Quantitatively do the research of the sampling method and likelihood measurement in GLUE methodology

In this paper the likelihood measurement and sampling method in Generalized Likelihood Uncertainty Estimation (GLUE) methodology were studied in an application to NASH model to the Yanduhe research catchment in Yangtze River basin, China. Uncertainty ratio is advanced to compare the Latin sampling method and MC sampling method quantitatively in the application of GLUE methodology, and the likelihood measurements are researched in the paper. The results showed that the sampling method do little influence to the uncertainty, however the likelihood measurements do great infect to the research result and multi-objective likelihood function was proved to be better than single-objective likelihood function.