Building the future by doing more together

HIDRIA - A multi-stage approach for addressing input data uncertainties in process-based rainfall-runoff modelling for small forested catchments upstream of the Ria de Aveiro
Coordinator - Jan Jacob Keizer
Programme - PTDC/CTE-GEX/71651/2006
Execution dates - 2009-04-01 - 2012-11-30 (44 Months)
Funding Entity - FCT
Funding for CESAM - 76746 €
Total Funding - 80000 €
Proponent Institution - Universidade de Aveiro
Participating Institutions
Escola Superior Agrária de Coimbra (ESAC/IPC)

During the last decade or so, hydrological and soil erosion modelling have seen major advances with the appearance of increasingly process-based and spatially-distributed models, like e.g. LISEM (LImburg Soil Erosion Model; and MEFIDIS ("Modelo de Erosão Físico e Distribuído" (Physical and Distributed Erosion Model); Whilst tools par excelence for testing knowledge of the physical processes underlying streamflow and erosion response as well as for projecting impacts of climate and/or land-use change scenarios, these models have elevated - parameterization and validation - input requirements that are not easily met with the required spatio-temporal definition.

Such constraints in available data and information also apply to, for example, the four small catchments in the Serra do Caramulo of nort-central Portugal that have been equipped, by the UAveiro team, with hydrometric stations since the mid-1980's and that, as such, assume an unique position amongst experimental basins in Portugal. These four catchments represent two contrasting land-cover situations, being predominatly covered by either Maritime Pine stands or commercial eucalypt plantations - i.e. the two forest types that typically dominate the landscape of the central Portuguese Serras.

Following-up on over two decades of hydrological research along more classical lines, the main aim of the present proposal is to advance event-based hydrological modelling for the above-mentioned, small forested catchments in central Portugal, with LISEM and MEFIDIS being used as reference models. The currently existing constraints with respect model input data and the resulting model prediction uncertainties are explicitly addressed through a novel, multi-stage approach. This approach envisages that initial model results will guide the gathering of model input data that are lacking or incomplete, on the one hand, and, on the other, are of significant import to the model output (or, in other words, induce noticeable model sensitivity). The collected additional data, in turn, are envisaged to help reduce the uncertainties of subsequential model runs (e.g. by allowing to eliminate one or more of the model parameters sets that, in the prior stage, presented elevated equifinality). The approach further foresees in verification of spatially explicit model predictons by field measurements, for example of streamflow within one of the subcatchments and of overland flow or throughfall in - in terms of predicted values - contrasting forest types or slope angles.

The other two key elements of the present proposal comprise, paraphrasing Beven (2001) from his book entitled "Rainfall-runoff modelling - the primer", the "ultimate test" for LISEM and MEFIDIS - i.e. its application to catchments that are treated as or, in other words, ungauged as lacking streamflow data for calibration purposes - as well as their benchmarking against two or three simpler models - with Unit-Hydrogram-based model as the "model to beat".

By assessing the applicability of detailed process-based models for hydrological modeling of headwater catchments upstream of the Ria de Aveiro, HIDRIA will provide hydrologists and watershed managers with relevant information for the analysis and management of available water resources and flood risks, in particular also with respect to the associated uncertainties due to constraints in widely available physical-environmental data as well as a general lack of gauged catchments.

Members on this project

Celeste Coelho
Jan Jacob Keizer
Principal researcher


CESAM Funding: