Answers are grounded in this paper and cite their sources.
A hydrological model can only be as reliable as the observed data against which its simulations are evaluated. The extended Generalised Likelihood Uncertainty Estimation (GLUE) methodology thus requires that limits of acceptability be defined, typically based on uncertainties in the stage–discharge rating curve, prior to any model runs and calibration against discharge data. This study proposes the use of an ensemble of rainfall–runoff ratio-based empirical benchmarks for constructing alternative behavioural limits that reflect the impact of errors in both precipitation and discharge data. The benchmarking exercise further allows for quantifying the time-varying consistency of discharge observations with long-term mass-balance expectations as a proxy for relative information content, which can later be explicitly accounted for in likelihood computation. The proposed approach is demonstrated for a wet, baseflow-dominated Philippine catchment, gauged at a station where a lowland alluvial river is prone to non-stationary rating shifts. Nearly half of the analysed streamflow record was classified as low-consistency, with such observations occurring disproportionately during or shortly after high-rainfall conditions, while high-consistency observations were associated with more quiescent periods. The derived limits widened with increasing flow magnitude and decreasing consistency. For intermediate- and high-consistency observations, this widening closely tracked probabilistic rating-curve uncertainty, but the relationship did not hold for the least informative observations, which were most susceptible to predominantly epistemic uncertainties. The limits of acceptability obtained are used, in the companion paper in this two-part series, as model-independent rejection criteria for testing rainfall–runoff models as multiple hypotheses of catchment functioning.
No takes yet. Share an insight, caveat, or question.
Cornelio et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: