Effects of data time-step on the accuracy of calibrated rainfall–streamflow model parameters: practical aspects of uncertainty reduction
Author(s) -
I.G. Littlewood,
Barry Croke
Publication year - 2012
Publication title -
hydrology research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.665
H-Index - 48
eISSN - 1996-9694
pISSN - 0029-1277
DOI - 10.2166/nh.2012.099
Subject(s) - streamflow , regionalisation , environmental science , scale (ratio) , hydrology (agriculture) , drainage basin , climatology , geology , geography , economic geography , cartography , geotechnical engineering
The effects of data time-step on the accuracy of calibrated parameters in a discrete-time conceptual rainfall–streamflow model are reviewed and further investigated. A quick-flow decay time constant of 19.9 hr, calibrated for the 10.6 km2 Wye at Cefn Brwyn using daily data, massively overestimates a reference value of 3.76 hr calibrated using hourly data (an inaccuracy of 16.1 hr or 429%). About 42 and 58% of the inaccuracy are accounted for by loss of information in the effective rainfall and streamflow data, respectively. A slow-flow decay time constant is inaccurate by about +111%, of which about 94 and 17 percentage points (85 and 15% of the absolute inaccuracy) are due to loss of information in the effective rainfall and streamflow data, respectively. Discrete-time rainfall–streamflow model parameter inaccuracy caused by data time-step effects is discussed in terms of its implications for parameter regionalisation (including database aspects) and catchment-scale process studies.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom