Type in a keyword, phrase, region, service, name, etc.
Tom Hultgreen

Given the level of investment required to get geothermal projects off the ground and see them through to completion, having a comprehensive and real-time view of a geothermal reservoir’s viability as early as possible is crucial.

But in the same way that the oil and gas industry deals with carbonate, basement and coalbed methane reservoirs, there are ways to model the subsurface approach to geothermal reservoirs – methods which allow us to better identify the projects presenting the biggest opportunities.

Reservoir modelling helps stakeholders build a robust business case, providing in-depth and live analysis of the drill site, its environmental make-up, how it might respond when work begins and, perhaps most importantly, its possibilities.

Needless to say, this insight is priceless to geothermal energy site developers and project stakeholders. Providing this level of understanding and confirming a reservoir’s commercial viability means geothermal energy developers can deliver successful projects and ultimately, bring sustainable geothermal energy sources onstream.

As experts in this rapidly growing space, we have set out our approach to geothermal reservoir modelling below, alongside some of the other established methods that are used to assess geothermal energy reservoirs.

Our approach to geothermal modelling

There are a number of techniques deployed to model geothermal reservoirs. At AGR, we employ the 3D and 4D static (or geological) and dynamic (or simulation) models, characterising the reservoir’s structural and property aspects to give us a comprehensive geological description of the energy reservoir and its dynamic behaviour over time. Constructing such models is a multidisciplinary effort between the fields of geophysics, geology, petrophysics, reservoir engineering and reservoir modelling.

The reservoir models for geothermal energy studies typically have an architecture signified by an integrated matrix-fractures duality. The matrix part of the model defines the main energy storage rock volume with its vital components and characteristics, whereas the fractures element represents the connectivity network, or transportation ‘highway’ for energy, with its physical signatures.

The modelling process includes gravimetric, magnetic, seismic data and well exploration and production data. These are analysed using both exploratory and confirmatory statistical methods and tools, often in labs.

What is a geothermal reservoir static model?

The static – or geological – 3D model is an integrated model containing the reservoir’s structural and property characteristics, expressed in a conglomerate of integrated sub-models.

The structural part comprises fault and fracture networks, geological formations, zones, segments and layering, and their specific attributes. The property part of the model includes various geological rock type ‘facies’ and expressions for their individual porosity, permeability, phase saturations, fracture intensity/density and aperture characteristics with specific attributes.

Static volume calculations using reservoir-specific energy and formation variables are performed, producing numeric and graphic estimates for aspects such as gross, net, pore, saturated and in-place energy volume figures, given a specific model realisation.

Alongside this, uncertainty and sensitivity analyses are also conducted, producing several stochastic model realisations using Monte Carlo algorithms. The output could be uncertainty ranges – such as P10-P50-P90 volumetric figures – or Tornado diagrams, to understand which aspects of the reservoir are more or less critical. This analysis guides and supports decision-making processes relating to reservoir development and management.


The resulting 3D geothermal reservoir static model, with its various property sub-models for the matrix-fractures duality, is used as a starting point for the dynamic modelling process.

Expanding on the dynamic geothermal reservoir model

To produce the dynamic geothermal reservoir model, we fit additional physical parameters to the static model, such as temperature and pressure fields, plus various dynamic aspects, financial guidelines, as well as constraints from reservoir engineering and production technology. The latter could be various layouts for the well design – or altogether different means for stimulating the reservoir, such as ‘Fish Bone’ and ‘Manara’ technologies.

As touched on earlier, the intention of the dynamic model is to evaluate the reservoir’s dynamic behaviour over time, optimising reservoir development and management. Geothermal dynamic reservoir modelling helps to maximise production, ensure safety and unlock new efficiencies all while minimising cost, risk and environmental impact.

There’s more to it, though. The dynamic model can also incorporate a historical time series of production data, if available. Then, history match simulation results can be used in a feedback loop to update and fine tune the vital parameters of the static model. This approach also improves the accuracy of forecasts around future geothermal reservoir performance and geothermal energy production, with greater precision and reliability in the model predictions.

The static and dynamic geothermal reservoir models are also used for planning infill well drilling campaigns, to optimise energy drainage throughout the reservoir’s lifecycle.

Overview of alternative geothermal reservoir modelling approaches

We’ve outlined our approach above, but various methods of geothermal reservoir modelling have emerged and evolved over the years. Different approaches are often combined and – depending on the unique characteristics of a given geothermal reservoir – these can be categorised as follows:

The Classic approach

The Classic approach relies on the location of interpreted or modelled faults, combined with geological top and base reservoir surfaces. This may also include various internal zones and segments.

Parameters derived from these – such as Distance To Faults (DTF) and Distance To Unconformity (DTU) – are used to guide property modelling, laterally and vertically. The result would typically be four main types of ‘reservoir facies’, namely the Fault Core Zone (FCZ), Fault Damage Zone (FDZ), Pseudo-Matrix (PM) and Weathering Zone (WZ). Establishment of the WZ may arrive from combining well data, DTU and interpretation of, for example, ‘low refraction’ seismic.

Typically, a matrix-fracture duality is established by this approach, where matrix and fracture properties are both dependent on the distance from the faults and top reservoir horizon. A Discrete Fracture Network (DFN), deterministic or stochastic, is established to express fracture properties like fracture intensity, fracture aperture, fracture porosity and fracture permeability.

The Seismic Inversion approach

The Seismic Inversion method trains seismic data on well data to estimate reservoir properties.

Over the two decades, we have witnessed the development of methods for Artificial Neural Network-trained (ANN) porosity and permeability prediction cubes, to be used for model conditioning. The idea is that the seismic cubes reflect varying mineralogy and a degree of weathering and fracture intensity in the geothermal reservoir.

The Geomorphology approach

The Geomorphology method is based on modelling original mineralogy variations in the reservoir, combining this with the one or more paleotopographic shape characteristics of the reservoir formation surfaces. For example, this could be their profile and or plane-form curvature expressions. Peaks, deeps, valleys and ridges in a reservoir formation surface are assumed to reflect paleo landforms, which are then used to guide areas of enhanced, say, erosion and or weathering in valleys and depressions.

The Host Rock approach

The Host Rock approach combines mineralogy, fracture intensity and weathering and or erosion characteristics of the reservoir rocks.

Criteria for, say, ‘weathering degree facies’ is established based on porosity and fracture intensity well logs, dividing the reservoir into segments and or zones classified by various degrees of weathering. DTF and DTU metrics, surface vs. deep weathering and or erosion, in addition to anisotropy may also be included in this approach.

There are, then, a number of ways to approach geothermal reservoir modelling, each of which helps bring clarity, confidence and accurate information to project stakeholders – whether investors, operators or the wider supply chain.

With years of experience in the geothermal energy space, AGR is helping energy businesses navigate the complex geological, technological and commercial challenges they may encounter. To learn more about our work in this area, please contact us.