[en] Full text: The purpose of this study is to develop a new indicator of fluid residence in complex geothermal systems. The method is based on a serious of assumptions in geothermal systems. Firstly, the Yangbajain geothermal field of Tibet is chosen to study the behaviors of strontium isotope and verify these assumptions when a deeper thermal fluid ascends. In the past years, two reservoirs have been identified at different depths in the Yangbajain field. A shallow reservoir is found at depths less than 450m, with temperature varying from 150 deg. C to 170 deg. C and decreasing toward the southeast. Parent rock in the shallow reservoir is altered Himalayan granite in the northwest part of the field and Quaternary alluvium in the southeast part. A deep reservoir is found at depths in the range from 750m to 1400m, with temperature higher than 250 deg. C. Parent rock in the deep reservoir is fractured Himalayan granite. Both thermal fluids from the shallow and deep reservoir and cold groundwater are collected for the analysis of chemical compositions and strontium isotope. Rock samples are also collected at different depths in wells for strontium isotope measurement. The results show there is a linear relation between chloride and boron concentration in the thermal fluids. It implies both reservoirs are hydraulically connected in the region. The shallow fluid is a mixture of the deep thermal fluid and cold groundwater. The Na/K ratios of the shallow and deep fluid do not keep a constant due to water-rock interaction at shallow. The rock samples are quite distinct in ⁸⁷Sr/⁸⁶Sr values, which indicates the granites are multiple intrusitions in the field. However, ⁸⁷Sr/⁸⁶Sr values are quite accordant to both the shallow and deep thermal fluid and can be identified easily from the cold groundwater. So it is reasonable to assure strontium component in both thermal fluids are dissolved from a particular granite at depth. The dilution/mixing process seems not to affect significantly the ⁸⁷Sr/⁸⁶Sr values because the cold groundwater is low concentration of strontium. In addition, when the deep thermal fluid approaches the surface, the pressure in the fracture zone decreases so that the thermal fluid can boil CO₂-rich steam escaping from the residual along its ascent channel. This process lead strontium carbonate in the condition of over-saturation. At this stage, it is dominant of precipitation but not dissolution in the fluid. It is clear that the ⁸⁷Sr/⁸⁶Sr value can preserve evidence of fluid residence at depth, where water-rock interaction occurs intensively and most of strontium is dissolved from the rock. The mixing/dilution can not affect the initial ⁸⁷Sr/⁸⁶Sr value apparently. Secondly, this method is applied to the Rehai geothermal field to confirm whether there is hydraulic connection between two hydrothermal areas in the field. The Rehai field is unique high-temperature geothermal system related to the magma chamber of mantle-derived. It is a un-exploited geothermal field in Yunnan Province, with a great potential for electricity generation. Previous researches demonstrate the thermal fluids in the region have similar hydrogen and oxygen composition. These thermal fluids are depleted in boron component. Although two hydromal areas can be easily distinguished on the basis of Na/K and Cl/SO₄ ratios, the ratios are easily changeable by water-rock interaction and oxidation-reduction reactions when the thermal fluid flows up. These two hydrothermal areas of the Rehai are located in different N-S striking faults, but most of people previously believe they are hydraulic connected through E-W striking faults. However, the thermal fluids of these two parts show obviously different in ⁸⁷Sr/⁸⁶Sr values. So it is reasonable to assure they are two independent hydrothermal systems in the field even they have a common heat source and the same meteoric origin. This conclusion will be of great benefit to geothermal exploration and assessment of resources in the future. (author)