Previous  |  Next

doi:10.2204/iodp.proc.311.203.2008

Summary and discussion

The goal of this study was to estimate gas hydrate saturation from the electrical logs and porosity estimators using Archie analysis while addressing uncertainties in (1) Archie parameters, (2) in situ salinity, and (3) in situ porosity. In this study, estimates of in situ salinity were obtained as accurately as possible and gas hydrate saturation estimates were reported with uncertainties related primarily to Archie parameter estimation. The analysis of the different available porosity measurements showed that a bias in porosity does not significantly affect the gas hydrate saturation estimate so long as the Archie parameters have been properly estimated. However, the most reliable porosity measurement was concluded to be the log density porosity because it is well calibrated to the core porosities and was measured in the same hole as was the resistivity. Uncertainties in the density porosity measurement arise mainly from the statistical uncertainty in the gamma ray count used to calculate the density and the error induced by using an average grain density in mapping bulk density to porosity. However, assuming that these sources of error produce random unbiased noise in the data, they are inherently included in the Archie parameter uncertainties and have therefore been accounted for.

Considering the estimates from density porosity to be the most accurate, gas hydrate saturations averaged over a 10 m window (Fig. F10) show distributed gas hydrate occurrence in many intervals (S_h = ~0.09 ± 0.07 at Site U1326 [170–200 mbsf]; S_h = ~0.10 ± 0.07 at Site U1325 [190–230 mbsf]; S_h = ~0.11 ± 0.07 at Site U1327 [140–225 mbsf]), with average concentrations of 5%–15% of the pore space, over depth intervals of 20–100 m. However, despite the interpreted distributed gas hydrate occurrences, the general agreement between C_cb and C_wb indicate that core salinity measurements can sample pore water that has not been freshened by gas hydrate dissociation upon core recovery. This finding suggests that, on the scale of individual log or core measurements, it is possible to sample gas hydrate–free zones, but at the meter scale, gas hydrate occurrence appears distributed.

One possible exception is at Site U1327, where general core salinity is much lower than seawater. It is concluded that most of the low core salinity is due to a source of low salinity fluid, probably from deeper in the section. However, the estimated C_wb is slightly higher than C_cb, suggesting a minor amount of regional freshening of core salinities from gas hydrate dissociation of pervasive gas hydrate upon core recovery. C_wb is greater than C_cb by ~1.5‰, corresponding to a background gas hydrate pore space saturation of ~3%, but the uncertainties are large. Based on the estimate of C_wb at the BSR, the difference between the background freshening measured in core (~22‰ salinity) and seawater (35‰ salinity) is attributed mostly to regional freshening from deeper sources of freshwater, with ~10% from pervasive gas hydrate freshening. A final answer to this question can only be obtained from direct pore water sampling in situ, which is proposed through Osmo-sampling as part of the second phase of Expedition 311.

Additionally, intervals of very high gas hydrate saturation of >40% of the pore space are observed at Sites U1326 (73–94 mbsf) and U1327 (120–138 mbsf) and were interpreted to be high-porosity sandy turbidite units in which large amounts of gas hydrate has formed. The correlation between these coarser-grained turbidite units and high gas hydrate saturation is the principal observation supporting the hypothesis that gas hydrate occurrence is largely controlled by sediment grain size and associated formation parameters, such as permeability (see the "Expedition 311 summary" chapter).

Comparison with previous interpretation

The gas hydrate saturation estimates from this more comprehensive study at Site U1327 differ significantly from the previous downhole resistivity study at nearby Site 889 of Hyndman et al. (1999). They estimated 25%–30% gas hydrate pore space saturation in the 100 m interval above the BSR, compared to the 5%–15% estimated in this study. This discrepancy is mainly attributed to the choice of pore water salinity baseline used in the analysis, the porosity measurement, and associated formulation of Archie's equation used to estimate gas hydrate saturations.

Hyndman et al. (1999) used a smoothed fit to core porosities in their analysis, whereas here the best results are taken to be those obtained from the log density porosity. In this analysis, using a smoothed core porosity trend yielded 15%–25% gas hydrate pore space saturation in the 100 m interval above the BSR at Site U1327. The difference is that in the 100 m above the BSR, the core porosity is (locally) biased relative to the log density porosity. Because the resistivity is the same, the porosity bias must, to first order, be compensated for by an equivalent volume of gas hydrate (because the Archie parameters are very similar), resulting in higher gas hydrate saturation estimates.

The second reason for the discrepancy is that Hyndman et al. (1999) estimated S_h directly from the in situ salinity method, whereas in this study C_wb is estimated from the in situ salinity method and used as the salinity profile in the core baseline salinity method to calculate S_h. The in situ salinity method, used on its own, overestimates S_h in areas where core salinities are freshened relative to C_cb. This occurs because the in situ salinity method is only truly accurate if the resistivity and core salinity are measured on the same physical sample. This difference accounts for a difference in S_h of 0–0.08 in the 100 m interval above the BSR.

Finally, a minor contribution to the difference in the results is that the simplification n = m adopted in the Archie analysis of Hyndman et al. (1999) accounts for an increase in gas hydrate saturation of ~4% (Riedel et al., 2005). All these factors combined account for a difference of 14%–22% gas hydrate saturation, which is approximately the difference between the reported estimates at Sites U1327 and 889.

Top of page  |  Previous  |  Next