3 NONHOMOGENEOUS POISSON MODELS FOR THE PROBABILITY OF BASALTIC VOLCANISM - APPLICATION TO THE YUCCA MOUNTAIN REGION, NEVADA

The distribution and timing of areal basaltic volcanism are modeled us ing three nonhomogeneous methods: spatio-temporal nearest neighbor, ke rnel, and nearest-neighbor kernel. These models give nonparametric est imates of spatial or spatiotemporal recurrence rate based on the posit ions and ages of cinder cones and related vent structures and can acco unt for migration and shifts in locus, volcano clustering, and develop ment of regional vent alignments. The three methods are advantageous b ecause (I) recurrence rate and probability maps can be made, facilitat ing comparison with other geological information; (2) the need to defi ne areas or zones of volcanic activity, required in homogeneous approa ches, is eliminated; and (3) the impact of uncertainty in the timing a nd distribution of individual events is particularly easy to assess. T he models are applied to the Yucca Mountain region (YMR), Nevada, the site of a proposed high-level radioactive waste repository. Applicatio n of the Hopkins F test, Clark-Evans test, and K function indicates vo lcanoes cluster in the YMR at the >95% confidence level. Weighted-cent roid cluster analysis indicates that Plio-Quaternary volcanoes are dis tributed in four clusters: three of these clusters include cinder cone s formed <1 Ma. Probability of disruption within the 8 km(2) area of t he proposed repository by formation of a new basaltic vent is calculat ed to be between 1 x 10(-4) and 5 x 10 in 10(4) years (the kernel and nearest-neighbor kernel methods give a maximum probability of 5 x 10(- 4) in 10(4) years), assuming regional recurrence rates of 5-10 volcano es/m.y. An additional finding, illustrating the strength of nonhomogen eous methods, is that maps of the probability of volcanic eruption for the YMR indicate the proposed repository lies on a steep probability gradient: volcanism recurrence rate varies by more than 2 orders of ma gnitude within 20 km. Insight into this spatial scale of probability v ariation is a distinct benefit of application of these methods to haza rd analysis in areal volcanic fields.