Natural gas hydrates are an icy solid consisting commonly of a methane molecule encased by a water lattice. Hydrates occur in vast quantities worldwide, and are found in marine and permafrost regions where their stability conditions are satisfied. Natural gas hydrates are a potential energy resource, a geohazard for drilling and submarine slope stability, and a source of methane a significant greenhouse gas. Traditionally seismic methods are used for hydrate detection since hydrates are often characterized by the bottom simulating reflector (BSR). The BSR often represents the phase change from solid hydrate above to free gas below at a depth controlled by the intercept of the hydrate stability field and the geothermal gradient. While a seismic BSR can be ubiquitous over an entire hydratebearing region, seismic methods do not indicate the gradational upper surface of the hydrate, nor do they provide a distinctive signature within the hydrate zone (Yuan et al., 2000). Furthermore, there are places where hydrates are known to exist yet exhibit no BSR, such as in the Gulf of Mexico (Sloan, 1998). Other methods for hydrate detection include resistivity measurements in well logs. Hydrates are electrical insulators and will be more resistive than the surrounding host sediment, thus providing an EM target. Although the resistivity contrast can be quite small (?2 Ohm-m, Hyndman et al, 1999) our forward modeling results (below) indicate we will be able to detect hydrate using standard electromagnetic methods. We investigate the potential of electromagnetic methods to map the extent and quantity of gas hydrates. A pilot marine EM study at Hydrate Ridge, Oregon was conducted in August, 2004. We collected an extensive data set consisting of magnetotelluric (MT), controlled source electromagnetics (CSEM), and controlled source magnetotelluric (CSMT) data. In this paper we present an outline of our experimental design, a description of our survey, and preliminary results from a subset of the CSEM data.

Figure 1. Forward Modeling study of hydrates compared to a background sediment without hydrate. Figure a (top) shows the ratio of the hydrate to the background sediment for various frequencies from 1 Hz to 35 Hz. The largest signal comes from the 35 Hz data. Figure b (below) shows the electric eld versus source-receiver range for the corresponding frequencies. The electric elds fall off quickly at the higher frequencies due to stronger attenuation at the higher frequencies. (Available in full paper)

Forward Modeling

Numerical forward modeling studies were used to design the CSEM experiment. Radial mode electric fields were modeled using Flosadottir and Constable's (1996) forward modeling code. Our model consisted of a 105 m thick hydrate layer of 2 Ohm-m sandwiched between a background sediment of 1 Ohm-m at a 45 m depth in 1200 m of seawater (0.3 Ohm-m). We found that in order to detect the hydrate anomaly (the ratio of the responses of sediments containing hydrate to the background response of sediment only) we required relatively high frequencies of about 35 Hz (Figure 1a). However, frequencies above 10 Hz attenuate very quickly (Figure 1b).

This content is only available via PDF.
You can access this article if you purchase or spend a download.