Impact of rainfall estimation uncertainty on streamflow estimations for catchments Wye and Tyne in the United Kingdom

The main sources of uncertainty in rainfall‐runoff models are found in the input variables where the largest impact on estimated streamflow is generally given by the catchment rainfall. Hence, by quantifying the uncertainty of catchment rainfall, it is possible to provide an estimate, albeit conserv...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of climatology Vol. 29; no. 1; pp. 79 - 86
Main Authors Ekström, Marie, Jones, Philip D.
Format Journal Article
LanguageEnglish
Published Chichester, UK John Wiley & Sons, Ltd 01.01.2009
Wiley
Subjects
Earth sciences
Earth, ocean, space
Exact sciences and technology
External geophysics
Hydrology
Hydrology. Hydrogeology
Meteorology
rainfall‐runoff modelling
simulated rainfall
spatial uncertainty
United Kingdom
Water in the atmosphere (humidity, clouds, evaporation, precipitation)
United Kingdom
Great Britain
Europe
Western Europe
hydrology
atmospheric precipitation
river discharge
United Kingdom
digital simulation
spatial uncertainty
Hydroclimatology
Rainfall rate
drainage basins
rainfall-runoff modelling
Risk management
uncertainties
Computing method
simulated rainfall
streamflow
Online AccessGet full text

Cover

More Information
Summary:The main sources of uncertainty in rainfall‐runoff models are found in the input variables where the largest impact on estimated streamflow is generally given by the catchment rainfall. Hence, by quantifying the uncertainty of catchment rainfall, it is possible to provide an estimate, albeit conservative, of streamflow estimation uncertainty. In this article, 42 series of catchment average rainfall were used to estimate monthly streamflow for two river catchments in the United Kingdom over a period of 8 years, these were as follows: a composite series based on weighted rain gauge series, a series based on interpolated gridded rainfall data and an ensemble of 40 series, based on simulated rainfall grids conditioned on observed rainfall statistics. This study shows that, on average, the uncertainty in monthly streamflow, as given from the mean range of monthly simulated data, were approximately 28 and 42% of the long‐term mean for the two catchments respectively. By quantifying uncertainty in streamflow due to catchment rainfall uncertainty it was possible to identify contributions of other factors to the total streamflow estimation uncertainty. For example, streamflow uncertainty was, in this experiment, shown to respond also to the absolute magnitude of the streamflow, a consequence of the model being optimized for low flows. Copyright © 2008 Royal Meteorological Society
ISSN:0899-8418
1097-0088
DOI:10.1002/joc.1665