Seismic Magnitudes, Corner Frequencies, and Microseismicity: Using Ambient Noise to Correct for High-Frequency Attenuation
Seismic Magnitudes, Corner Frequencies, and Microseismicity: Using Ambient Noise to Correct for High-Frequency Attenuation
复制标题
DOI：
10.1785/0120190032
复制
发表时间：
2020-06
影响因子：
3
通讯作者：
A. Butcher;Richard Luckett;Jeffrey M. Kendall;B. Baptie
中科院分区：
文献类型：
--
作者：
A. Butcher;Richard Luckett;Jeffrey M. Kendall;B. Baptie
近年来，随着越来越多地将地下用于工业活动，如水力压裂和强化地热计划，对低震级地震的重视程度越来越高。在管理这些活动的相关地震风险时，震级分布和震源属性是关键的输入，然而微震活动通常观察到不一致和不一致(M<；2)。这在一定程度上是因为他们的脉冲响应是由介质控制的，而不是由源控制的。本文提出了一种从环境地震噪声(κ0_Noise)的频谱衰减估计高频幅度衰减参数的方法。该估计不需要预先存在的地震目录，并且独立于震源属性，因此避免了基于地震的方法的一些主要限制。然后，我们将κ0_Noise引入到Brune(1970年)震源模型中，并计算了英国新奥勒顿附近记录的与煤矿相关的微震事件的震源特性和震级关系。这会产生大约10到100米的破裂半径，这与Verdon等人的发现一致。(2018)，并导致应力降值在0.1至10兆帕之间。在不使用κ0_Noise的情况下，计算这些性能可以得到更高的断裂半径(100~500m)和更低的应力降(∼1×10−2 Mpa)。最后，我们发现用这些源属性估计参数的组合κ0-Brune模式准确地捕捉到了新Ollerton的ML-MW关系，并且应力降对这种关系的梯度有很大的影响。
Over recent years, a greater importance has been attached to low-magnitude events, with increasing use of the subsurface for industrial activities such as hydraulic fracturing and enhanced geothermal schemes. Magnitude distributions and earthquake source properties are critical inputs when managing the associated seismic risk of these activities, yet inconsistencies and discrepancies are commonly observed with microseismic activity (M<2). This, in part, is due to their impulse response being controlled by the medium, as opposed to the source. Here, an approach for estimating the high-frequency amplitude decay parameter from the spectral decay of ambient seismic noise (κ0_noise) is developed. The estimate does not require a pre-existing seismic catalog and is independent of the source properties, so avoids some of the main limitations of earthquake-based methods. We then incorporate κ0_noise into the Brune (1970) source model and calculate source properties and magnitude relationships for coal-mining-related microseismic events, recorded near New Ollerton, United Kingdom. This generates rupture radii ranging approximately between 10 and 100 m, which agrees with the findings of Verdon et al. (2018), and results in stress-drop values between 0.1 and 10 MPa. Calculating these properties without κ0_noise produces much higher rupture radii of between 100 and 500 m and significantly lower stress drops (∼1×10−2 MPa). Finally, we find that the combined κ0-Brune model parameterized with these source property estimates accurately capture the ML–Mw relationship at New Ollerton, and that stress drop heavily influences the gradient of this relationship.