Spectral Analysis of Geophysical Data - Advances in Data, Methods, Models and Their Applications in Geoscience - page 27

2
Spectral Analysis of Geophysical Data
Othniel K. Likkason
Physics Programme,
Abubakar Tafawa Balewa University, Bauchi,
Nigeria
1. Introduction
The usefulness of a geophysical method for a particular purpose depends, to a large extent,
on the characteristics of the target proposition (exploration target) which differentiate it
from the surrounding media. For example, in the detection of structures associated with oil
and gas (such as faults, anticlines, synclines, salt domes and other large scale structures), we
can exploit the elastic properties of the rocks. Depending on the type of minerals sought, we
can take advantage of their variations with respect to the host environment, of the electric
conductivity, local changes in gravity, magnetic, radioactive or geothermal values to
provide information to be analysed and interpreted that will lead to parameter estimation of
the deposits.
This chapter deals with some tools that can be used to analyse and interpret geophysical
data so obtained in the field. We shall be having in mind potential field data (from gravity,
magnetic or electrical surveys). For example, the gravity data may be the records of Bouguer
gravity anomalies (in milligals), the magnetic data may be the total magnetic field intensity
anomaly (in gammas) and data from electrical survey may be records of resistivity
measurements (in ohm-metre). There are other thematic ways in which data from these
surveys can be expressed; depicting some other attributes of the exploration target and the
host environments.
Potential fields for now are mostly displayed in 1-D (profile form), 2-D (map form) or 3-D
(map form and depth display). Whichever form, the 1-D and 2-D data are usually displayed
in magnitudes against space (spatial data). When data express thematic values against space
(profile distance) or thematic values against time, they are called time-series data or time-
domain data. The changes or variations in the magnitudes (thematic values) with space
and/or time may reflect significant changes in the nature and attributes of the causative
agents. We therefore use these variations to carefully interpret the nature and the structural
features of the causative agents.
Most at times the picture of the events painted in the time-domain is poor and
undiscernable possibly because of noise effects and other measurement errors. A noise in
any set of data is any effects that are not related to the chosen target proposition. Even
where such noise and measurements errors are minimized, some features of the data need
to be gained/enhanced for proper accentuation. The only recourse to this problem is to
make a transformation of the time-domain data to other forms or use some time-domain
tools to analyse the data for improve signal to noise ratio; both of which must be