Observed correlation between the depth to base and top of gas hydrate occurrence from review of global drilling data

Abstract

A global inventory of data from gas hydrate drilling expeditions is used to develop relationships between the base of structure I gas hydrate stability, top of gas hydrate occurrence, sulfate‐methane transition depth, pressure (water depth), and geothermal gradients. The motivation of this study is to provide first‐order estimates of the top of gas hydrate occurrence and associated thickness of the gas hydrate occurrence zone for climate‐change scenarios, global carbon budget analyses, or gas hydrate resource assessments. Results from publically available drilling campaigns (21 expeditions and 52 drill sites) off Cascadia, Blake Ridge, India, Korea, South China Sea, Japan, Chile, Peru, Costa Rica, Gulf of Mexico, and Borneo reveal a first‐order linear relationship between the depth to the top and base of gas hydrate occurrence. The reason for these nearly linear relationships is believed to be the strong pressure and temperature dependence of methane solubility in the absence of large difference in thermal gradients between the various sites assessed. In addition, a statistically robust relationship was defined between the thickness of the gas hydrate occurrence zone and the base of gas hydrate stability (in meters below seafloor). The relationship developed is able to predict the depth of the top of gas hydrate occurrence zone using observed depths of the base of gas hydrate stability within less than 50 m at most locations examined in this study. No clear correlation of the depth to the top and base of gas hydrate occurrences with geothermal gradient and sulfate‐methane transition depth was identified.

The Volume of Gas Hydrate‐Bound Gas in the Northern Gulf of Mexico

“…The parameter f was directly based on gas hydrate occurrences observed in 124 wells (Majumdar et al, 2017). Similar observation of gas hydrate occupying only a fraction the HSZ has also been modeled and observed in other basins around the world (e.g., Clennell et al, 1999;Collett et al, 2014;Paganoni et al, 2016;Riedel & Collett, 2017;Tsuji et al, 2009;Wang et al, 2014;Xu & Ruppel, 1999). Thus, this additional parameter in our model likely increased the accuracy of our result.…”

Mechanisms of Methane Hydrate Formation in Geological Systems

“…The top of the HSZ is usually located in the water column above the seafloor in marine environments (Figure a) and as shallow as 100–200 m below the ground surface in terrestrial systems (Figure b). In fact, the top of the actual gas hydrate occurrence can be tens to hundreds of meters below the top of the HSZ because the ocean water and the shallow sediments are undersaturated in methane (Riedel & Collett, ). The methane hydrate stability zone can also be illustrated in a pressure versus temperature plot as the zone where the in situ state (red dashed line) is to the left of the phase boundary (green line; Figures c and d).…”

Global Estimates of Biogenic Methane Production in Marine Sediments Using Machine Learning and Deterministic Modeling

“…However, qualitatively many of our predictions and subsequent ensembles yield little to no methane hydrate in the subsurface. While comparing our results/input to observed values (herein referred to as point observations), we note our inputs and results (herein referred to as cell averages) are representative of ∼10 × 10‐km areas, whereas point observations from Riedel and Collett (2017) are very discrete sampled points (scale of cm). Our cell averages represent areas and not necessarily discrete points; therefore, variability is inherent as our estimates actually represent a 10 × 10 km area average.…”

“…To further quantitatively validate our estimated BHSZ, we compare our estimated HSZ thicknesses to observed drilling sites collated by Riedel and Collett (2017) shown in Table 2. Site locations are presented in their regional context within the Supplemental Information (Figures S1-S2 in Supporting Information S1).…”

“…Temperature within the subsurface was determined by calculating a geothermal gradient derived from predicted heat flow at the seafloor using observations primarily from the Global Heat Flow Compilation Group (2013). We assimilated several new observations from Hornbach et al (2020) and Riedel and Collett (2017). Using these observations, we predicted heat flow and uncertainty at the seafloor using a KNR machine learning algorithm following the workflow outlined in Lee et al (2019).…”