Angle of Incidence: Snell's Law, Critical Angle, and AVO Analysis
The angle of incidence is the acute angle formed between an incoming seismic ray and the normal (perpendicular) to a reflecting or refracting interface at the point of intersection, measured in degrees from the normal rather than from the interface plane itself. This geometric convention is the universal standard in seismic wave physics because Snell's Law and the Zoeppritz equations, which govern how energy is partitioned between reflected and transmitted waves at an interface, are formulated in terms of the normal-referenced angle. At normal incidence (angle of incidence = 0 degrees), the incoming ray strikes the interface perpendicularly, generates a reflected ray returning exactly along its incoming path, and transmits a refracted ray straight through without deflection; the amplitude of the reflection depends only on the contrast in acoustic impedance between the two layers. As the angle of incidence increases from 0 toward 90 degrees, the transmitted ray bends away from the normal in the faster medium (per Snell's Law: V1 / sin θ1 = V2 / sin θ2), and the partitioning of energy between reflected and transmitted waves evolves in ways described by the full Zoeppritz equations involving both compressional and shear velocities and densities of both layers. At the critical angle of incidence (θc = arcsin(V1/V2), applicable only when V2 > V1), the transmitted compressional wave travels exactly along the interface and becomes a head wave that radiates energy back to the surface; beyond the critical angle, total internal reflection occurs and all incident energy is returned as reflected waves with a phase shift. Understanding the angle of incidence at every point in a seismic acquisition geometry is central to amplitude-versus-offset analysis, seismic migration, depth conversion, full-waveform inversion, acoustic borehole log interpretation, and the design of surface seismic surveys targeting hydrocarbon indicators in the Western Canada Sedimentary Basin and worldwide.
Key Takeaways
- Snell's Law and ray bending at interfaces: When a seismic wave crosses an interface between two media with different velocities, the angle of incidence in the upper medium and the angle of refraction in the lower medium are related by Snell's Law: V1 / sin θ1 = V2 / sin θ2. If V2 is greater than V1, the refracted ray bends away from the normal, meaning θ2 is greater than θ1. If V2 is less than V1, the refracted ray bends toward the normal, and θ2 is smaller than θ1. In seismic exploration, Snell's Law is the foundation of ray-based seismic modeling, which predicts how wavefronts propagate through layered velocity models built from well logs and inversion results. Velocity modeling errors that shift the assumed velocity contrast at a key interface cause the modeled angle of incidence at that interface to deviate from the true angle, introducing amplitude errors in AVO analysis and depth errors in pre-stack depth migration, both of which affect well location decisions in structurally complex Foothills plays.
- Critical angle and head wave generation: The critical angle of incidence (θc) is the angle at which the refracted wave in the faster lower medium is bent exactly to 90 degrees from the normal, meaning it travels along the interface. This occurs only when the lower layer's velocity exceeds the upper layer's velocity, and the critical angle equals arcsin(V1/V2). For a Paleozoic carbonate (V2 = 6,000 m/s) overlain by Cretaceous shale (V1 = 3,200 m/s), the critical angle is arcsin(3,200/6,000) = 32.2 degrees. At source-receiver offsets that produce angles of incidence exceeding 32.2 degrees at that interface, the reflected amplitude jumps dramatically because post-critical reflections carry total energy plus a phase shift. This transition zone is observable on pre-stack common-midpoint gathers as a change in the character of the reflection wavelet, and distinguishing post-critical reflections from legitimate AVO anomalies is a key quality-control step before AVO attribute extraction in deeper WCSB plays.
- Normal incidence reflection coefficient and acoustic impedance: At zero angle of incidence, the reflection coefficient R0 depends only on the acoustic impedances of the two layers: R0 = (Z2 - Z1) / (Z2 + Z1), where Z = density times P-wave velocity. Positive R0 means the lower layer has higher impedance (harder), producing a reflection with the same polarity as the incident wave; negative R0 means the lower layer has lower impedance (softer), producing a polarity reversal. In seismic interpretation, identifying anomalously low R0 values in pre-stack data at known sand tops provides the foundation for bright-spot analysis and AVO attribute mapping. A typical Cardium sandstone saturated with 37-degree API crude (impedance approximately 6.2 x 10³ g/cm²·m/s) overlying Ireton shale (impedance approximately 7.8 x 10³ g/cm²·m/s) gives R0 = (7.8 - 6.2)/(7.8 + 6.2) = 0.114, a moderately positive hard kick, but gas saturation of the same sandstone lowers its impedance to approximately 4.9 x 10³ and produces R0 = 0.23, a 100 percent amplitude increase detectable in high-quality 3D data.
- AVO analysis and pre-stack angle gathers: Amplitude variation with offset analysis uses the angle of incidence at the target reflector, calculated for each source-receiver offset in a common-midpoint gather using ray-traced velocity models or approximated by the formula sin²θ = offset² / (offset² + 4 x V²rms x t²0/4), to fit the Shuey two-term approximation R(θ) = R0 + G x sin²θ and solve for the intercept R0 and gradient G independently. The gradient captures how the reflection amplitude changes with increasing angle of incidence and is sensitive to the contrast in Poisson's ratio (related to Vp/Vs ratio) across the interface. Gas sands typically display strongly negative gradients (Class II-IV AVO), while brine sands and shales display near-zero to positive gradients. Misidentifying the correct angle of incidence, for example by using an inaccurate RMS velocity field for the gather moveout correction, shifts the AVO fitting axis and can produce false anomalies or suppress real ones in Viking or Glauconitic channel plays.
