Uncertainty propagation in a modelling chain of climate change impact for a representative French drainage site

Analysis of the uncertainty propagation along a hydroclimatic modelling chain has been performed by few studies to date on subsurface drainage hydrology. We performed such an analysis in a representative French drainage site. A set of 30 climate projections provided future climatic conditions for three representative concentration pathways (RCPs): RCP2.6, RCP4.5, and RCP8.5. Three hydrological models for drainage systems, MACRO, DRAINMOD for DRAINage MODel, and SIDRA-RU for SImulation du DRAinage - Reserve Utile in French, on the three different parameter sets were used to quantify uncertainties from hydrological components. Results showed that the RCP contribution to total uncertainty reaches almost 40% for air temperature, does not exceed 15% for precipitation, and is almost negligible for hydrological indicators (HIs). The main source of uncertainty comes from the climate models, representing 50-90% of the total uncertainty. The contribution of the hydrological components (models and parameter sets) to the HI uncertainty is almost negligible too, not exceeding 5%.

Automated summary

To date, there have been few studies analyzing the propagation of uncertainty in a hydroclimatic modelling chain for subsurface drainage hydrology. This study conducted such an analysis in a representative French drainage site using 30 climate projections under three representative concentration pathways. Three hydrological models and parameter sets were used to quantify uncertainties in hydrological components. Results showed that the main source of uncertainty came from climate models, while the contribution of hydrological components to the uncertainty was negligible.