FIELD AND MODELING CONSTRAINTS ON THE THERMAL HISTORY OF A SWARM OF COLUMBIA RIVER BASALT DIKES, CORNUCOPIA STOCK, NE OREGON

Mapping of a swarm of 35 Columbia River Basalt dike segments that cut tonalites and trondhjemites of the Cornucopia Stock in the Wallowa Mountains of northeastern Oregon provides field constraints for a thermal model of diking. We have mapped dike geometry, distribution, and thermal features such as cooling joints, quenched versus eroded dike margins, grain size distribution across dikes, and presence or absence of country rock melt adjacent to dike margins or in wallrock xenoliths or screens trapped within dikes. A two-dimensional thermal model was developed for parameterized trondhjemitic and basaltic melt fraction versus temperature relationships. To insure that phase change and temperature change are consistent with each other, a predictor-corrector method was employed (Voller and Swaminathan, 1991) and has been validated through comparison with canonical conduction results. Sidewalls have reflecting thermal boundaries to mimic a larger section of the crust and the resolution of the model is 1 m².

The results of the model suggest that simple injection and cooling of a 10-m-thick dike cannot induce wallrock melting, but that the same volume of basalt injected incrementally can generate significant wallrock melt. The volume of country rock melt correlates with the frequency of dike intrusions. We have also modeled the injection and cooling of two cross-cutting dikes, such as observed in the Cornucopia swarm, in order to place constraints on the timing of sequential dike injection and cooling. A minimum time of 120 days between injections is required to allow quenching of the second dike against the first dike; this case generates about 60% melt in the wallrock hosted in the 15 degree angle between the dikes. If more than 1200 days elapse between injections, no melt is generated in the wallrock. Comparison to field data, where 5-15% melt is found at the dike intersection, suggests 2-3 years elapsed between dike injections.