Introduction and Statement of Problem

Of all of the geophysical methods investigated so far, the acquisition, processing, and interpretation of seismic observations tend to be the most time, equipment, and manpower intensive. As a consequence, compared to the other geophsyical techniques, seismic methods tend to be rather expensive. Even though the seismic method investigated in this particular project, refraction seismology, is inexpensive compared to some of the more comprehensive seismic investigations (i.e., reflection seismic methods), given its expense, you will rarely employ even a refraction seismic investigation before doing other, less expensive, reconnaissance investigations.

In the investigations undertaken thus far, the relevant observations have been measured by sophisticated instrumentation and provided to the user as a single number: the gravitational acceleration, magnetic field intensity, or the apparent resisitivity at some spatial location. In collecting seismic observations, no such simple number is provided. In refraction surveying, for example, the instrument does not simply display the time required for a wave to propagate from the source to a fixed receiver location. Rather, seismic instruments measure and provide records of the time-history of ground motion at a particular location. This has two consequences.

• First, the amount of data collected during a seismic investigation is orders of magnitude larger than collected while conducting the other geophysical investigations described. Along with this increase in data flow comes logistical concerns regarding data storage and data access that we haven't had to deal with in the past. Also, with the increased data flow collected from seismic investigations comes a much greater ability to constrain the details of the underlying geologic structure than we have seen with the other methods.
• Second, it is up to the data processor or interpreter to determine from the records of ground motion the relevant physical parameters; in a refraction survey these would be the times of arrival of particular seismic waves. Thus, unlike the other techniques, the determination of the relevant physical parameters from seismic observations requires human intervention and, as a result, is in itself an interpretative process.

Although reliable interpretations from all geophysical observations require an accurate knowledge of field parameters, the equipment and logistical scope involved in conducted seismic surveys make this issue particularly thorny. Extensive field notes, known as OB (Observer) sheets, are routinely recorded while conducting ALL seismic investigations. The quality and accuracy of these OB sheets can vary greatly and can directly affect the interpretability of the observations collected. For example, the personnel responsible for the processing and interpretation of seismic data may only interact with the acquisition crew through the OB sheets produced during data acquisition. If the information included on these sheets is incorrect or incomplete, it may be impossible to process or interpret the observations or the final interpretation may be biased.

To gain a better appreciation of the importance of providing adequate documentation of the field activities, the interpretive process that is required to extract travel-time information from the field observations, and the noise inherent to all seismic observations, a series of shot records were recorded in the CSM survey field using a variety of seismic sources and array configurations. With this knowledge, and adequate observer sheets, you should be able to determine everything you need about this survey, extract the relevant travel-time information, and construct estimates of the typical uncertainties associated with travel-time observations.

Procedure

Pointed to in the table below are the first three observer sheets from the survey described above.

Observer Sheets - CSM Noise Test

Page 1

Page 2

Page 3

In writing these field notes the following nomenclature has been used:

• Record Length - The total time duration after the initiation of the source that ground motion is recorded,
• Sample Rate - Ground motion is not recorded continuously. Rather, every so often the ground motion that the seismometer senses is sampled by the recording instrument and stored onto tape. This process is called digitizing. Sample rate specifies the time interval, in milliseconds, at which measurements are made,
• Group Interval - In the context of this particular set of OB sheets, this number specifies the distance between geophones,
• Src Cnt - Simply an incremental counter of the number of sources initiated,
• Reel # - The tape number on which the data referred to is stored,
• First File - The Field File Identification Number by which the data stored on the tape is referred,
• Src Loc - The location of the source in station numbers, and
• Comments - This simply contains comments about the particular recording. It may contain information on the source, information on noise, or other information the recorder felt was relevant.

Examine the three observer sheets and then address the following questions:

• Comment on the general quality of the OB sheets. Could you read everything? Were the comments clear? Did you notice any errors?
• From the OB sheets describe the experiment that was performed. Produce a sketch of the geophone and source geometries. What sources were used? Did they all work?
• List the FFID's, or Field File Identification Numbers, of those shot records that you believe will contain reliable observations.
• Is the OB sheet lacking any information that you think should have been included?

Now let's look at some of the data collected during this particular survey. Pointed in the table below are images of the shot records associated with the FFID's described in the OB sheets given above.

Shot Records

FFID 1	FFID 2	FFID 3	FFID 4	FFID 5	FFID 6
FFID 7	FFID 8	FFID 9	FFID 10	FFID 11	FFID 12
FFID 13	FFID 14	FFID 15	FFID 16	FFID 17	FFID 18
FFID 19	FFID 20	FFID 21	FFID 22	FFID 23	FFID 24

Using the OB sheets and the shot records address the following questions:

• Different sources were used in the survey described above. Give a brief discussion of the types of sources used. Your discussion should include a description of the pros and cons of using each source, a description of the expected signal-to-noise ratio of the data obtained by each source, and a cost-benefit analysis of using each source.
• From the well-recorded shots, estimate the time of arrival of the first arriving seismic wave observed at each station. For shot records using the same source at the same source location, average the estimated arrival times at each station and compute the standard deviation of the observed arrival times. Comment on the source/offset variation of these numbers.
• One of the most common methods used to enhance the signal-to-noise ratio of seismic observations is to sum the traces recorded at each station for repeated shots. This process is called stacking. Stacking helps to reduce some types of noise, but it can not reduce other common types of noise.
  

  ---

  1. What do you believe are the most common sources of noise in the shot records given above?
     
     
  2. Which types of noise can be effectively removed by stacking and which types will not be removed by stacking?
     
     
  3. For those types of noise that can not be removed by stacking, what would be your recommendation on designing a survey to minimize their effects?