One of the more challenging (and satisfying) experiences is using your seismic data to create an accurate prognosis for your proposed well. Typically this exercise involves a top-down analysis, starting with near surface geologic markers and working one’s way deeper through the stratigraphic column. A few of the nearest wells are used, establishing ranges for formation tops as well as velocities for converting relevant seismic picks to depth.
How about building a prognosis from the bottom up? I imagine starting with a deeper marker, whose depth might be more predictable than first thought, or at least more predictable than a complex overburden above the target zone. How can a deeper marker be more predictable? In an intercratonic basin like the Western Canadian Sedimentary Basin the Palaeozoic, for instance, dips steadily and predictably from the foreland into the foredeep of the Rocky Mountain thrust belt. Palaeozoic depth structure maps, based on regional well control, can provide a relatively accurate datum below your target zone. Seismic data can be used to create an isochron from this deeper datum up to one’s target, and converting this isochron to an isopach yields a depth to one’s target. This approach has been used by Foothills explorationists for decades, especially when complex structuring in the overburden challenged a more conventional top-down prognosis. Not only do you end up with a prognosis, but you’ve also created a regional structure map on a deep marker, also characterizing the regional strike at this stratigraphic level.
When working with deeper datums it is true that one is dealing with generally higher velocity sections, where two-way seismic time uncertainty means greater isopach uncertainty. In compensation, these higher velocity sections tend to be more vertically and areally consistent, and have a more consistent lithology, and are thus more predictable.
Another Western Canadian Sedimentary Basin example that can use a bottom-up prognosis is the Mississippian subcrop play family. The deeper datum often used here is the Mississippian–Devonian boundary of the Exshaw, a good seismic marker. Once again, the isochron from the Exshaw up to one’s prospect top can yield an isopach, which can accurately predict how much Banff, or Pekisko, or Shunda, or Elkton Formation (or all of these) is present at your proposed location. Just as important as the depth prediction is the determination of the stratigraphic interval left at the Mississippian subcrop, some of these intervals being more prone to porosity development than others.
What other seismic phenomena might be visible and predictable in depth or stratigraphic level?
- Detachment surfaces in thrust terranes are often stratigraphically consistent, occurring in the more ductile intervals of the section (e.g. anhydrites in the Wabamun, ductile shales such as the Exshaw or Nordegg).Such detachments may be identified on seismic, confirming a stratigraphic level, hence aiding in the prognosis.
- Synclines, in thrusted terrains, are not only ‘pin’ points for palinspastic restoration, but are also imaged preferentially on seismic, and their depth (at their base) may be prognosed from the sum of all the overlying section, based on both well control and surface outcrop measurements (a powerful combination).
- In extensional environments, growth-style sole faults will often sole out in a common stratigraphic level.
- Salt welds may also occur at stratigraphically consistent levels, at the base of the original salt layer or the base of the remobilized salt pillow.
- Metamorphic or volcanic zones are usually good seismic markers, and their depth may be predicted using non-seismic data such as gravity, magnetics, or magnetotellurics.
Yes, many of these deep markers may have larger uncertainty in their depth predictions, as a result of very sparse well control at these depths, but it is always instructional to examine your prognosis from the bottom up, as a complement to your more conventional top-down prediction.