Abstract
New heat-flux observations made along two seismic reflection profiles on 6 Ma crust of the Costa Rica Rift flank show an inverse correlation between heat-flux and sediment thickness similar to that observed on other sedimented ridges and young ridge flanks. Extrapolation of the seafloor heat-flux values to the top of the igneous crust—justified by comparing seafloor and borehole determinations where observations are colocated—shows the surface of the crust to be of uniform temperature despite large local sediment thickness variations. This is consistent with observations made at DSDP/ODP Holes 504B and 896A where basement temperatures are observed to differ by less than 2 K, also despite contrasting sediment thicknesses which would cause a difference of more than 15 K under conditions of conductive heat transport. Efficient lateral heat exchange via vigorous crustal hydrothermal circulation is required to create the degree of uniformity inferred and observed. Permeability measurements at the two drill sites have shown that this vigorous circulation may be restricted to the uppermost 100 m of the crust. The laterally uniform temperatures in the uppermost igneous crust here and elsewhere are remarkable given the small apparent depth extent of the circulation that so efficiently distributes heat. Modeling suggests that the effective formation-scale permeability must be of the order of 10
−9 m
2, higher by several orders of magnitude than borehole packer and flow determinations, and that fluid fluxes are of the order of many tens of m year
−1. Deeper in Hole 504B, permeability previously measured is too low to permit thermally significant flow, and this is consistent with geochemical observations which have shown that deep borehole fluids retain characteristics of drilling fluid years after episodes of drilling. Paradoxically, deep temperature logs are best explained by the presence of advective heat transport, and suggest that thermally significant flow may indeed extend throughout the 1800 m section penetrated at Site 504, and cool the deep crust by many tens of K. Permeability in the deeper extrusives and sheeted dikes estimated via numerical models using the constraint of the thermal structure is >10
−15 m
2. Fractures and other zones of high permeability in both the uppermost and deeper crust must be underrepresented by the borehole. While certainly not as vigorous as the inferred flow in the uppermost crust, the deeper circulation suggested by the deep temperature logs is noteworthy in light of previous conclusions based on permeability and fluid chemistry that the deeper crustal water at this site is stagnant. Model results suggest average fluxes of up to 0.1 m year
−1, and an average time for exchange between the deeper and shallow layers of <5000 years.
