What I mean is what are the procedures that should be followed to detect the problem of bad seismic data? If I have bad data and I want to investigate why this area specifically has such issues..

Dear Ahmed, interesting topic to try to help you, as a seismic interpreter, but following the Adriana´s answer, your question is too vast to give you a single answer, for a couple of consideration. Let me explain:

• · When you use the word “bad”, I understand that you are talking about the SNR (signal-to-noise ratio), and in its place, the resolution of the seismic data.
• · The signal-to-noise ratio is a standard to measure the quality of seismic data and plays an important role in seismic processing and interpretation.
• · The “reasons behind bad seismic quality” This could be due to a myriad of reasons, to name a few:

The cost, the amount of money you are available to spend: the most energy injection to ground, the most money, because you are increasing the points of sample by unit area (Vp´s / km2)

environmental or logistical conditions,

ground conditions in the acquisition (for example, the weathering thickness and velocity contrast),

the type of final seismic processing,

the type of migration, just one item, inside the processing...

or for many other reasons that you are not comfortable as a seismic interpreter: because a particular seismic dataset was good for illuminate the faults, but it doesn’t have the azimuths and offset distribution to go to reservoir characterization, and so on...

• · Anyway, when you say “…the procedures that should be followed to detect the problem of bad seismic…”, there are a lot of answers, for instance, if you are talking about seismic design, seismic acquisition, seismic processing or seismic interpretation, as I wrote above.
• · For example, if you are in seismic processing stage, to improvement of resolution after migration, deconvolution is an important point, but on the other hand, this is often met with resistance as it is considered to have a negative effect after migration.
• · But if you are in seismic interpretation, maybe you are interesting to characterize your reservoir doing seismic sequence-stratigraphy-examination, and, a noise be deliberately adding to the seismic to improve the resolution. Therefore, you are comfortable, but for another interpreter the seismic may appear wormy and she / he will try to reduce the add noise…

Therefore, if you want to know the origin of the possible causes of your "bad" seismic data, you should start by reviewing:

1- Which was the geological objective of the original work (structural, stratigraphic, reservoir characterization, etc)

2- What were the design, acquisition and final processing parameters applied. Would you use another sequence of processing? Would you need a closer SLI-RLI? Would you need a small bin-size? Would you need an increase in the coverage fold at target?

3- What are the responses based on the SNR signal-to-noise ratio in your window of interest, is it according your target size and thickness for example?

Best regards Ahmed, Mario E. Sigismondi