Quantile-Based Hydrological Modelling

Predictive uncertainty in hydrological modelling is quantified by using post-processing or Bayesian-based methods. The former methods are not straightforward and the latter ones distribution-free (i.e. assumptions on probability distribution of model's output necessary). To alleviate possible limitations related to these specific attributes, this work we propose calibration model quantile loss function. By following methodological approach, one can directly simulate pre-specified quantiles predictive streamflow. As a proof concept, apply our method frameworks three models 511 river basins contiguous US. We illustrate show how an honest assessment performance be made proper scoring rules. believe that help towards advancing field uncertainty.