Characteristics of fluid-induced resonances observed during microseismic monitoring (original) (raw)
Related papers
On the interpretation of resonance frequencies recorded during microseismic experiments
2013
Summary Continuous passive seismic recordings during hydraulic fracture treatments can be used to map the frequency content of wavefields emitted by microseismic events during fluid injection. Recent studies have shown that the frequency content of continuous recordings contains information on the fluid injection. In particular, spectral lines can be caused by different phenomena leading to similar resonance frequencies. The first step is then to identify and separate the possible sources (receiver, path and source effects) to facilitate the interpretation of the specific resonances due to fluid injection. We here report two case studies where resonance frequencies are detected. In the case of the first case study, some low-frequency (5-50 Hz) resonance frequencies are found on the borehole geophones as well as two arrays of broadband stations on the surface. The spatial distribution of the stations that have recorded these resonances suggests that they could originate from the vert...
Potential use of resonance frequencies in microseismic interpretation
The Leading Edge, 2012
Continuous passive recordings of microseismic experiments can be used to detect and trace the modifications in frequency content during hydraulic fracturing or heavy-oil steam injection. We analyze the performance of four different time-frequency representations, namely the short-time Fourier transform, the S-transform, the continuous wavelet transform, and the autoregressive method, on a real microseismic data set of intermediate quality. We show that time-frequency transforms provide an efficient tool to highlight time-varying resonance frequencies occurring during reservoir fracturing. Four distinct resonance frequencies at ∼17, ∼35, ∼51, and 60 Hz are observed during two experiments using the same experimental setup.
Direct correlations between volume injection rates and microseismic resonances
2014
Summary Microseismic experiments are commonly monitored by geophones deployed in observation wells. Rather than studying microseismic events, we employ time-frequency pictures of continuous recordings to delineate several resonances in a 2-stage microseismic experiment. Variations in resonance frequencies are strongly correlated with variations in volume injection rates, especially the slurry flow and the nitrogen injection rate. This strong correlation suggest fluid flow-related mechanisms, like non-Darcian flow, at the perforation location or in the reservoir could be their source. This example shows that resonances interpretation could lead to new ways for reservoir monitoring during hydraulic stimulations, such as for hydrocarbon and geothermal reservoirs, CO2 sequestration or even volcano monitoring.
Interpretation of Resonance Frequencies Recorded during Hydraulic Fracturing
London 2013, 75th eage conference en exhibition incorporating SPE Europec, 2013
Hydraulic fracturing treatments are often monitored by strings of geophones deployed in boreholes. Instead of picking discrete events only, we here use time-frequency representations of continuous recordings to identify resonances in two case studies. This paper outlines an interpretational procedure to identify their cause using a subdivision into source, path, and receiver-side effects. For the first case study, two main resonances are observed both at depth by the downhole geophones and on the surface by two broadband arrays. The two acquisition networks have different receiver and path effects, yet recorded the same resonances; these resonances are therefore likely generated by source effects. The amplitude pattern at the surface arrays indicates that these resonances are probably due to pumping operations. In the second case study, selective resonances are detected by the downhole geophones. Resonances coming from receiver effects are either lower or higher frequency, and wave propagation modeling shows that path effects are not significant. We identify two possible causes within the source area, namely, eigenvibrations of fractures or non-Darcian flow within the hydraulic fractures. In the first situation, 15-30 m long fluid-filled cracks could generate the observed resonances. An interconnected fracture network would then be required, corresponding to mesoscale deformation of the reservoir. Alternatively, systematic patterns in non-Darcian fluid flow within the hydraulic fracture could also be their leading cause. Resonances can be used to gain a better understanding of reservoir deformations or dynamic fluid flow perturbations during fluid injection into hydrocarbon and geothermal reservoirs, CO 2 sequestration, or volcanic eruptions.
