Authors: Torsten Lange¹, Alexander Kissinger², Martin Sauter¹, Rainer Helmig², Michael Heitfeld³

¹Georg-August-Universität Göttingen
²Universität Stuttgart
³Ingenieurbüro Heitfeld - Schetelig, Aachen

This text relies on the articles “Hydraulic fracturing in unconventional gas reservoirs: risks in the geological system part 1 & part 2” that was originally published in Environmental Earth Sciences, Part 1: September 2013; Part 2: May 2013.

Introduction

The study presented here is part of the ExxonMobil information and dialog process on the technology of hydraulic fracturing (Ewen et al., 2012). It is the first larger German study on the quantification of possible risks of groundwater contamination caused by substances transported into the groundwater area by frack fluids, initiated by the Injection of these fluids during the process of gas production from unconventional reservoirs. The topics discussed in the study are the results of the general duty of care principle for groundwater stipulated in the European Water Framework Directive and the corresponding laws and regulations on the one hand and the in-depth analysis of the topic of different societal interest groups on the other hand. In the US, where the technology has already been used on an industrial scale and under different legal framework conditions for a long time, there is evidence of possible or actual risks to groundwater. The goal of this and similar studies is therefore to prepare a catalog of requirements before hydraulic fracturing activities are carried out in Germany that evaluates the risks to the groundwater and minimizes these risks if the technology is applied.

Topics

The topics discussed in this study are related to the assessment of the risk of groundwater contamination caused by the transport of the fluids and additives in the overburden that are released during hydraulic fracturing and via existing permeable fault zones. Two time and space scales (depending on the process) are important for the transport of substances: the short-term (approximately 12 hours), local, vertical transport under high fluid pressures during hydraulic fracturing (model scenario 1) and the long-term (30 years) regional transport under the conditions of the regional hydraulic gradients (model scenario 2) in a deep aquifer system. The study also examines the possible migration of methane from its reservoirs into near-surface aquifers and into the atmosphere (climate relevance, model scenario 3) after the production phase. The study considers the different geological structures of the Lower Saxony Basin and the Münsterland Cretaceous Basin for different geological settings. These settings cover a spectrum of different geological constellations, where the size of the overburden and the presence of salt horizons and permeable fault zones were varied.

Methods

The deterministic approach selected for this study is based on the development and simulation of scenarios that require a basic understanding of the processes and adequate knowledge of system geometries and variables. A probabilistic approach was not possible due to the fact that the amount of comprehensive data sets that would be required for determining occurrence probabilities, extent of damage and risks was not available. In addition to the regional and local relevance of the study, approaches that can be applied in a more general context, that is to say to other sites, were developed. This includes the development of geological settings and the use of a specifically conservative approach while taking into account the impact of cumulative effects of factors that are unfavorable for the prognosis, i.e. that encourage substance migration.
This is why specific transport processes that are not determined by advection, such as matrix diffusion, sorption processes, and degradation processes were not included (1) even though they have a considerable "favorable" impact on contaminant migration. Instead, (2) while taking into account the hydrogeological plausibility of highest permeabilities and lowest effective porosities for (3) permeable fault zones between the actual fractured horizon and the near-surface aquifers were taken into account. The operational framework conditions were selected such (4) that the maximum pressures can be maintained over the planned drilling period despite a loss of fluids into highly permeable structures above the fractured horizon which would be prevented by technical control mechanisms under real operating conditions. This means that the worst case scenarios are assumed in which large amounts of frack fluids would reach a permeable fault zone of the overburden. Moreover, (5) maximum possible upward-directed vertical pressure potential differences were considered for the simulation of regional transport in the Münsterland Cretaceous Basin.

The described cumulations of unfavorable factors for the distribution of substances in the individual scenarios are extreme cases that are physically possible and theoretically conceivable and are thus within the upper limit range of potentially negative effects of hydraulic fracturing.

One decisive factor for the migration of fluids is the fracture height, which is one of the framework conditions for estimating transport of substances. Based on microseismic measurements of 3000 individual hydraulic fracturing operations in different areas and under varying operational conditions, the Pinnacle Halliburton study (Fisher & Warpinski, 2011) shows that the hydraulically induced fracture heights do not exceed 1500 ft (approx. 500 m) and are usually significantly lower.

Aside from the comprehensive geological and hydrogeological inventory of the investigated areas, geological settings were selected based on the geological conditions that are significant for the distribution of substances in order to ensure that the natural geological heterogeneity is taken into account. The main criteria were the size of the overburden and the existence or absence of fault zones and salt horizons. The selection of the effective, advective transport parameters, the permeability and the effective porosity for lithologically clean units and faults was carried out based on comprehensive literature research. On the other hand, transmissivities for possible fault zones in the Münsterland Cretaceous Basin were estimated by means of mixture balance calculations. Here, the salinities of the near-surface groundwater, that are partly increased in the surroundings of the selected settings, and the salinity degree of the deep Cenoman/Turon aquifer were used. The usually increased salt contents of the overburden of the "Emscher Mergel", which is several hundreds of meters thick, were ignored in the light of the conservative approach.

