Key points are not available for this paper at this time.
Summary Hot springs in continental geothermal areas, thermal springs which are not associated with known high-temperature areas, and hot springs associated with hydrothermal circulation on ocean ridge crests are described by models based on fluid flow in fractures. A steady state model shows that fractures of the order of a few millimetres wide can carry a substantial convective flow and that the convective flow depends upon the third power of the fracture width. The steady state model also furnishes estimates of conductive temperature losses in springs and gives estimates of the depths of circulation for thermal springs in the southeastern United States which are in good agreement with available field data. In many cases the temperature and flow rate of springs is non-stationary. This is particularly true of hot springs in high temperature geothermal areas and it is expected to be true of springs on mid-ocean ridge crests. Time-dependent models for springs show that the main effect of the circulation is to lower the regional geothermal gradient. Non-stationary convection controlled by fractures can explain the variability of heat flow data obtained near ocean ridge crests. A numerical example shows that convection in a block 3 km wide, containing fractures 3 mm wide and 5 km deep, and circulating for 104 years gives rise to a hot spring with a temperature of 125 °C and a flow rate of 0.14 kg m−1 s−1. Such a spring discharging at the sea floor would give rise to an unmeasurably small temperature anomaly in the sea water. The convective heat transfer due to such a circulation system is roughly 200 times greater than the heat transfer that would have been achieved by conduction alone.
R. P. Lowell (Sat,) studied this question.