Review of scenario selection and downscaling methods for the assessment of climate change impacts on hydrology in the United States pacific northwest

摘要: This paper reviews methods that have been used to evaluate global climate simulations and downscale scenarios for the assessment of impacts on hydrologic systems in Pacific Northwest, USA. The approach described has developed facilitate integrated research support regional resource management. Global model are evaluated selected based historic 20 th Century simulations. A statistical downscaling method is then applied produce a data set. To use projections assessment, additional mapping may be generate synthetic station time series. Finally, results presented from indicate important differences response what captured by models downscaling. 1. Introduction Some most anticipated change expressed through processes such as streamflow, snowpack, flooding. Modeling these requires high- resolution future temperature precipitation. science at scales quite advanced typically downscaled fine 10-50 km grids or locations. While there remains significant done fully understand dynamics bolster confidence scenarios, current modeling adequate many applications hydrology. principal challenge linking existing computational tools institutional mechanisms within an assessment. For example, under change, system impact complicated constantly shifting underlying trends large year-to-year variability (Arnell, 1996). analysis water their reliability, yield, specific event frequency, generally assumes static state can statistically using series events depends observed record past estimate probability events. assumed stationary so all equally probable probabilities carry into future. Typically, transient multiple projected emissions models. this number various it does not correspond well Climate Impacts Group (CIG) University Washington United States. focuses four diverse yet connected natural Northwest (fresh water, forests, salmon coasts) socioeconomic and/or political associated with each. Hydrologic central sectors; thus, forms basis quantitative analyses. Many approaches we empirical corrections simulated data. These relationship between statistics parameter simulation equivalent conditions. correct In its simplest form, could simple perturbation bias. quantile mapping, however, full distribution taken account. given present-day conditions location 5°C too cold compared observations. climate, one would add values bias simply lapse-rate correction unresolved topography stem deficiency