Results

Based on the presented approach, the substance transport was simulated for the three model scenarios described above, the input variables and parameters were varied, and the following results were derived: 

The scenario 1 simulations (transport during hydraulic fracturing) resulted in a maximum vertical fluid transport of 50 m under unfavorable conditions. This value is used to estimate minimum distances between the fracturing head and the near-surface groundwater.

For scenario 2 (regional transport, deep aquifer) the Lower Saxony Basin was not considered because it can be presumed that there is a closed hydraulic system in the exploration areas and/or relevant depths. However, there is a regional hydraulic gradient in the Münsterland Cretaceous Basin facing southwest, which theoretical allows annual transport distances in the 1 to 25 meter range depending on the model used. With high vertical pressure gradients, the worst-case scenario simulations exhibit a vertical migration for the considered time period. However, if factors such as matrix diffusion and sorption and degradation processes are realistically taken into account, it must be said that a significant reduction of the transported organic components is to be expected. Due to the overestimates that are the result of the specifically conservative approach, it is therefore not yet possible to make a final statement regarding the long-term migration of substances.

A long-term migration (scenario 3) of methane from a depleted gas reservoir through the overburden and into the atmosphere is possible based on conservative estimates: permeable fault zone, low residual saturation with low effective porosity, large volumes of gas that can be released, thin overburden, absence of salt horizons. However, there are still significant uncertainties regarding the input parameters. A site-specific evaluation is required in any case.

Recommendations

A recommended minimum distance of initially 1000 m between the perforation in the piping and the ground level can be derived from the maximum transport distances and hydraulically induced fracture heights: 500 m fracture height + 200 m vertical migration distance (= double mobilization of frack fluids à 50 m x safety factor of 2). The top 300 m (stress-relieved to loose near-surface zone of the overburden), including the 100 m thick near-surface aquifer, are not regarded as barrier-relevant.

To protect usable deep aquifers, a safety distance between the base of the groundwater aquifer and the piping perforation of 600 m is recommended: 500 m fracture height + 100 m vertical migration distance of the frack fluids (= 2 x fracking à 25 m (based on model calculations) x safety factor 2).

There are no recommendations for methane migration because there is not enough information on the source term. Generally, thick overburdens with a low permeability, in particular evaporite horizons, are always an effective barrier.

It is recommended to avoid hydraulic fracturing in drinking water protection zones 1 and 2, in mineral spring protection areas, in areas that are tectonically critically stressed, close to heavily-fissured or brittle zones or near old wells/shafts. Prior to each hydraulic fracturing procedure, a site analysis including a documentation of the existing condition is to be carried out and the site is to be monitored accordingly.

References

Lange, T., Sauter, M., Heitfeld, M., Schetelig, K., Brosig, K., Jahnke, W., Kissinger, A., Helmig, R., Ebigbo, A., Class, H. (2013): Hydraulic fracturing in unconventional gas reservoirs: risks in the geological system, Part 1. Environmental Earth Sciences, 70 (8), 3839-3853

Kissinger, A., Helmig, R., Ebigbo, A., Class, H., Lange, T., Sauter, M., Heitfeld, M., Klünker, J., Jahnke, W. (2013): Hydraulic fracturing in unconventional gas reservoirs – Risks in the geological system, Part 2. Environmental Earth Sciences, 70 (8), 3855-3873

Literature

Fisher, K., & Warpinski, N. (2011): Hydraulic fracture-height growth: real data. Soc Petrol Eng SPE 145949

Ewens, C. Borchardt, D., Richter, S. Hammerbacher, R. (2012): Hydrofracking risk assessment – executive summary.

Related Literature

Sauter, M., Helmig, R., Klünker, J., Lange, T., Kissinger, A., Brosig, K., Jahnke, W., Heitfeld, M., Scheltig, K. (2012): Risiken im Geologischen System bei der Fracking-Technologie. Wasser und Abfall, 6, pp. 16-20

Sauter, M., Helmig, R., Scheltig, K., Brosig, K., Kissinger, A., Lange, T., Heitfeld, M., Klünker, J., Jahnke, W., Ebigbo, A., Paape, B. (2012): Gutachten zur Abschätzung der Auswirkungen von Fracking-Maßnahmen auf das oberflachennahe Grundwasser –  Generische Charakterisierung und Modellierung.