Page 268 - Origin and Prediction of Abnormal Formation Pressures

240 H.H. RIEKE, G.V. CHILINGAR AND J.O. ROBERTSON JR.

(G.V. Chilingar, personal communication, in Fertl, 1976, p. 190). This is a valid
observation by Chilingar as demonstrated by the plot of salinity values for log and
laboratory results in Fig. 10-5. Hermanrud et al. (1998, 1999) raised additional doubt
by expanding the question to include other well logging data used to establish such
trends. They evaluated sonic, resistivity, neutron, and density well-log data from 80
wells on the Norwegian continental shelf in the North Sea to show that there is no clear
correlation between the well-log response (abnormally high porosity) and interpreted
fluid pressure. They did not find a depth in these wells below which porosity ceases to
decrease (overpressure indicator); therefore, undercompaction in these shales associated
with the dominating clastic sediments was not demonstrated.
In the offshore, Atlantic Haltenbanken area of Norway, however, both the resistivity
and sonic logs responded to high fluid pressures present in the Jurassic intra-reservoir
shales of the Ror and Not formations. The shales separate reservoir sandstones deposited
in deltaic and shallow marine environments. Hermanrud et al. (1998) suggested that the
well logging tools were responding to textural changes in the shales or microfracturing
rather than elevated porosity induced by the overpressuring of the heterogeneous shales.
One can raise the following question, however: was their choice of formation water
resistivity and matrix transit time values used for evaluating the 'Not Formation'
high-pressure regimes correct? These values were determined using an average porosity
for a low-pressure reference well in the Not Formation, rather than using actual chemical
measurements. Would there have been a more precise demarcation of the porosity-depth
relationship if fluid samples were used?
Burrus (1998) noted that the conversion of log measurements in shales into porosity
values is not straightforward. Density logs are sensitive to changes in lithology, neutron
logs are sensitive to changes in mineralogy, and the sonic logs are not linearly related to
porosities.
The above brief discussion raises concern about the present trend in the literature
to place specific well data from wide areas into a lumped-parameter evaluation plot.
Well-log interpretation techniques should be performed in conjunction with water
analysis of in-situ formation test fluid samples and from the cores of shales and their
associated sandstones. Such analyses should be carried out on a well-by-well basis
rather than making the assumption that the hydrochemical facies hold from one well to
the other.

Hackberry and Manchester fields, Louisiana, U.S.A.
Schmidt (1973) performed an important field case study. He analyzed the pore waters
of both shales and sandstones from the Manchester and Hackberry fields in the south
Louisiana Gulf Coast Basin by determining the concentration of various cations and
anions together with base-exchange capacity, type of exchangeable cations, and mineral
composition of the clays.
Similar data from sandstones, in the A-5 Farmers Land and Canal well in the
Manchester Field, were calculated by Schmidt (1973) using the spontaneous potential
(SP) electric log curve. Sandstone salinity values are based on 56 water sample analyses
for major cations and anions. The sampling was from normally pressured producing
zones in the Hackberry Field, and highly pressured zone in the Manchester Field.