This article, written by Technology Editor Dennis Denney, contains highlights of paper OTC 17168, "Hydrocarbon Exploration Using Marine EM Techniques," by S. Constable, Scripps Inst. of Oceanography, U. of California San Diego, prepared for the 2005 Offshore Technology Conference, Houston, 2-5 May. Recent interest in marine electromagnetic (EM) methods has been driven, in large part, by the challenges in deepwater exploration and associated efforts to reduce risk through the acquisition of additional data streams. The catalyst for the latest developments in marine EM technology was the problem of subsalt exploration in the Gulf of Mexico. Investigations were made of the use of controlled-source EM (CSEM) methods for direct hydrocarbon detection. Field trials were carried out offshore Angola. Methodology Fig. 1 shows the basic methods of marine EM studies. Seafloor recorders make time-series measurements of the electric and magnetic fields at discrete locations. The ionospheric and magnetospheric marine magnetotelluric (MT) source fields propagate into the sea, and a seafloor MT response can be processed from the data much like a land MT impedance, with two significant differences. Attenuation in the conductive seawater layer dramatically reduces the electric- and magnetic-field energies at higher frequencies. Although MT responses on land can be processed routinely to several hundred hertz, in 1000 m of seawater it is essentially impossible to obtain MT data at frequencies higher than 1 Hz. Until the development of a broadband seafloor instrument, there were few attempts to collect seafloor MT data at frequencies greater than 0.01 Hz. Attenuation is less in shallower water (signal strength decreases with water depth squared), but the MT source fields decrease in magnitude with frequency, offsetting the shallow-water improvement. The seafloor 1D MT response is the same as would be observed on land over the same geological structure without the ocean layer. However, the magnetic field is affected significantly by seabed conductivity, even in the 1D case (in stark contrast to land, where the magnetic field is essentially uniform, especially away from lateral contrasts in conductivity). A hybrid MT response obtained with the seafloor electric data and nearby land magnetics (equivalent to a sea-surface magnetic record) was used in the early development of the method to finesse the effects of motional noise in the seafloor magnetic data). This response illustrates the effect of seafloor conductivity on the magnetic field. In the case of 2D and 3D seafloor conductivity structure, a vertical electric field is possible in the seawater layer near the seafloor, although this effect is usually very small.