Computer modelling retrospective on Wairakei and Ohaaki

Abstract

Our modelling studies of the Wairakei and Ohaaki geothermal fields have been ongoing for more than 30 years and it is now possible to check back on the predictions made from early studies carried out in the 1990s and early 2000s and determine how well the predictions made by those models compare with reality. There are several problems with a retrospective assessment of the 1990s and early 2000s models. The most significant is that the future scenarios considered then are different from the actual production and injection strategies followed. We overcame this difficulty by using the actual production /injection history with the old models. In general the models perform quite well in matching the production history but in both cases they are too conservative and the actual performance of Wairakei and Ohaaki has been better than that predicted by models from the late 1990s or early 2000s. In both cases the number of make-up wells used in the original future scenarios turned out to be more than have actually been required The main discrepancies arise because of the development of new and/or deeper production zones, not included in the old models. In some cases, e.g. the shallow steam zone at Te Mihi (Wairakei), the permeability has turned out to be even higher than the high value used in the old model and production has continued past the point when the old model predicts failure should occur due to a large pressure decline.