The air temperature taken at the bridge closest to the time the core was on deck was compared to the average core liner temperature for the core (Fig. F2). In addition, comparisons of day and night and camera operators (noon to midnight vs. midnight to noon watches) were conducted to determine a possible sampling bias (Fig. F3). No correlations were found. Core liner temperatures from top to bottom of the 10-m cores were examined to investigate any effect (particularly warming) associated with the method of scanning or the time elapsed during the scan (~1 min). Ten scans (five each from Sites 1226 and 1230) were randomly selected and compared. No consistent warming of the cores was observed (Fig. F4).
The first independent evidence of gas hydrate at Site 1230 was visual observation of fizzing sediment (small white bubbles) interpreted to be decomposing hydrate in Section 201-1230A-15H-5; 123.5 meters below sea floor (mbsf). A subsequent review of the IR scan for that core revealed that core liner temperatures of the fizzing section were only a few degrees cooler (average = 4°C cooler) than the surrounding sediment (Table T1). Based on this observation, camera span and level were set so that the 15° to 25°C range was visible, which simplified the subsequent identification of cold spots caused by gas hydrate dissociation. Cold spots were first identified with the camera and then visually confirmed to be hydrate nodules or fizzing sediment in Cores 201-1230A-26H and 1230B-12H. Figure F5 illustrates the thermal contrast between an interval of fizzing sediment and the surrounding sediment.
A thorough examination of the downcore temperature plots revealed a greater variability in cores with hydrate nodules or fizzing sediment. Figure F6 illustrates the difference between cores that did not contain hydrate and cores that did contain hydrate or fizzing sediment. The standard deviation of the minimum temperatures in hydrate-bearing cores are generally >1°C, whereas in nonhydrate-bearing cores it is usually <1°C (Table T2). This can be explained by the contrast between low temperatures in the hydrate-bearing sediments and high temperatures typical of associated gas expansion voids (voids warm to ambient temperatures rapidly). This is not a fixed rule, however, and emphasizes the current need for careful core-by-core interpretation of both the thermal plots and the original image files. Although visual confirmation of hydrate is necessary for positive identification, careful analysis of the downcore temperature plots and thermal images suggest other hydrate occurrences at Site 1230 in Cores 201-1230A-11H, 13H, 21H, and 35X and 201-1230B-11H (Figure F7; see also "Appendix"). Hopefully, developments will be made in the future to normalize core temperatures for the wireline trip, allowing confirmation of in situ temperatures and presence of hydrate using thermal imaging data alone.
To conduct a comparison of the thermal data with other physical property measurements, composite downhole plots were generated. The curated depths of the two hydrate nodule occurrences and three intervals of fizzing sediment observed at Site 1230 were compared to the depths assigned based on the thermal image analyses (Table T1). Although imperfect, the two depth scales are comparable, suggesting that the thermal image depths may be useful for the generation of composite downhole plots.
Resistivity, P-wave velocity, natural gamma ray (NGR) emission, and core liner temperature all illustrate increasing variability between ~80 and 165 mbsf (Fig. F8). The dissociation of hydrate with increased temperature and decreased pressure alters the core by altering the water content of surrounding sediments and creates gas expansion voids. This results in depth discrepancies between downhole log (wireline) data and shipboard measurements. However, although centimeter-scale correlation is not yet feasible, zones of identified and potential hydrate occurrence are recognizable. These correlations are more fully investigated in "Physical Properties" in the "Site 1230" chapter. Interestingly, the thermal data from Site 1226 also show a correlation with other physical property measurements (Fig. F9). This could be explained by differential warming of distinct lithologies during wireline recovery.