Monitoring and analysis of reservoir pressure changes in a producing geothermal field is a very important component of a resource management program as it helps determine how the reservoir is reacting to production and/or injection and therefore helps inform possible changes in the production and/or injection strategy to mitigate issues that are being encountered. In the Tiwi geothermal field, Philippines, reservoir pressures have been monitored since before commercial production began in 1979 by combining data from intermittent downhole pressure surveys run in individual wells. Although the data are relatively sparce, by combining measurements from all available wells in a defined sector, a reasonable overview of the pressure changes occurring with time is obtained, although the precision of the measurements is limited. In November 2005, capillary tubing was installed in Mat-25, an observation well located in the Matalibong production sector of the Tiwi field, to monitor pressure changes in the deep liquid reservoir on a more continuous basis and with improved precision. Since then, three other wells (Mat-29, 30 and 32) have also been used as observation wells at various times and the collected data from the four wells have provided a reasonably continuous record of downhole pressure changes from 2005 to the present. Since 2008, wellhead pressures have also been monitored to provide data on the pressure changes in the overlying steam production zone, which in the past has been an important source of production for the Tiwi field. From 2005 to 2013, the measurements indicate that deep liquid reservoir pressure was generally increasing, while short-term changes correlated well with injection flow changes to the MatRidge Hot Brine Injection System (MRHBIS) wells, even though these wells are 1.2 to 1.8 miles (2 to 3 km) from the production area where the observation wells are located. However, it was possible to successfully match this interaction using both analytical and numerical models. The main concern with the increasing deep pressure has been the associated rise in the steam-liquid interface, which has allowed the liquid to infiltrate the steam producing zones so that wells that previously produced superheated or saturated steam then started producing two-phase or liquid, as indicated by the change in measured discharge enthalpies. With the apparent correlation between pressure changes and injection and with the positive results in matching the data with both analytical and numerical models, it was concluded that it should be possible to mitigate the rise in pressure and the steam-liquid interface and possibly reclaim steam production from some wells by reducing injection. A pilot field test conducted in 2011 also appeared to confirm that this was possible, and the decision was made to construct a pipeline to divert injection to the SouthEast Hot Brine Injection System (SEBHIS) from the Bariis production sector, which had previously been injected to the MRHBIS. The pipeline was commissioned in July 2013. Modelling results indicated that it would be necessary to reduce the MRHBIS injection by at least 50% to stabilize the pressure. However, this was difficult to attain due to the design of the Tiwi injection system, which relies on gravity feed, but even with the reduction that has been achieved since 2013, deep liquid pressures have continued to rise, resulting in essentially all the Matalibong steam producing wells now producing at two-phase or liquid enthalpies. With regard to defining recharge, the analytical model relies on flow from the overall defined area of the model to the pressure sink while for the numerical model, there is an attached shallow aquifer that was included to initially provide a pathway for outflow from the Matalibong sector and this later reversed to provide a source of inflow once the pressure dropped sufficiently. For both models, as mentioned above, the resulting recharge was sufficient to match the data to 2013. More recent modelling work has shown that even though the reduction in injection has been less than the 50% target, reservoir pressure should have declined, which is contrary to the observed continued increase in pressure. To match the pressure increase from 2013 to 2016, a run using the numerical model indicated that up to 1,500 kph additional influx was required and this would have needed to continue or increase since then. Considering that the pressures are still significantly lower than initial pressures prior to production, it is certainly possible there are sources of recharge, from both shallow groundwater and laterally extensive aquifers that could provide the additional influx. So, the question arises, why did the divergence in pressure occur and why was it only seen after 2013, considering that reasonable matches to the historical data had been obtained with both analytical and numerical models? We still don’t know the answer, but it is a lesson to reservoir engineers and modellers that even with well-matched models, there will continue to be uncertainty in forecasting how a natural system will evolve.

Press the Back button in your browser, or search again.

Copyright 2025, Stanford Geothermal Program: Readers who download papers from this site should honor the copyright of the original authors and may not copy or distribute the work further without the permission of the original publisher.