SEG Technical Program Expanded Abstracts 2011, 2011
Simple models of fault displacement yield acceleration and displacement spectra from which seismic moment and corner frequency can be determined. The classic Brune model is used here to predict the shape of the source spectrum and to provide scaling relationships between spectral and source parameters. In order to obtain reliable estimates of the source spectrum, the effects of attenuation need to be estimated and corrected. In this preliminary study, Q P = 25 7 and Q S = 22.5 10 are obtained by applying the spectralratio method to a selected set of 10 microseismic events with high signal-to-noise ratio. The Brune model is then fit to several observed S waveforms, yielding source parameters that depend strongly on the geophone sensitivity, instrument gain and Q. More work is needed to reduce uncertainty in these estimates, but these initial results show promise for spectral characterization of microseismic events. The attenuation considerations in this work provide potentially useful constraints for predicting magnitude detection distance relations and suggest that, even neglecting near surface attenuation, only low-frequency (~ 50 Hz) P waves could be observed at propagation distances of ~ 3 km required for surface monitoring.
Identification of Microseismic Attributes Through Spectral Analysis
Unconventional Resources Technology Conference, 2015
The canonical monitoring geometry for hydraulic fracturing processes consists of a horizontal treatment well and a roughly parallel observation well that contains an array of acoustic sensors. This is a costeffective method because it allows for relatively highresolution monitoring with a low number of crosswell geophones. However, there is a significant drawback with this type of monitoring technique. Specifically, moment tensor inversion suffers based on a small solid angle, which is due to limited aperture. While the current literature suggests that additional monitoring wells are necessary to gain a better understanding of microseismic attributes, that solution is accompanied by a much higher cost. Another proposed solution to this problem is through the use of large surface monitoring arrays; however, what is gained in azimuthal coverage is often lost due to the introduction of significant noise. In an effort to minimize overall cost and improve the understanding of microseismic source mechanisms with this basic monitoring geometry, we propose a new approach that relies primarily on spectral analysis. Though careful exploration in the frequency domain, specific parameters like center frequency and bandwidth, combined with knowledge of pumping parameters, lead to inferences regarding microseismic source mechanisms that would be otherwise unavailable. The benefits of this approach are significant due to the fact that an understanding of microseismic source mechanisms can lead to a greater awareness of new fractures and associated permeability of the fracture network. Moreover, circumventing the limitations accompanied by the traditional survey geometry due to limited aperture while minimizing overall cost can have a significant impact on the viability of new hydraulic fracturing processes.
Felt Seismicity Related to Hydraulic Fracturing
London 2013, 75th eage conference en exhibition incorporating SPE Europec, 2013
Hydraulic fracturing is well known to generate seismic events, most of these events are very small magnitude and are of great use when recorded by a properly calibrated array of geophones to delineate the geometry or the fractures and in other microseismic monitoring applications. However, there have been increasing numbers of reports of larger magnitude seismicity. In this presentation, we discuss the instrumentation aspects of properly recording and these larger magnitude events. We discuss a case study where seismicity was recorded by a near-surface network of 4.5 Hz geophones and forcebalanced accelerometers that corresponded to events up to moment magnitudes of 3, large enough to both be felt on surface and to be recorded by distance regional seismic stations up to 100 km away. These events are also accompanied by hundreds of events seen on the near-surface network with magnitudes between 1 and 3. The presence of these events has implications for previous microseismic studies where generally highfrequency (15Hz) geophones are employed to derive the locations. In these cases, the magnitudes of large events together with parameters like the radius of the rupture, will be systematically underestimated. Therefore, this saturation effect will cause a general mis-estimation of the discrete fracture network activated during the fracture. Events that may appear isolated below zone, if they are large enough, can have size dimensions in the range of hundreds of meters to kilometers. Features of such scales can have dramatic effects on the observed seismicity and so their accurate identification using instruments in the appropriate bandwidth is critical to obtaining an accurate picture of the DFN and the potential for seismic hazard associated with hydraulic fracturing.
