Holistic versus monomeric strategies for hydrological modelling of human-modified hydrosystems

Abstract

Abstract. The modelling of human-modified basins that are inadequately measured constitutes a challenge for hydrological science. Often, models for such systems are detailed and hydraulics-based for only one part of the system while for other parts oversimplified models or rough assumptions are used. This is typically a bottom-up approach, which seeks to exploit knowledge of hydrological processes at the micro-scale at some components of the system. Also, it is a monomeric approach in two ways: first, essential interactions among system components may be poorly represented or even omitted; second, differences in the level of detail of process representation can lead to uncontrolled errors. Additionally, the calibration procedure merely accounts for the reproduction of the observed responses using typical fitting criteria. The paper aims to raise some critical issues, regarding the entire modelling approach for such hydrosystems. For this, two alternative modelling strategies are examined that reflect two modelling approaches or philosophies: a dominant bottom-up approach, which is also monomeric and, very often, based on output information, and a top-down and holistic approach based on generalized information. Critical options are examined, which codify the differences between the two strategies: the representation of surface, groundwater and water management processes, the schematization and parameterization concepts and the parameter estimation methodology. The first strategy is based on stand-alone models for surface and groundwater processes and for water management, which are employed sequentially. For each model, a different (detailed or coarse) parameterization is used, which is dictated by the hydrosystem schematization. The second strategy involves model integration for all processes, Correspondence to: I. Nalbantis (nalbant@central.ntua.gr) parsimonious parameterization and hybrid manual-automatic parameter optimization based on multiple objectives. A test case is examined in a hydrosystem in Greece with high complexities, such as extended surface-groundwater interactions, ill-defined boundaries, sinks to the sea and anthropogenic intervention with unmeasured abstractions both from surface water and aquifers. Criteria for comparison are the physical consistency of parameters, the reproduction of runoff hydrographs at multiple sites within the studied basin, the likelihood of uncontrolled model outputs, the required amount of computational effort and the performance within a stochastic simulation setting. Our work allows for investigating the deterioration of model performance in cases where no balanced attention is paid to all components of human-modified hydrosystems and the related information. Also, sources of errors are identified and their combined effect are evaluated.

Continental hydrosystem modelling: the concept of nested stream–aquifer interfaces

“…Although the number of papers concerning stream-aquifer interfaces exponentially increased over the last 15 years , they mostly focus on local scale issues, following a classic bottom-up scientific approach (Nalbantis et al, 2011). The lack of models aiming at quantifying stream-aquifer exchanges at large basins' scale was already alleged by Fleckenstein et al (2010) and Krause et al (2011).…”

Nitrogen and phosphorus transformations and balance in a pond-ditch circulation system for rural polluted water treatment

“…Because of progressively severe problems with water resources, strengthening research on water resources management and hydrological forecast has become increasingly urgent (Nalbantis et al, 2011). The purpose and focus of hydrology research have gradually progressed and shifted from simple flood prediction to comprehensive water resource management that requires more information on detailed hydrological processes (Jarsjo et al, 2012).…”

How can expert knowledge increase the realism of conceptual hydrological models? A case study in the Swiss Pre-Alps

“…As a further development of FUSE, Clark et al (2015) developed the SUMMA approach to provide a framework for both modellers and experimentalists to test alternative model discretisations, parameterisations, and numerical schemes. Nalbantis et al (2011) compared a bottom-up and a top-down modelling approach with a focus on catchments with high 5 human impact.…”

“…In contrast, top-down models are conceptual models. With regard to the level of modelling detail, Nalbantis et al (2011) linked monometric approaches, where some components are examined in detail and other ones are only roughly described, to the bottom-up philosophy and the holistic approach, when all components are 5 modelled with the same degree of detail, to the top-down one. Sivapalan et al (2003), in contrast, classify approaches according to the scale considered: if the modelling is performed first at the small scale of e.g.…”