids to the top of the bars protecting the upward-looking observation window. It has a displacement of 11 tonnes. It can accommodate three people, including the pilot. There is a second observation window in the bow that faces downward and forward. Visibility in the lake is typically about 2–3 m. The submarine carries exterior lights, TV, a mechanical arm for sampling, a pressure gauge, and a recording echosounder. There are 100 connections through the hull that are made up to suit users. Power is available at 24 v, and the normal electrical capacity available for scientific use is 24 Ahr. The submarine is connected by an underwater radio link to a 12m mother vessel, Black Prince, which carries a global positioning system (GPS). The submarine tracks over the ground are recovered and plotted after dives, giving the submarine’s horizontal position to within about 2 m and its speed over the ground, typically 0.5 6 0.05 m21, to within 0.5 cm s21. It has a normal operational capacity of 8 h and can work to a depth of 500 m, although our observations were limited to the upper 90 m of the lake. Navigation just above the bottom, following isobaths along the sloping boundaries of the lake, is difficult because of the channels and poor visibility, and in this mode of operation, the submarine is prone to make occasional contact with the soft sediment, which is a hazard to sensors carried ahead of the submarine. The majority of our studies near the bottom were therefore restricted to measurements obtained during runs at constant depth, when we were either approaching or reced-