| Title | Random Simulation with Geologic Control in Assessment of Geothermal Resources of the State of Goi·s, Central Brazil |
|---|---|
| Authors | Ferreira, L.E.T. and Hamza, V.M |
| Year | 2005 |
| Conference | World Geothermal Congress |
| Keywords | Random Simulation, Resource estimates, Goias (Central Brazil) |
| Abstract | Recent results obtained in evaluation of geothermal resources of the state of Goi·s (central Brazil) are presented. The procedure adopted is based on random simulation of the main geothermal parameters over a regular mosaic of elemental areas defined over the study area. The observed values of temperature gradients and available data on flow rates of shallow wells were used as bounding limits in the simulation process. In addition, weighting factors were assigned depending on the geotectonic characteristics of the study area. Heat flow data available from 18 localities were used in conjunction with hydrogeologic data from 636 wells in generating a network of 6400 regular grid points. The table below provides a summary of the results obtained in such random simulation schemes. Scheme Search Energy Resource Recoverable Radius Flux(GW) Base(J) Fraction(MW) 1 0.50 18.39 1.13 Not Estimated 2 1.50 27.32 1.75 Not Estimated 3 2.00 29.30 1.89 Not Estimated Volumetric -- 36.31 2.54 8.1x1021 The results obtained indicate that resource estimates based on random simulation process are better representative of in-situ conditions those based on the conventional volumetric methods. However the estimates are to some extent dependant on the search radius and on the interval used in random simulations. More meaningful results can be obtained through the use of finer grids but such improvements are dependant on the availability of detailed geologic maps. Another advantage is that the random schemes can easily be adjusted for eventual changes in the constraints and in the parameters used. It also allows virtual experimentation of resource estimates. The method is potentially useful in cases where the geographic distribution is poor and non homogeneous. |