DSD-INT 2017 The use of additional sources of data for hydrological modelling in the Meuse basin - Bouaziz
0 likes1,722 views
The document discusses the importance of additional data for designing a hydrological model of the Meuse River, focusing on the HBV model and various data sources such as historical timeseries and satellite-based soil moisture. It highlights different modeling approaches for water balance, soil moisture, and groundwater levels, emphasizing the need for creative data usage to refine the model structure and parameters. The conclusion encourages a shift from merely fitting parameters to understanding underlying processes.
DSD-INT 2017 The use of additional sources of data for hydrological modelling in the Meuse basin - Bouaziz
- 1. The value of additional data to design the hydrological model of the Meuse Laurène Bouaziz 24 oct 2017
- 2. Currently: HBV model • Developed for Sweden • Works well for high discharges • Models as tools
- 3. Meuse model • Models as hypothesis • How should the model look like?
- 4. Available data • Historical timeseries (P, T, Q, E) – 45 basins • Satellite based soil moisture – VanderSat – 100m resolution • Piezometer levels (ADES, PIEZ’EAU) • Landuse, elevation, geology
- 5. What I’ll show: 1 Historical series of P,E,Q & the Budyko framework Loss term 2 Satellite based soil moisture Summer and winter processes & constrain the parameter beta 3 Satellite based soil moisture & groundwater levels Different model concepts for different parts of the landscape
- 6. 1. Plotting the long term water balance Andréassian and Perrin 2012
- 7. 1. Plotting the long term water balance Andréassian and Perrin 2012 Ea = 0.97 * Ep
- 8. 1. Estimating Ea with Budyko Budyko 1974, Gentine 2012 Ea = 0.88 * Ep
- 9. 1. Consequences for the model • Deep percolation losses – Karstification of limestone, chalk • Abstractions
- 10. 2. Satellite based soil moisture • VanderSat • 100 m resolution • Almost daily 2015-2017
- 11. 2. Previous work • Van Dijk et al. used Microwave and MODIS (5x5km) • Worked good in natural rivers (monthly timestep) but no good signal in rivers in western Europe
- 12. 2. Soil moisture based rating curve
- 13. 2. Predict discharges using soil moisture Separate rating curve for summer and winter One rating curve for entire period
- 14. 2. Constraining the beta parameter 𝑅 𝑃 = 𝑆𝑢 𝑆𝑢𝑚𝑎𝑥 𝑏𝑒𝑡𝑎
- 15. 3. Groundwater and soil moisture
- 16. 3. Consequences for the model • Coupled soil moisture and groundwater schematization • Different model concept for the riparian zones
- 17. Summary 1 Historical series of P,E,Q & the Budyko framework Loss term 2 Satellite based soil moisture Summer and winter processes & constrain the parameter beta 3 Satellite based soil moisture & groundwater levels Different model concepts for different parts of the landscape
- 18. Conclusion • Additional sources of data and creative use of data may help to define the model structure and/or constrain the model parameters • Shift the focus from parameter fitting towards process understanding
- 19. Merci