Seasonal streamflow forecasts in the Ahlergaarde catchment, Denmark: the effect of preprocessing and post-processing on skill and statistical consistency

Abstract

Abstract. In the present study we analyze the effect of bias adjustments in both meteorological and streamflow forecasts on the skill and statistical consistency of monthly streamflow and yearly minimum daily flow forecasts. Both raw and preprocessed meteorological seasonal forecasts from the European Centre for Medium-Range Weather Forecasts (ECMWF) are used as inputs to a spatially distributed, coupled surface–subsurface hydrological model based on the MIKE SHE code. Streamflow predictions are then generated up to 7 months in advance. In addition to this, we post-process streamflow predictions using an empirical quantile mapping technique. Bias, skill and statistical consistency are the qualities evaluated throughout the forecast-generating strategies and we analyze where the different strategies fall short to improve them. ECMWF System 4-based streamflow forecasts tend to show a lower accuracy level than those generated with an ensemble of historical observations, a method commonly known as ensemble streamflow prediction (ESP). This is particularly true at longer lead times, for the dry season and for streamflow stations that exhibit low hydrological model errors. Biases in the mean are better removed by post-processing that in turn is reflected in the higher level of statistical consistency. However, in general, the reduction of these biases is not sufficient to ensure a higher level of accuracy than the ESP forecasts. This is true for both monthly mean and minimum yearly streamflow forecasts. We discuss the importance of including a better estimation of the initial state of the catchment, which may increase the capability of the system to forecast streamflow at longer leads.

Subseasonal hydrometeorological ensemble predictions in small-and medium-size mountainous catchments: Benefits of the NWP approach

“…The QM method used here is a popular pre-processing method for hydrometeorological ensemble forecasts (e.g. Kang et al, 2010;Lucatero et al, 2018;Verkade et al, 2013) but does not come without limitations. In particular Zhao et al (2017) point out the inability of QM to provide fully reliable ensembles for post-processing GCM precipitation.…”

“…The QM technique is a 5 simple and widely used method for pre-processing hydrometeorological forecasts (e.g. Kang et al, 2010;Lucatero et al, 2018;Verkade et al, 2013). For a given target day of a reforecast the correction is derived from the distribution of all the reforecasts within a three weeks window around the same lead day and the corresponding observations, hence the correction depends both on the lead time and on the period of the years.…”

“…In this approach, observed historical meteorological conditions are used to generate the streamflow forecasts. Several studies have shown that Ensemble Streamflow prediction (ESP) can provide skillful seasonal streamflow prediction (Harrigan et al, 2017) and seasonal hydrometeorological prediction systems have difficulties to outperform the ESP beyond one month lead time (Arnal et al, 2018;Lucatero et al, 2018). Irrespective of the methodology used, the performance of ensemble streamflow forecasts depend to a large degree on 30 the initial condition within the catchment especially for small catchments (Van Dijk et al, 2013;Thirel et al, 2010).…”

Seasonal hydroclimatic ensemble forecasts anticipate nutrient and suspended sediment loads using a dynamical-statistical approach

“…GCM outputs from the CFSv2 are often characterized by the presence of systematic biases (Saha et al 2014), which propagate through the hydrological model to result in biased streamflow forecasts (Yuan and Wood 2012). Statistical preprocessing techniques are used to correct such biases (Shukla and Lettenmaier 2011, Yuan and Wood 2012, Crochemore et al 2016, Lucatero et al 2018. We use a logistic regression model to statistically preprocess the CFSv2 forecasts (Messner et al 2014a, 2014b, Yang et al 2017, Sharma et al 2018.…”

Improving hydrological post-processing for assessing the conditional predictive uncertainty of monthly streamflows

“…While our research focused on the predictive uncertainty of projected streamflows, future works will investigate the effect of pre-processing inputs (i.e. precipitation and temperature) and post-processing streamflows to identify the main source of predictive uncertainty, such as Lucatero et al, (2018) obtained for seasonal forecasts. Forthcoming studies will also implement the ABC post-processor in other catchments with different hydrological conditions to be able to draw more generalised conclusions.…”