- Wide-angle reflections and seismic survey design: Wide-angle seismic data, acquired at source-receiver offsets that produce angles of incidence of 30 to 60 degrees at target depth, carry information about both P-wave and converted P-to-S wave reflection coefficients that is absent from conventional short-offset data. Designing a seismic survey to achieve a target maximum angle of incidence requires calculating the offset-to-depth ratio needed at the shallowest and deepest target horizons and specifying receiver line spacing accordingly. In Duvernay shale plays at 3,500 to 4,000 metres depth, achieving angles of incidence of 40 to 45 degrees for reliable AVO gradient extraction requires maximum source-receiver offsets of 4,500 to 5,500 metres, which in turn drives patch geometry, the number of simultaneously active receiver lines, and the total number of recording channels per shot, all of which scale the program cost. A 200 km2 Duvernay 3D program with 5 km maximum offset and 300-metre receiver station spacing requires approximately 2,000 receiver channels and costs CAD 6 to CAD 9 million for field acquisition, not including processing or interpretation.
AVO Analysis in Practice: Pre-Stack Gather Interpretation
AVO analysis begins with pre-stack common-midpoint gathers in which each trace represents a different source-receiver offset and therefore samples the target reflector at a different angle of incidence. The gathers are sorted into offset or angle bins, with angle conversion performed by ray-tracing through the picked interval velocity function from the shallowest reflector to the target. Gather quality is assessed by examining the flatness of reflections after NMO correction: residual moveout on a key calibration horizon indicates that the velocity used for NMO correction is inaccurate, which introduces an angle-dependent time shift that distorts the AVO response and mimics a false gradient anomaly. Quality pre-stack gathers show flat, well-aligned reflections on target horizons with minimal stretch at far offsets, coherent noise suppressed by trace balancing, and amplitude spectra that remain consistent across offset bins after scaling for geometric spreading and inelastic attenuation.
Shuey's two-term approximation is valid for angles of incidence up to approximately 30 degrees; beyond this range, the three-term Shuey or full Zoeppritz inversion is required to avoid systematic errors in the extracted R0 and G attributes. In the WCSB, many Cretaceous targets (Viking, Glauconitic, Mannville) lie at depths of 600 to 1,800 metres, and maximum usable offsets of 1,200 to 2,400 metres produce maximum angles of incidence of 20 to 35 degrees, which places most practical analysis within the two-term range. Deeper Devonian carbonate targets and Duvernay shale plays at 3,000 to 4,500 metres depth, paired with 4,500-metre maximum offsets, reach angles of 30 to 45 degrees and require three-term or Zoeppritz inversion, with Vp/Vs ratio constrained by multicomponent seismic or dipole sonic logs in offset wells.
The AVO crossplot of intercept R0 versus gradient G is the primary display for hydrocarbon indicator identification. Gas sands plotting in the third quadrant (negative R0, negative G) represent Class III AVO anomalies, the classic bright-spot response where amplitude increases with offset in a sand with lower impedance than surrounding shale. Class II anomalies, where R0 is near zero and G is strongly negative, produce near-zero near-offset amplitude but large far-offset amplitude and are diagnostic of gas sands with impedance close to the encasing shale, a situation common in the deep Montney. The product R0 x G and the scaled Poisson's ratio reflectivity (R0 + G) are AVO attributes derived from the angle-of-incidence-indexed gather that highlight hydrocarbon indicators while partially suppressing the background shale trend. Calibrating these attributes against known gas, oil, and brine wells in the area before using them to rank undrilled prospects is the standard workflow in every major AVO play.
In the Ferrier Cardium play of west-central Alberta, a 3D seismic program over a 72 km2 block uses pre-stack angle gathers with maximum angles of incidence of 32 degrees at the Cardium target (depth 1,750 metres) to map a Class IIp AVO anomaly associated with a tight, low-porosity sand containing minor gas saturation at the crestal position. The operator applies two-term Shuey inversion to 288 CMP gathers over the anomaly, extracting R0 and G attributes on a 40-metre x 40-metre grid. The resulting AVO anomaly maps a connected channel body of 3.4 km2 with internal amplitude variations consistent with 6 to 14 percent porosity inferred from Vp/Vs ratio. Two horizontal wells drilled into the highest-amplitude portion of the anomaly at locations selected by the AVO attribute map encounter average net Cardium pay of 5.8 and 6.4 metres respectively, within the pre-drill prediction of 5 to 7 metres, confirming that the angle-of-incidence-based AVO analysis correctly characterized the reservoir extent at a 3D program cost of CAD 3.1 million against an average well cost of CAD 3.8 million per horizontal.
Fast Facts
Normal incidence (angle of incidence = 0 degrees) is the only geometry at which the reflection coefficient equals the simple two-layer acoustic impedance contrast formula; at all other angles, both P-to-P and P-to-S conversions contribute energy and the full Zoeppritz system must be solved or approximated. Critical angles for common WCSB rock pairs range from 20 to 35 degrees at Cretaceous sand-shale contacts (V1 = 2,800 to 3,400 m/s, V2 = 3,500 to 4,200 m/s) to 30 to 45 degrees at Devonian carbonate contacts (V2 up to 6,500 m/s). AVO analysis is sensitive to angle-of-incidence errors of more than 2 to 3 degrees, which corresponds to NMO velocity errors of 3 to 5 percent at offsets beyond half the target depth; velocity quality control is therefore mandatory before gradient extraction. The Zoeppritz equations, published by Karl Zoeppritz in 1919, remain the exact analytical solution for plane-wave reflection and transmission coefficients at a flat interface as a function of angle of incidence, P and S velocities, and densities of both half-spaces.