Dealing With Ion LET Uncertainties: An Application of Generalized Linear Models

Abstract

Although most single-event effect (SEE) rate estimation methods presume a fit to SEE cross section versus linear energy transfer (LET), fitting SEE data is challenging because the data are not compatible with the assumptions of many common fitting techniques (e.g., linear regression). The difficulty of fitting such data is compounded when the LET of the ion responsible for an SEE is uncertain. We modify a generalized linear model, SEE data fitting methodology, to accommodate uncertain LET and apply the method to the problem of backside heavy-ion SEE testing to demonstrate its utility, explore the dependence of systematic errors that arise from improper treatment of LET uncertainty, and develop guidelines for minimizing such systematic errors when proper treatment is not possible. Additional applications are suggested and assessed for suitability to treatment by the model.