Javascript verkar inte påslaget? - Vissa delar av Lunds universitets webbplats fungerar inte optimalt utan javascript, kontrollera din webbläsares inställningar.
Du är här

Monthly runoff simulation: Comparing and combining conceptual and neural network models

Publiceringsår: 2006
Språk: Engelska
Sidor: 344-363
Publikation/Tidskrift/Serie: Journal of Hydrology
Volym: 321
Nummer: 1-4
Dokumenttyp: Artikel i tidskrift
Förlag: Elsevier


Runoff estimation is of high importance for many practical engineering applications so that, e.g. power production, dam safety and water supply can be ensured. The methods and time step relevant for runoff simulations vary depending on the location and the application. Long-term runoff simulation for Scandinavia is of high importance as its hydropower production is affected by climate variability, which strongly influences winter temperature and precipitation. This work investigates the possibility of modelling monthly runoff for two Norwegian river basins. Two methodologies-artificial neural networks (NN) and conceptual runoff modelling (CM)-are compared and NN offer the best estimations of monthly runoff for both tested basins with R-2 = 0.82 and 0.71, respectively. The combination of NN and CM by using snow accumulation and the soil moisture calculated by the CM as input to the NN proved to be an excellent alternative to perform high quality monthly runoff simulations and improved the simulations skill for both basins (R-2=0.86 and 0.75, respectively).


  • Water Engineering
  • monthly runoff
  • combination
  • modelling
  • conceptual
  • hydrological modelling
  • artificial neural networks


  • ISSN: 0022-1694

Box 117, 221 00 LUND
Telefon 046-222 00 00 (växel)
Telefax 046-222 47 20
lu [at] lu [dot] se

Fakturaadress: Box 188, 221 00 LUND
Organisationsnummer: 202100-3211
Om webbplatsen