Evaluation of Regionalization Methods for Hourly Continuous Streamflow Simulation Using Distributed Models in Boreal Catchments

Abstract

Regionalization for prediction in ungauged basins at hourly resolution is important for water resources management (e.g. floods and hydropeaking). In this paper, calibration of 26 catchments (39-3090 km 2 ) in mid-Norway was performed using hourly records and three spatially distributed (1x1 km 2 ) precipitation-runoff models: a first-order nonlinear system model (hereafter Kirchmod), the HBV model and the Basic-Grid-Model (BGM). Four regionalization methods for each model namely parameter set yielding maximum regional weighted average (MRWA) performance measures (PM), regional median of optimal parameters (RMedP), nearest neighbor (NN) and physical similarity (PS) were evaluated and compared with three benchmarks. Parameter transfer from best regional donor (BRD) and from an ideal best arbitrary single-donor (BASD), and local calibration (LC) were as benchmarks. The physical similarity attributes include hypsometric curves (PSH), land use (PSL), drainage density (PSD), catchment area (PSA), terrain slope (PSS), bedrock geology (PSR), soil type (PSSO) and combination of all (PSC).Comprehensive evaluation of single-and multi-donors, simple benchmarks and more advanced regionalization methods using multi-models, two PM and their statistical evaluation indicate that the identification of regionalization methods is dependent on the models, the PM and their statistical evaluation. In general, the PSH, PSC and BRD methods performed better for the Nash-Sutcliffe efficiency (NSE) based on boxplots and regional median values of both the NSE and relative deterioration or improvement of the NSE from the local calibration due to the regionalization. The methods also performed better for the individual catchments. The PSS, RMedP, MRWA and BRD methods performed better for the log-transformed streamflow 2 (NSEln) based on the same evaluation criteria. Similar performance to the more advanced regionalization methods of transfer of homogeneous parameter sets across the whole region from the BRD for both NSE and NSEln indicate the potential of the simple regionalization approach for predicting runoff response in the region.

Regional Statistical and Precipitation–Runoff Modelling for Ecological Applications: Prediction of Hourly Streamflow in Regulated Rivers and Ungauged Basins

“…Wagener and Wheater, ) and uncertainty in identification of suitable regionalization methods (e.g. Hailegeorgis et al, ). The regression based prediction in the present study transfers characteristics of the observed streamflow from gauged to ungauged catchments while the regional calibration of the P–R model derives regional model parameters by utilizing the available streamflow data in the region.…”

“…Hence, the algorithm is useful to calibrate large number of catchments at once. Hailegeorgis et al () obtained acceptable performance of the regional calibration based on transferring regional parameter set that provides maximum regional weighted average (MRWA) NSE compared to other advanced regionalization methods. In the present study, we also evaluated the performance of regional transfer of parameter sets that provide the best MRWA NSE compared to the regional regression model: where N Ca is the total number of catchments included to compute the NSE MRWA ( N Ca = N C − 1) since a donor catchment is excluded for the leave one out cross‐validation, n ia is the length of timestamp with non‐missing observed streamflow series for each catchment i and N TS is the total length of timestamp for the calibration period.…”

“…Various regionalization methods that are useful for either regional transfer of calibrated parameters of the P–R models or for transfer of streamflow information (e.g. Parajka et al, ; Farmer et al, ; Hailegeorgis et al, ) are reported in hydrological sciences literature. For the regression model in the present study, we proposed a simple method to derive streamflow time series (hydrographs) for ungauged catchments from relationships between flow percentiles and drainage area and observed streamflow data at the gauged catchments.…”

“…Hence, the algorithm is useful to calibrate large number of catchments at once. Hailegeorgis et al (2015) obtained acceptable performance of the regional calibration based on transferring regional parameter set that provides maximum regional weighted average (MRWA) NSE compared to other advanced regionalization methods. In the present study, we also evaluated the performance of regional transfer of parameter sets that provide the best MRWA NSE compared to the regional regression model:…”

Infiltration from the Pedon to Global Grid Scales: An Overview and Outlook for Land Surface Modeling

“…Other researchers also developed and evaluated methods that are potentially applicable for transferring infiltration parameterizations across scales. Hailegeorgis et al (2015) evaluated four regionalization methods for continuous streamflow simulation. Their regional calibration method uses the maximum weighted average of a performance measure such as the Nash-Sutcliffe efficiency to identify specific parameter sets per subpixel by the DREAM algorithm (Vrugt et al, 2009).…”