Evaluating the effect of the joining platform on seismic data acquired by geophones mounted to rigid media Paul D. Vincent 1 , Georgios P. Tsoflias ⁎, Don W. Steeples, Ross Black The University of Kansas, Department of Geology, 120 Lindley Hall, 1475 Jayhawk Blvd., Lawrence, KS 66045, USA abstract article info Article history: Received 29 November 2006 Accepted 30 October 2008 Keywords: Near-surface seismology Seismic acquisition High-resolution shallow seismic data can be acquired efficiently by using rigid media to rapidly deploy geophones. Minor changes occur to the recorded wavefield when geophones and spikes are rigidly attached to a bar of square steel tubing. We show through forward modeling and frequency-domain inverse filtering that the bar acts as a linear filter on the data acquired. This filter contains both a phase shift and an amplitude scalar and is a function of spike-to-receiver offset on the bar. Filter characteristics are independent of phase- moveout direction or spike and receiver placement. By combining the phase-shift and the amplitude scalar, frequency-domain filter generation proved more computationally efficient than previously employed time- domain algorithms. While statistical methods show the amplitude scalar in the filter to vary from shot to shot, the phase shift appears to remain constant. This research shows that rigidly mounted geophones have the ability to record the seismic wavefield with high fidelity, thus making the acquisition of ultra-shallow seismic data more efficient. © 2008 Elsevier B.V. All rights reserved. 1. Introduction Ultra near-surface 2D seismic surveys can be executed efficiently and effectively when geophones attached to rigid media are planted using hydraulic force (Steeples et al., 1999a,b; Schmeissner et al., 2001). Rigidly mounted accelerometers can be used to assess concrete slab integrity and material elastic properties (Goupil et al., 2002). Spikes et al. (2005) used geophones attached to channel iron to conduct a common-midpoint (CMP) survey to image reflectors at depths ranging from 25 to 95 m. Blair et al. (2003) and Clark et al. (2004) tested a variety of steel media shapes and discovered that the ground-coupled airwave was reduced both in amplitude and coherency when data were recorded by geophones attached to hollow, square steel tubing. For brevity, the hollow, square steel tubing to which the geophones are mounted for automated planting will simply be referred to as the ‘bar’ (Fig. 1). Data acquired for the referenced research also showed that the bar has an effect on the recorded ground motion. While the targeted reflectors were recorded in both Blair and Clark's experiments, portions of the wavefield were distorted. The difference between data acquired by individual hand-planted geophones and by geo- phones mounted on the bar appears to be most pronounced when the recorded wavefront has a relatively low phase-velocity or a relatively high dominant frequency. Two phases so affected are high-frequency, low-velocity airwave and low-velocity surface waves. Vincent et al. (2004) found that the reduction in airwave coherency in bar-mounted geophone data is dependent on the azimuth of the source location in relation to the receiver line. Data recorded from shot locations along a line orthogonal to the bar orientation exhibited a larger amplitude airwave that also remained coherent in the data for a longer period of time than when hand- planted geophones were used to record data. Furthermore, these experiments indicated a ~60° azimuthal zone off both ends of the bar in which the coherency and amplitude of the recorded airwave were sharply reduced compared to hand-planted geophones. Airwave energy is a major source of noise in shallow seismic data, therefore its suppression is desirable. An example of the azimuthal variability of airwave coherency is shown in Fig. 2. Ralston (2002) used a least-squares time-domain approach to produce time-domain filters for removing bar effects for three- component seismic data acquired by Gal'perin geophones mounted on channel iron. He concluded that the filters were site dependent. This paper presents a solution to the square-tubing filter in the frequency domain and shows that the proposed filter is variable and must be determined for each shot in the survey. 2. Methods 2.1. Modeling the bar effect Data acquired for the above-mentioned research indicate that the rigid medium acts as a linear filter on the acquired data. In this paper we test the model that geophones mounted to rigid media record a summed wavelet consisting of the scaled and time-shifted spike Journal of Applied Geophysics 68 (2009) 146–150 ⁎ Corresponding author. Tel.: +1 785 864 4584; fax: +1 785 864 5276. E-mail address: tsoflias@ku.edu (G.P. Tsoflias). 1 Present address: Chevron Corporation, 1500 Louisiana St., Houston, TX 77002, USA. 0926-9851/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jappgeo.2008.10.010 Contents lists available at ScienceDirect Journal of Applied Geophysics journal homepage: www.elsevier.com/locate/jappgeo