This paper describes observations of the travel times from a series of 5-ton shots fired by the United States Geological Survey in Lake Superior during July 1966 (Project Early Rise). The observations were made along three profiles: (1) the Wichita profile, extending 2250 km from Lake Superior to the Texas-Mexico border, (2) the Little Rock profile, extending 1650 km from Lake Superior to the Arkansas-Louisiana border and (3) an arc profile at a radius of about 1200 km from the shot point from West Virginia to the Kansas-Nebraska border. These records and others of the nuclear explosion GREELEY at points in the central United States, taken in conjunction with the times for GNOME, have been analyzed and model calculations made.
The analysis of the Early Rise observations showed that any low-velocity layer beneath the central United States was either very thin or at a depth of more than 150 km below the surface. Times from the Nevada Test Site were fitted by a model with a marked low-velocity layer at a depth of 104 km. The travel times from GREELEY to the central United States were consistent with a model having the Nevada model upper layers at one end and the Early Rise model at the other. The deeper structure has velocity discontinuities (or steep velocity gradients) between 362 and 382 km and between 623 and 645 km as was proposed by Niazi and Anderson (1965). The Greeley observations between 1200 and 2200 km suggest that the low-velocity layer thins out from about 106°W over a distance of 600 km.