An assessment of simulated runoff from global models

Giuntoli, Ignazio (2017). An assessment of simulated runoff from global models. University of Birmingham. Ph.D.

PDF - Accepted Version

Download (10MB)


This thesis assesses long-term runoff projections from global multi-model ensembles used in hydrological impact studies. Firstly, the study investigates global-scale changes in frequency of high and low flow days towards the end of the current century, quantifying the relative contribution to uncertainty from global climate (GCMs) and global impact models (GIMs). Results show increases in high flows for northern latitudes and in low flows for several hotspots worldwide. Overall, GCMs provide the largest uncertainty; but GIMs are the greatest source of uncertainty in snow-dominated regions. Secondly, the ability of a set of GIMs to reproduce observed runoff is evaluated at the regional scale, indicating that GIMs capture well trends in low, medium, and high flows, but differ from observations with respect to medium and high flows timing. Thirdly, the contribution to uncertainty from GCMs, GIMs, Representative Concentration Pathways (RCPs), and internal variability is quantified for transient runoff until 2099. Over the USA, GCMs and GIMs are responsible for the largest uncertainty. Efforts to improve runoff projections should thus focus on GCMs and GIMs. In particular, GIMs should be evaluated in the region of study, so that models reproducing unrealistic runoff can be excluded, potentially yielding greater confidence in ensemble projections.

Type of Work: Thesis (Doctorates > Ph.D.)
Award Type: Doctorates > Ph.D.
College/Faculty: Colleges (2008 onwards) > College of Life & Environmental Sciences
School or Department: School of Geography, Earth and Environmental Sciences
Funders: Natural Environment Research Council, Other
Other Funders: The University of Birmingham
Subjects: G Geography. Anthropology. Recreation > G Geography (General)


Request a Correction Request a Correction
View Item View Item


Downloads per month over past year