Currently three imaging spectrometer architectures, tunable filter, dispersive, and Fourier transform, are viable for imaging the universe in three dimensions. There are domains of greatest utility for each of these architectures. The optimum choice among the various alternative architectures is dependent on the nature of the desired observations, the maturity of the relevant technology, and the character of the backgrounds. The domain appropriate for each of the alternatives is delineated; both for instruments having ideal performance as well as for instrumentation based on currently available technology. The environment and science objectives for the Next Generation Space Telescope will be used as a specific representative case to provide a basis for comparison of the various alternatives.

A robust bearing estimation process for 3-component stations has been developed and explored. The method, called SEEC for Search, Estimate, Evaluate and Correct, intelligently exploits the in- herent information in the arrival at every step of the process to achieve near-optimal results. In particular, the approach uses a consistent framework to define the optimal time-frequency windows on which to make estimates, to make the bearing estimates themselves, to construct metrics helpful in choosing the better estimates or admitting that the bearing is immeasurable, andjinally to apply bias corrections when calibration information is available to yield a single final estimate. The method was applied to a small but challenging set of events in a seismically active region. The method demonstrated remarkable utility by providing better estimates and insights than previously available. Various monitoring implications are noted fiom these findings.

Fermilab is in the midst of a program to raise the luminosity in the Tevatron proton-antiproton collider by at least a factor of five above the currently achieved level of 1 .6x10{sup 31} cm{sup -2}sec{sup -l} Components of this program include the construction of a new synchrotron, the Main Injector, a new antiproton storage ring, the Recycler, and a variety of improvements to the existing Antiproton Source and Tevatron. Commissioning of these components will be underway in early 1999 with the next Tevatron collider run scheduled to start in April 2000. Initial commissioning experience with these systems will be described, along with developments underway to support further Tevatron luminosity enhancements over the coming decade.

Long term performance requirements for a geologic repository for spent nuclear fuel and high-level waste are based on assumptions concerning water use and subsequent deaths from cancer due to ingesting water contaminated with radio isotopes ten thousand years in the future. This paper argues that the assumptions underlying these requirements are faulty for a number of reasons. First, in light of the inevitable technological progress, including efficient desalination of water, over the next ten thousand years, it is inconceivable that a future society would drill for water near a repository. Second, even today we would not use water without testing its purity. Third, today many types of cancer are curable, and with the rapid progress in medical technology in general, and the prevention and treatment of cancer in particular, it is improbable that cancer caused by ingesting contaminated water will be a sign&ant killer in the far future. This paper reviews the performance requirements for geological repositories and comments on the difficulties in proving compliance in the face of inherent uncertainties. The already tiny long-term risk posed by a geologic repository is presented and contrasted with contemporary every day risks. A number of examples of technological progress, including cancer treatments, …