Space filling and shape of fractures were investigated on rock sections of the Tamagawa Welded Tuffs distributed around Hachimantai, Northeast Japan. Various scales of natural fracture patterns which were obtained from 10m×10m, 5m×5m, 50cm×50cm and 1.6mm×1.6mm observation areas were used for fractal analysis. Space filling of natural fracture patterns in sections of the Tamagawa Welded Tuffs shows a statistical self-similarity, and its fractal dimension determined by a box-counting algorithm lied in the range from 1.08 to 1.48. Mean values of fractal dimension in various scales of fracture pattern of the Tamagawa Welded Tuffs were approximately equal. These facts suggest that the fractal dimension of space filling of the zoom sequences had an unique fractal property and the value of the fractal dimension was about 1.28. Fracture shape in the section which is sliced in random orientation also showed fractal property, and its fractal dimension was in the range between 1.01 and 1.05. The fractal observed in space filling and shape on natural rock suggests that the characteristic fractal dimensions could be estimated by fracture pattern of other scales. In order to compare the fractal dimensions of fracture space filling among various observation areas of different scales, it is necessary that the ratio of the sensitivity of measurement (lmin: the minimum length of between ends of fracture) to the maximum cell dimension (R) has the same value with the ratio of the minimum cell dimension for box-counting rd, min to R in each observation area. When the values of lmin/R and rd, min/R was about from 0.02, we could reasonably compare the fractal dimensions of fracture space filling in the rage from microscope scale to decimeter scale of different observation areas.
Temperature in a test pit of a metal mine was measured during a half month by a distributed temperature sensor. The sensor was optical fiber jacketed by a metal tube. The fiber bound along boring rods was inserted for 145m into the pit. Because of the metal jacket, the fiber sustained less damage by friction against rock around the pit. It stood also corrosion by hot water in the pit for a half month. The measured temperature was biased from the true one because the optical character of the fiber was different from that of the reference fiber constructed in the measuring system. In addition, owing to heat conduction along the fiber, temperature of a hot spot like inflow of hot water into a borehole was affected by temperature in surroundings. In spite of these errors, continuous measurement of temperature distribution by optical fiber helped us to understand the condition in the pit. It showed the change of flow rate in the pit and the difference of geological features of rock surrounding it.
Heat extraction tests through natural joints at high temperature were conducted at Toyoha Mine. At these tests, cold water was injected to a borehole at constant flow rate. Water which flowed through natural joints was produced from other boreholes. Flow rate, temperature and pressure of water at production borehole were measured during these tests. To understand the flow paths and the nature of heat extraction zone, computer simulation model was developed using FEHM (Finite Element Heat and Mass Transfer) code. And also we calculated the case of the continuous injection by using FEHM model. The major results obtained from simulation are as follows. 1) The natural water flow through the vein was about 0.182kg/s in the vertical direction and there was small amount of natural water flow in the horizontal direction. 2) As the natural water flow through the vein was larger than the injection flow rate of 0.67kg/s, the temperature and the flow rate of water from production boreholes did not change clearly by injecting water. 3) However, the temperature at the point where the production borehole No. 456 intersects the vein decreased a little as indicated by temperature measurement using the optical fiber. 4)Even in case of the continuous injection, cold area of Toyoha mine model was very narrow.