What Geophysics Actually Is
Geophysics is the study of how physical properties vary inside the Earth and other planetary bodies. It uses measurements of gravity, magnetism, seismic waves, electricity, and heat flow to build models of what lies beneath the surface. The field sits between geology and physics, which means you need both a working knowledge of rock mechanics and enough mathematics to handle wave propagation. I started using Introduction To Geophysics methods during graduate school when we mapped shallow fault zones for a landslide hazard project. The core idea is simple enough in theory, but the field data never behaves like the textbook example. Noise gets everywhere. Equipment fails at inconvenient times. And the inversion problems are often underdetermined by design.
Introduction To Geophysics for Beginners
The entry point into this field is usually a course that covers the governing equations for elastic waves, electrical resistivity, and magnetic anomalies. From there you learn how to convert raw instrument readings into a subsurface model. That conversion step is where most people struggle, because it involves solving an inverse problem that rarely has a unique solution. Seismic methods are the most common starting point. You generate waves at the surface using a hammer plate, a small explosive charge, or a vibroseis truck. Sensors called geophones or hydrophones record the returning signals. P-waves travel faster through dense, stiff rock. S-waves cannot pass through fluids, which makes them useful for identifying water-saturated layers. The travel times between source and receiver give you velocity structure. Electrical resistivity tomography is another standard technique. You drive metal electrodes into the ground and inject a current. The measured voltage drop tells you how resistant the subsurface is to electrical flow. Clay and saline water conduct well, so they appear as low-resistivity zones. Freshwater aquifers and dry bedrock show up as high-resistivity features. This method works best for shallow investigations, typically down to about 100 meters depending on electrode spacing.
Magnetotellurics measures natural electromagnetic fields generated by solar wind interacting with the ionosphere. You record electric and magnetic field components at the surface over several hours or days. The frequency content of those signals controls penetration depth. Low frequencies reach deep. High frequencies stay shallow. The resulting conductivity profile can extend to hundreds of kilometers.
Get the Full Details
How to Read a Geophysical Dataset
Raw data always looks messier than the processed version. A typical seismic shot gather contains direct arrivals, reflections, refractions, surface waves, and random noise all overlapping. Your first task is to separate the signal you care about from everything else. Field static corrections account for near-surface velocity variations that would otherwise distort deeper reflections. When I ran a 2D reflection survey over a karst terrain, the limestone dissolved cavities created hyperbolic diffractions that looked like coherent events to automated picking algorithms. I ended up manually editing the gather and applying a pre-stack migration instead of post-stack. The difference in image quality was immediate and noticeable. Automated picks saved time but cost accuracy in complex geology. Inversion is the standard approach for turning measurements into a geological model. You assume a starting model, calculate what the data should look like, compare it to the real data, and adjust the model to reduce the misfit. The process repeats until convergence or until the iterations stop improving the fit. Regularization prevents the solution from oscillating wildly between data points, but it also smooths real features. Choosing the right regularization parameter is a judgment call that depends on your survey and your geological expectations.
Common Tools and Software
SEGYview and OpendTect are common choices for seismic data visualization and basic processing. OpendTect is free for academic use and handles SEGY files directly. It includes built-in filtering, migration routines, and attribute analysis tools. For resistivity and induced polarisation data, Res2DInv and Res2DMod remain the standard for 2D inversion. They run on Windows and require a license, but the trial version lets you process a limited number of profiles. GeoEFT is a free alternative that handles resistivity and IP inversion with a simpler interface. Python has become the default environment for custom processing workflows. Libraries like ObsPy for seismic data, Emu for magnetotellurics, and SimPEG for unified geophysical inversion give you full control over every processing step. The learning curve is steep, but the flexibility pays off when standard software cannot handle your specific problem.
I recommend downloading OpendTect for initial data inspection and Python scripts for anything that requires custom filtering or inversion strategies. Most academic institutions provide OpendTect licenses at no cost.
Pitfalls That Cost Time and Money
The biggest mistake beginners make is treating a geophysical model as a geological map. A low-resistivity zone could be clay, saline water, or conductive mineralization. The data cannot distinguish between those without additional constraints. Always tie your model to borehole logs or outcrop observations whenever possible. Near-surface weathering layers are another common source of error. A variable thin layer of dry sand over consolidated material can delay seismic arrivals by milliseconds and shift interpreted depths by several meters. Measuring the weathering thickness at multiple points along your profile and applying a static correction removes most of that error. Instrument drift is easy to overlook. Magnetometer readings change with temperature. Gravimeters respond to small elevation differences and local mass anomalies. I once spent two days trying to interpret a gravity low that turned out to be caused by an unrecorded change in battery voltage on the instrument. Always log instrument metadata alongside your field measurements.
Geophysics surveys are expensive to run and interpret. A single 2D seismic line with 24 channels can cost between $2,000 and $8,000 per kilometer depending on terrain and mobilization. Resistivity profiles with multi-electrode arrays are cheaper but still require significant field time. Budget accordingly and define your survey objective before committing resources.
What the Field Actually Feels Like
You spend more time preparing for a survey than executing it. Cable layout, electrode placement, ground coupling, and environmental interference all require attention before you collect a single reading. In urban areas, power line noise contaminates electrical methods almost completely. You either schedule measurements during low-load periods or switch to a different method. Data processing usually takes longer than acquisition. A day of fieldwork might produce a dataset that requires a week of cleaning, editing, and inversion before it is usable. The bottleneck is almost always the inversion step. Convergence depends heavily on your starting model, and picking the wrong starting model can send the solver into a local minimum that looks reasonable but is geologically wrong. The payoff comes when the model matches independent observations. A seismic reflector aligning with a borehole horizon, or a resistivity boundary matching a known fault trace, validates the entire workflow. That alignment is what makes the tedious processing worth the effort.
