[en] The Severe Accident Risk Reduction Program (SARRP) has completed a rebaselining of the risks to the public from a boiling water reactor with a Mark I containment (Peach Bottom, Unit 2). Emphasis was placed on determining the magnitude and character of the uncertainties, rather than focusing on a point estimate. The risk-reduction potential of a set of proposed safety option backfits was also studied, and their costs and benefits were also evaluated. It was found that the risks from internal events are generally low relative to previous studies; for example, most of the uncertainty range is lower than the point estimate of risk for the Peach Bottom plant in the Reactor Safety Study (RSS). However, certain unresolved issues cause the top of the uncertainty band to appear at a level that is comparable with the RSS point estimate. These issues include the modeling of the common-mode failures for the dc power system, the likelihood of offsite power recovery versus time during a station blackout, the probability of drywell failure resulting from meltthrough of the drywell shell, the magnitude of the fission product releases during core-concrete interactions, and the decontamination effectiveness of the reactor enclosure building. Most of the postulated safety options do not appear to be cost effective, although some based on changes to procedures or inexpensive hardware additions may be marginally cost effective. This draft for comment of the SARRP report for Peach Bottom does not include detailed technical appendices, which are still in preparation. The appendices will be issued under separate cover when completed. This work supports the Nuclear Regulatory Commission's assessment of severe accidents in NUREG-1150. (author)