Work

Constraining a Tracer‐Aided Hydrological Model With Multi‐Objective Calibration and Young Water Fractions: Toward Improved Process‐Based Simulations

Year: 2025
Type: article
Abstract: ABSTRACT Tracer‐aided hydrological models are powerful tools for revealing hydrological processes of a watershed of interest, but their performance is often limited by parameter uncertainty. This study aims to develop a transferable, process‐consistent parameter‐constraint framework by combining multi‐objective calibration strategies with observed young water fraction ( F obs ) constraints within the Two‐Reservoir StorAge Selection (TRSAS) model. The model was applied to a humid hilly watershed in southern China to assess how weighting different observational targets (i.e., discharge, streamflow δ 18 O, soil water δ 18 O, and groundwater δ 18 O) affects model performance and parameter identifiability. Calibrating solely to discharge yielded optimal discharge simulations but poor reproduction of streamflow δ 18 O, whereas prioritising streamflow δ 18 O led to consistently strong performance across variables (NSE > 0.70), except for groundwater δ 18 O (NSE < 0.55). The best‐performing scenario assigned the highest weights to discharge (0.4) and streamflow δ 18 O (0.3), achieving NSE values of 0.76 and 0.66 for the calibration and validation periods, respectively. Streamflow δ 18 O was especially effective in constraining both flow and transport module parameters, while soil and groundwater δ 18 O provided limited or localised constraint. Groundwater storage parameters remained poorly identifiable across all calibration strategies. In addition, incorporating F obs did not significantly improve performance metrics but enhanced internal consistency, such as simulated young water fraction ( F sim ) and runoff source apportionment. Overall, the proposed calibration–constraint framework strengthens the process realism and diagnostic capability of tracer‐aided hydrological models and shows strong potential for application in other humid watersheds.
Language: English
Open Access status: bronze
OpenAlex ID: W4416236225
FWCI: 0.258
Cites: 38
Cited by: 1
Related to: