Comment on “Upper crustal structure as a function of plate age” by Robert Houtz and John Ewing

Evidence for hydrothermal circulation in the upper oceanic crust on ridge flanks is derived from heat flow patterns [e.g., Anderson and Hobart, 1976; Langseth et al., 1988] and from the occurrence of low-temperature alteration products at numerous Deep Sea Drilling Project (DSDP) and Ocean Drilling Program (ODP) sites [e.g., Alt et al., 1986].The primary evidence that hydrothermal alteration is an evolutionary process that affects the seismic properties of the uppermost basement over millions of years was presented by Houtz and Ewing [1976] and Houtz [1976].Based on slope/intercept interpretations of more than 200 air gun/sonobuoy records from the Atlantic and Pacific, Houtz and Ewing [1976] proposed that (1) the seismic structure of layer 2 consists of three sublayers 2A, 2B, and 2C having p wave velocities of 3.6+ 0.4, 5.2+ 0.4, and 6.1 + 0.2 km/s, respectively; (2) layer 2A velocities increase with age over tens of millions of years until they cannot be distinguished from velocities characteristic of layer 2B; and (3) layer 2A apparently thins with age, disappearing when velocities in layer 2A approximate those in layer 2B after about 40 m.y.Purdy and Ewing [1986] have pointed out that the Houtz and Ewing data set is the only clear evidence for a systematic, age-dependent variation of crustal structure.Further evidence for agedependent upper crustal velocities is provided by highresolution studies on ridge crests and borehole measurements in older crust (see Jacobson [1992] for a review).For that reason, and because the apparent evolution of layer 2A is thought to be a consequence of hydrothermal alteration, the conclusions reached by Houtz and Ewing continue to be cited by numerous investigators [e.g., Salisbury et al., 1988; Carlson and Herrick, 1990; Wilkens et al., 1991, Jacobson, 1992].Though Houtz and Ewing were quite certain that layer 2A velocities increase with age, they were ambivalent regarding its apparent thinning.They discussed thinning of layer 2A at length, but they noted in the abstract (and in the paper by Ewing and Houtz, [1979]) that the apparent thinning of layer 2A is likely to be a consequence of its increasing velocity.This ambiguity has never been resolved.The primary purpose of this comment is to suggest that, in fact, the thinning of layer 2A an artifact of the interpretation.In our analysis of the problem we accept the velocities reported by Houtz and Ewing [1976], and we consider only slope/intercept interpretations that approximate the seismic structure as a sequence of homogenous layers.These assumptions are comparatively primitive.Ewing and Houtz [1979] noted, and other studies have shown [e.g., Whitmarsh, 1978; Spudich and Orcutt, 1980; Vera et al., 1990], that gradient models are more realistic than simple layer models.