Resonance in downhole microseismic data and its removal
SEG Technical Program Expanded Abstracts 2016, 2016
We identified resonance due to poor geophone-borehole coupling in downhole microseismic data and proposed to use spik-ing deconvolution and relative spectrum analysis to remove its effect. The resonance may hinder the arrival time picking and contaminate microseismic waveform spectrum. We designed a spiking deconvolution filter to recover the impulse response of the earth. Also, we proposed to use relative spectrum analysis to study microseismic source parameters. The application of deconvolution on Marcellus shale dataset improved the identifiability of the multiple arrivals. Additionally, the relative spectrum analysis is more effective in revealing the actual spectrum characteristics of microseismic waveforms.
Introduction to this special section: Passive seismic and microseismic—Part 2
The Leading Edge, 2012
An introduction to this special section: Passive seismic and microseismic-Part 2 W elcome to the second half of TLE's two-part special section on passive seismic and microseismic. This month, we focus again on monitoring hydraulic fracturing with microseismic with five articles, but also expand beyond "micro"seismicity, to include unintended "induced" seismicity that may occur during injection. Five articles in this special section focus on induced-seismicity topics. In this introduction, we will highlight various issues related to undesired induced seismicity which may be caused by hydraulic fracturing and deep, underground salt water disposal. Why should you care about induced seismicity? The large increase in unconventional plays and hydraulic fracturing, discussed in the November special section introductory article (Goodway, 2012), has been accompanied with an increase in the generation of wastewater, which is a byproduct resulting from flowback after the stimulation procedure. Induced seismicity from the wastewater injection is extremely rare, occurring in less than 1% of the wells (NAS report, 2012, Shemeta et al., this issue). Induced seismicity (M>1) associated with hydraulic fracturing is even more rare. However rare, and regardless of the cause, induced seismicity
Phenomenology of tremor-like signals observed over hydrocarbon reservoirs
Journal of Volcanology …, 2003
We have observed narrow-band, low-frequency (1.5^4 Hz, amplitude 0.01^10 Wm/s) tremor signals on the surface over hydrocarbon reservoirs (oil, gas and water multiphase fluid systems in porous media) at currently 15 sites worldwide. These 'hydrocarbon tremors' possess remarkably similar spectral and signal structure characteristics, pointing to a common source mechanism, even though the depth (some hundreds to several thousands of meters), specific fluid content (oil, gas, gas condensate of different compositions and combinations) and reservoir rock type (such as sandstone, carbonates, etc.) for each of those sites are quite different. About half of the sites are fully explored or even developed and producing fields, and hard quantitative data on the reservoirs are available (well data, reservoir monitoring data, seismic surveys, etc.). The other areas are essentially either explored prospect areas where we did not have access to hard reservoir data or (in only one case) areas where no exploration wells have been drilled at all. The tremor signal itself was observed over ALL locations investigated so far. The signals weaken at the rim of the reservoirs and are not observed outside the reservoir area. There is a strong correlation of the tremor power with the thickness of the hydrocarbon-bearing layers ('pay zone thickness') determined by borehole log measurements. The overall correlation between surface tremor measurements and accessible subsurface well data is higher than 90%. The phenomenological comparison of hydrocarbon tremor signals with volcanic tremor signals from Stromboli and Arenal volcanoes using both conventional spectral analysis tools and non-linear dynamics methods reveals fundamental similarities between those two phenomena as well as their close relation to bandpass filtered noise. Nevertheless, the specific signal sources are expected to be different for volcanoes and hydrocarbon reservoirs. Using the currently available data we present possible concepts (active or passive mechanisms) on the nature of the hydrocarbon tremor source. Our data lead us to conclude that we are most likely observing a characteristic filtering/ mixing effect, with the energy input supplied by the natural seismo-acoustic background. The reservoir, i.e. the hydrocarbon-water-multifluid system contained in a porous medium, is expected to be the oscillatory element able to act as a filter/mixer (resembling essentially a in-reservoir path effect) for the natural seismo-acoustic background. Most intriguing seems the application aspect, i.e. the practical usability of this spectroscopic approach as a direct from-the-surface, non-invasive hydrocarbon indicator. ß