Premium
A new uncertainty analysis in the climate change impact assessment
Author(s) -
Lee JaeKyoung,
Kim YoungOh,
Kim Yongdai
Publication year - 2017
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.4957
Subject(s) - downscaling , climatology , uncertainty analysis , climate change , environmental science , projection (relational algebra) , streamflow , stage (stratigraphy) , precipitation , sensitivity analysis , uncertainty quantification , climate model , econometrics , statistics , mathematics , meteorology , geography , geology , drainage basin , paleontology , oceanography , cartography , algorithm
A majority of existing studies on uncertainties in climate change impact assessments carried out the uncertainty analysis independently at each stage without quantifying the total uncertainty and thus it was seldom possible to assess the relative contribution of each stage to the total uncertainty and also to see how the uncertainty is propagated as the stage proceeds. To overcome these shortcomings, this study proposes a simple yet new approach, which can quantify the total uncertainty as well as the incremental uncertainty at each stage. Employing the maximum entropy as an uncertainty measure, the new approach was applied to a case study that consists of two emission scenarios, four global climate model GCM scenarios, two downscaling techniques, and two hydrological models. The difference was noteworthy: in case of the water streamflow projection, the conventional approach identified the GCM stage as the largest contributor (89.34%) to the total uncertainty while this new approach concluded the emission scenario stage the largest (58.66%). In case of the precipitation projection, the downscaling stage produced the largest uncertainty indicating that the relative uncertainty contribution of each assessment stage can vary depending on the projection variable which of uncertainty is examined. The case study also compared the projection uncertainty with the natural variability that exists in the observed data and concluded that the uncertainty generated by the future climate change projection is about two times larger than that of the past natural variability.