Towards Simulating Streamflow in Ungauged Basins Utilizing isoWATFLOOD: Time-Series Distribution of Oxygen-18 and Deuterium across Canada
Abstract
With growing awareness of the impact of climate change on environmental processes, hydrological modelling is a critical tool for the sustainable management and development of water resources. However, problems such as model validation and quantification of uncertainty arise in calibrating hydrological models over the large, often remote basins for which these predictions are required. Therefore, new tools are being sought that are capable of providing reliable, large-scale predictions of current and future water availability. The isoWATFLOOD model has the ability to simulate both quantity and isotopic composition of streamflow and runoff generation processes. Water mass and isotopes simulated in tandem can further constrain model simulations to measured isotopic signals in streamflow, facilitating calibration and constraining parameterization. isoWATFLOOD simulations of isotopes in streamflow first require isotopic forcing from climate, which is provided from the distribution of stable water isotopes in precipitation (δ18O and δ2H) observations. This study will highlight the development of seasonal time-series prediction models for precipitation δ18O and δ2H for the Canadian domain. It has been well documented that the isotopic composition of precipitation is dependent on climate, geographic location, topography, and moisture sources, among other factors. Based on this understanding, climate parameters obtained from the NCEP-NARR (National Centre for Environmental Prediction North American Regional Reanalysis) data set are stepwise regressed with CNIP (Canadian Network for Isotopes in Precipitation) isotopic observations to create monthly time-series multivariate prediction models for the distribution of precipitation δ18O and δ2H across Canada. By modelling both oxygen-18 and deuterium, d-excess and delta-delta space act as diagnostic tools to quantify model performance, alongside statistical methods. Stepwise regression results show significant correlation between δ18O and δ2H with predictor variables. The simulated Canadian local meteoric water line shows considerable agreement with the measured. Analysis of model residuals suggests a reasonable replication of δ18O and δ2H across the Canadian domain, with the majority of simulations within 15% of measured data. There are several instances where simulated d-excess shows large discrepancies from measured; indicating that the parameter combination or possibly the simplistic modelling approach may not be able to account for the physical processes behind the seasonal variation in isotope signatures at particular locations. Overall, these findings suggest that the δ18O and δ2H multivariate models are sufficient to reproduce isotopic signals in precipitation across the Canadian domain on a monthly basis and warrant implementation within the isoWATFLOOD model.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFMPP21A1769D
- Keywords:
-
- 1041 GEOCHEMISTRY / Stable isotope geochemistry;
- 1847 HYDROLOGY / Modeling;
- 3354 ATMOSPHERIC PROCESSES / Precipitation