Seismic, Well Tests, and Production Data

Different Data, Different Scales

No single measurement sees the whole reservoir. A core sees centimeters, a log sees the rock within a foot of the borehole, a well test (flowing the well and watching the pressure respond) senses permeability over tens to hundreds of meters around the well, seismic images tens of meters vertically but blankets kilometers laterally, and production data integrate the behavior of the whole field over time. Each fills a gap the others leave.

The Seismic-to-Well Tie

Seismic is the only data that covers the volume between wells, so tying it to the wells is essential. Seismic reflections come from contrasts in acoustic impedance (velocity times density). At each boundary the reflection coefficient is $R = \frac{Z_2 - Z_1}{Z_2 + Z_1}$, and the recorded trace is that reflectivity convolved with a seismic wavelet. The widget builds the synthetic trace from the well; lower the frequency and watch thin beds blur together.

That blurring is the key limit. Seismic cannot resolve beds thinner than about a quarter of the dominant wavelength, the tuning thickness $\frac{\lambda}{4} = \frac{V}{4f}$. A 30 Hz survey in 3000 m/s rock cannot separate beds thinner than roughly 25 m, even though a log sees centimeters. That scale gap, fine detail at the wells and coarse coverage in between, is exactly why we need geostatistics to fill the volume.