Quantifying uncertainty in hydrological system responses to climate variability is of fundamental importance for development strategy of water resources adaptation to climate change. It is well established that the most important source of hydrological uncertainty comes from the uncertainty in the future climate projections which has the largest impact on the projected hydrological change. Uncertainty in climate projections derives from three main sources: forcing, model response, and internal variability. He we study how the uncertainty of meteorological parameters caused by internal atmospheric variability is transformed by a hydrological system.
An approach is proposed to assess uncertainty of hydrological simulations originating from the internal variability of the atmosphere. This internal variability, also termed “weather noise”, induces an uncertainty that is inherent to the climate system and that is the lowest level of uncertainty achievable in climate change studies. It is, therefore, used as a threshold to define the significance of the hydrological modelling induced uncertainty.
The weather noise was estimated performing 45 ensemble simulations with Atmospheric Global Circulation Model (AGCM) ECHAM5 with identical boundary conditions (sea surface temperatures and sea ice concentrations) prescribed as observed during 1979-2012 and different initial conditions. AGCM output from all 45 numerical experiments was used (after bias correction post-processing) for friving the hydrological models. As a result, the corresponding ensemble of multi-year hydrograph was modelled and analyzed. Two world wide tested hydrological models were utilized for the simulations: ECOMAG and SWAP. Case study was carried out for two large rivers of the Arctic basin: Lena and Northern Dvina.
Hydrological models forced by ECHAM5 AGCM output were integrated to provide information on the statistical properties of runoff characteristics. Mean annual and monthly runoff values as well as their standard deviations were estimated from the simulated ensemble of hydrographs and compared with corresponding statistics obtained from the observed discharge series. Simulated and observed trends in the mean annual runoff were also analyzed. The runoff statistics uncertainty caused by the ”weather noise” was estimated. We found that the runoff characteristics of the smaller Northern Dvina River basin are more sensitive to the internal atmospheric variability. Our results indicate that the internal variability of the climate system is important when interpreting hydrological sequences of climate change.
The study was supported by the Russian Ministry of Education and Science (grant No. 14.B25.31.0026). The present work has been carried out within the framework of the Panta Rhei Research Initiative of the International Association of Hydrological Sciences (IAHS).