Petroleum geologists often think that information about seismic velocity is something best left to geophysicists. But there are many reasons for geologists to get involved with velocities. Let’s look at two important uses of velocity data: depth conversion and basin analysis.
Velocity models are geological interpretations
The largest volumetric uncertainty for petroleum exploration prospects, and throughout the appraisal and early development of fields, is usually the size and shape of the ‘container’ — the gross rock volume. This uncertainty might be related to many aspects of seismic acquisition, processing, and interpretation. But seismic depth conversion, traditionally seen as the final step in seismic interpretation, is frequently the source of most of the seismic-derived uncertainty. Even anisotropic pre-stack depth migration does not guarantee the best depth-conversion model.
There is uncertainty (and a need to quantify it) about the relative depths of a horizon across an area: this controls the size and shape of any discovered or potential hydrocarbon accumulation and is required for volumetrics. There is also uncertainty about the depth at a specific location of a certain horizon or horizons: this is required for well prognoses to aid in safe and cost-effective drilling.
Geologists unfamiliar with the principles of depth conversion are often happy to leave this crucial interpretation step to their geophysical colleagues, especially if they see that some mathematical capability may be required. This can be a big mistake. Any variations in velocity in the overburden above your prospect or field are caused by changes in the geology: thus, a depth conversion procedure that isn’t founded on or consistent with a sound geological model is unlikely to be optimum, and may be badly wrong.
This is simply an extension of a general principle of subsurface mapping that the contouring of formation tops or rock properties from well data without a geological model is unlikely to give the best map interpretation. The fewer the data points the more important it becomes to have a good geological model. Coincidentally, having a statistically supported geological model is likely to give an even better interpretation — this is why geologists and geophysicists need to combine forces.
Thinking about depth conversion should be a very early step in a seismic interpretation project because we may need to interpret several shallower horizons, possibly solely for depth conversion purposes. Having a clear understanding of the geological history of an area prior to starting detailed interpretation is an essential starting point. Moreover, as more and more seismic data are necessarily processed in depth to handle rapid spatial changes in velocity, understanding how the velocity of the shallow geology varies is crucially needed very early in the seismic processing.
Three things to look for when considering depth-conversion uncertainties are:
- Low structural relief. Small velocity variations can cause large changes in the size and shape of low relief closures.
- Highly variable overburden. Large water depth variations, laterally varying lithologies (e.g. carbonate buildups, sand-filled channel systems, or salt), and complex geological structure can all contribute to lateral velocity variations.
- Complex geological history. Sediment velocities reflect their maximum depth of burial, so the recognition of different amounts of uplift across an area is vital to predicting velocity variations accurately.
So, geologists, talk to your geophysical colleagues about depth conversion and be aware that they may need your help to develop the optimum velocity model.
Velocities can help your geological interpretation too!
Analysing velocities from wells and seismic processing can also provide vital clues to help your geological understanding of an area. In exploration projects we often have limited knowledge of the lithologies present. Seismic stratigraphy and seismic facies analysis can only take us so far. Velocities give us an additional dimension. For example, if the ‘pull-up’ under a salt dome doesn’t seem to be as large as you would expect this might indicate the presence of carbonates rather than clastics in the adjacent rim synclines. Equally, lateral velocity trends beneath an unconformity may help us quantify differing amounts of eroded section and may provide a vital clue to understanding the history of petroleum generation and migration in a basin.