Open AccessTesting G reenfield run‐off estimation techniques using high‐resolution field observations
Author(s) -
Rodda H.J.E.,
Hawkins J.
Publication year - 2012
Publication title -
journal of flood risk management
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.049
H-Index - 36
ISSN - 1753-318X
DOI - 10.1111/j.1753-318x.2012.01156.x
Subject(s) - calibration , estimation , environmental science , flow (mathematics) , flooding (psychology) , hydrology (agriculture) , drainage , high resolution , statistics , computer science , mathematics , operations research , geography , geology , remote sensing , ecology , biology , economics , geotechnical engineering , psychology , geometry , management , psychotherapist
The current UK guidelines for sustainable drainage systems design [ D epartment of the E nvironment F ood and R ural A ffairs/ E nvironment A gency ( DEFRA / EA ), 2005; C onstruction I ndustry R esearch and I nformation A ssociation ( CIRIA ), 2007] require that the design peak flow from a new development should not exceed that which would occur under the pre‐development or G reenfield conditions. Although these guidelines are fairly recent documents, the standard methodology for estimating G reenfield flows dates back to the results of studies from the 1970s. The recommendation of the guideline method was not the result of any testing and calibration but rather the assumption that it produced G reenfield flow estimates that were in the order of what would be expected. Experience of G reenfield peak flow estimates using the approved method has shown flows to be very low compared with what would be expected where a detailed knowledge of the site hydrology was available or from observational evidence of flooding events. Recent monitoring of surface run‐off and drainage using a 1‐min temporal resolution at the R owden M oor D rainage E xperiment at N orth W yke R esearch, D evon, UK has provided an ideal dataset through which to test the current approved methodology for G reenfield flow estimation. The results show that the methods significantly underestimate the G reenfield peak flows.