Title of the dataset:
Data underlying the research of: Improving forecast skill of lowland hydrological models using ensemble Kalman filter and unscented Kalman filter

Authors:
Yiqun Sun, Albrecht Weerts


Related publication:
Improving forecast skill of lowland hydrological models using ensemble Kalman filter and unscented Kalman filter, submitted to WRR, 2019WR026616.

Description:
For operational water management in lowlands and polders (for instance in the Netherlands), 
lowland hydrological models are used for flow prediction, often as an input for a real-time control 
system to steer water with pumps and weirs to keep water levels within acceptable bounds. 
Therefore, proper initialization of these models is essential. The ensemble Kalman filter (EnKF) has been widely 
used due to its relative simplicity and robustness, while the unscented Kalman filter (UKF) has received little attention 
in the operational context. Here, we test both UKF and EnKF using a lowland lumped hydrological model. 
The results of a reforecast experiment in an operational context using an hourly time step show 
that when using nine ensemble members, both filters can improve the accuracy of the forecast by 
updating the state of a lumped hydrological model (Wageningen Lowland Runoff Simulator, WALRUS) 
based on the observed discharge, while UKF has achieved better performance than EnKF. 
Additionally, we show that an increase in the ensemble members does not necessarily 
mean a significant increase in performance. WALRUS model with either UKF or EnKF could be 
considered for hydrological forecasting for supporting water management of polders and 
lowlands, with UKF being the computational leaner option.

This dataset contains the following files:
Zip file with all simulations and forecasted data for both filters:
	EnKF_Ens18_2000_2001 contains netcdf-cf files starting with T0 of the forecast with forecasted ensemble streamflow (simulated_discharge) and ensemble WALRUS model states (depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit)
	EnKF_Ens36_2000_2001 contains netcdf-cf files starting with T0 of the forecast with forecasted ensemble streamflow (simulated_discharge) and ensemble WALRUS model states (depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit)
	EnKF_Ens72_2000_2001 contains netcdf-cf files starting with T0 of the forecast with forecasted ensemble streamflow (simulated_discharge) and ensemble WALRUS model states (depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit)
	EnKF_Ens9_2000_2011 contains netcdf-cf files starting with T0 of the forecast with forecasted ensemble streamflow (simulated_discharge) and ensemble WALRUS model states (depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit)
	EnKF_UKF9_2000_2011 contains netcdf-cf files starting with T0 of the forecast with forecasted ensemble streamflow (simulated_discharge) and ensemble WALRUS model states (depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit)
	Walrus_2000_2011 contains a netcdf-cf files with the WALRUS simulation (simulated_discharge,depth_quickflow_reservoir, depth_groundwater, depth_surface_water_reservoir, storage_deficit
	Obs_2000_2011.csv contains observed streamflow (Q in mm/hour) and forcing (P mm/hour ,T oC, ET mm/hour)

Explanation of variables:
n.a.

Methods, materials and software:

The experiments were conducted within Delft-FEWS (Werner et al. [2013]) using OpenDA-SOBEK. 
Delft-FEWS has its origin in flood forecasting and flood warning. In this study, Delft-FEWS is used to import 
data, run the forecasts, and export the results. The Delft-FEWS system used in this study (FEWS Vecht) 
is a standalone copy of the platform used by the water board for operational water management. 
OpenDA is a toolbox that is designed for data assimilation and calibration of numerical models 
(Rakovec et al., 2015). Both filtering algorithms used in this study (i.e., UKF and EnKF) have been included in OpenDA. 
SOBEK is a modeling software that mainly includes sub-modules for open water flow and rainfall-runoff processes in rural 
and urban environments, which can be used for flood forecasting, optimization of drainage systems, and control of irrigation systems. 
(Betrie et al., 2011; Bruni et al., 2015; Haile and Rientjes, 2005; Prinsen and Becker, 2011). 
WALRUS is available as a rainfall-runoff module (in C) within SOBEK. 
A wrapper for OpenDA-SOBEK-WALRUS is implemented to link the filter algorithms and the hydrological model.

The forcing data, precipitation, and potential evapotranspiration have been acquired from the 
Royal Netherlands Meteorological Institute (KNMI). Precipitation was measured at weather station Twenthe, 
and potential evapotranspiration was estimated using global radiation and temperature measured at the 
same station and the method of Makkink. Discharge observations were used for both calibration (by Loos, 2015) and verification. 
The discharge was measured by the local water authority, Water board Vechtstromen. All data were provided with hourly resolution.

This dataset is published under the CC BY-SA (Attribution ShareAlike) license.
