Source: INMM Conference Paper
This paper considers the challenges and policy choices associated with verifying a North Korean declaration about its past nuclear activities. In terms of the plutonium programme, there are two key choices facing policy makers. First, in light of North Korea’s reported statement that its October 2006 test involved only 2 kg of plutonium, is the IAEA’s standard definition of a significant quantity appropriate? Second, using open source information, this paper argues that the uncertainty in the material unaccounted for, σ(MUF), will be between 5 and 10 kg, depending on the degree of cooperation afforded by North Korea. In order to ensure that verification produces a conclusive answer, the magnitude of σ(MUF) will necessitate some combination of decreasing the detection probability and increasing the false alarm rate from their standard IAEA values, or adopting entirely different criteria against which to assess North Korean compliance. The paper also includes a discussion of the process for verifying the absence or existence of a clandestine uranium enrichment programme by formulating it as a Bayesian inference problem. This framework, together with an analysis of the politics of inspections at undeclared locations suggests that, unless detailed intelligence pinpointing the location of a clandestine centrifuge facility is available, inspection rights at undeclared facilities may be of limited utility.