S12: Developing new hydrological models - towards merging top-down and bottom-up approaches

Organised by: Barry Croke, David Post, Ian Littlewood, John Pomeroy

The Prediction in Ungauged Basins (PUB) Initiative (http://cee.uiuc.edu/research/pub/) is a major international program of the IAHS (http://www.cig.ensmp.fr/~iahs/) aimed at reducing the uncertainty in predicting water and contaminant fluxes at ungauged sites. While there has already been considerable research into predicting fluxes at ungauged sites, results have been typically poor due to limitations in both models and data used.

The focus of this session is on the development of new models (PUB science theme 6). Contributions addressing this problem from the top-down and bottom-up approaches (or a combination of the two) are strongly encouraged. This includes developments based on the analysis of existing models or development of new models from theory and/or field data. Contributions related to the development of new datasets (e.g. using remote sensing techniques) and techniques for estimating parameter values for ungauged sites are also welcome. As the aim of PUB is to reduce predictive uncertainty, all authors are strongly encouraged to discuss the uncertainty in their results, techniques, models or data.


To discuss current and potential future work in development of new models for prediction in ungauged basins.


The problem in predicting flows at ungauged sites is four-fold:
  1. identifying a suitable model structure for a particular catchment
  2. determining key attributes that drive the hydrologic response of a catchment
  3. estimating parameter values for the chosen model, based on the selected attributes
  4. what is the uncertainty in the predictions, and what are the main contributors to this uncertainty

This session is targeted at addressing (at least partially) the first of these, with a focus on development of improved models through a merging of the top-down and bottom-up approaches.


The primary question is:

Can the understanding of catchment-scale behaviour represented in top-down models be combined with the process understanding built into bottom-up models to provide a reduction in predictive uncertainty for ungauged basins?

Contact: Barry Croke, barry.croke@anu.edu.au


Part 1, Tuesday 15:45 - 17:30

15:45. Conceptual modelling of individual HRU’s as a trade-off between bottom-up and top-down modelling, a case study, Rudi van den Bos

16:00. The Use of Inductive and Deductive Reasoning to Model Snowmelt Runoff from Northern Mountain Catchments, Pablo Dornes

16:15. Modeling Soil-Water Dynamics for Diverse Environmental Needs, Pierre Martin

16:30. Comparison of Top-Down and Bottom-Up Models for Simulation of Water Balance as affected by Seasonality, Vegetation Type and Spatial Land Use, Enli Wang

16:45. Evaluation of a physically based distributed hydrological model, BTOPMC, for different physiographic zones of Nepal, Sangam Shrestha

17:00. Uncertainty analysis associated with Rainfall Spatial Distribution in an Experimental Semiarid Watershed, Northeastern Brazil, Lucio Flavio Ferreira Moreira

17:15. Hydrological characterisation of four Brazilian catchments using a simple rainfall-streamflow model, Ian Littlewood

Part 2, Wednesday 9:45 - 11:00

9:45. Consistency versus Optimality in Environmental Model Identification under Uncertainty, Thorsten Wagener

10:00. Preliminary Indications of the Transferability of IHACRES Model Parameters in Mountainous Rainfall Driven Rivers, Paul Whitfield

10:15. River Flow Simulation within Ungauged Catchments in Lebanon using a semi-distributed rainfall-runoff model, Antoine Hreiche

10:30. Regionalization of IHACRES Model Parameters for In-tegrated Assessment across the Lake Erie, northern Ohio USA basin, Richard Anderson

10:45. Development of a National, Landuse-Based Water Balance Model for Australia, Wendy Welsh

Part 3, Wednesday 11:15 - 12:30

11:15. Model averaging, equifinality and uncertainty estimation in the modelling of ungauged catchments. Julian Reichl, Francis Chiew, Andrew Western

11:30. Rainfall–streamflow–air temperature datasets (and catchment information) available internationally to assist with PUB Decade top-down modelling, Barry Croke

11:45. Deriving a spatially-explicit hillslope sediment delivery ratio model based on the travel time of water across a hillslope, David Post

12:00. Questions and general discussion