1D joint inversion of MT and TEM data from Ngozi geothermal prospect, SW-Tanzania - an integrated interpretation of geoscientific results

Abstract

Ngozi geothermal prospect is the flagship prospect of TGDC, located in the Mbeya region in the southwestern part of Tanzania. The prospect is within the southern part of the East Africa rift triple junction of Rukwa, Usangu and Karonga basins, which are located where the eastern and western branches of the East African Rift System converge and form an important volcanic zone, the Rungwe volcanic province. This report summarizes the integrated findings of resistivity models based on 1D joint inversion of TEM – MT sounding data and the previous geological and structural work of the area together with geochemical results.

Based on the integrated findings from this study three drilling sites are proposed to hit the reservoir, which is assessed to be 4 km wide and 7 km long. The low resistivity within the fault zone at a depth between 4 and 5 km below ground level (b.g.l.) south of Ngozi Caldera Lake is an evidence of the existence of the localized, relatively young, shallow plutonic bodies which may constitute a suitable heat source for the geothermal system. The eastern and western limits of extension of the reservoir correspond to the fault zones observed in the MT cross-sections south and north of Ngozi Caldera Lake. The eastern limit of the geothermal reservoir also corresponds quite well with the fault mapped during the surface geological survey. It has been observed that the shear zones, including most of the lineaments and inferred faults previously mapped in the area, cause displacement of the conductive anomalies in MT data. Based on these findings, the geothermal conceptual model has been revised and a first priority zone of 28 km2 was defined which has the maximum probability of hosting a geothermal system of 232°C reservoir temperature. The resource capacity of 49.4 MWe with a P90 probability has been evaluated through a volumetric assessment using Monte Carlo approach.