There is a newer version of the record available.

Published November 17, 2021 | Version 5.1.0
Software Open

The Hydrological Discharge Model - a river runoff component for offline and coupled model applications

  • 1. Helmholtz-Zentrum Hereon

Description

The Hydrological Discharge (HD) model

The Hydrological Discharge (HD) model calculates the lateral transport of water over the land surface to simulate discharge into the oceans. It has been validated and applied in many studies since the publication of its original global 0.5° version (Hagemann and Dümenil, 1998; Hagemann and Dümenil Gates, 2001). Recently, a new version 5.0 has been developed that can be applied globally at a 5 Min. (~8-9 km) resolution. This HD version was applied and validated over Europe by (Hagemann et al., 2020). The HD model has been coupled to several global and regional ESMs (cf. Hagemann et al., 2020). It separates the lateral water flow into the three flow processes of overland flow, baseflow, and riverflow. Overland flow and baseflow represent the fast and slow lateral flow processes within a grid box, while riverflow represents the lateral flow between grid boxes. The HD model requires gridded fields of surface and subsurface runoff as input for overland flow and baseflow, respectively, with a temporal resolution of one day or higher. The present version 5.1 comprises several changes that are described in the files history.md that should be previewed below as well as in the file scr/history.txt.

Installation

You have have to uncompress the tarfile hdmodel_5_1.tgz in your desired code directory and then follow the instructions in ./scr/readme_hdmodel.txt to modify and run the compile and run script in a linux/unix environment. In order to run the HD model, you need to download the HD model parameter files Vs. 5.1 that are published as dataset on Zenodo.This dataset comprises parameter fields at 0.5° global resolution and at 5 Min. resolution (global, Europe).   

The script hdpara_cut.ksh cuts out a regional subdomain from the global 5 Min. HD parameter file (see readme_hdmodel.txt). Gridboxes that may laterally flow out of the domain at the regional boundaries are
converted into sinks to avoid numerical exceedances of array bounds. An examples is provided for Europe. Further regional domains should be implemented analogously.

Remark for coupling purposes

Note that this HD model distribution should include everything that is necessary to run the standalone/offline version of the HD model. The HD model is also equipped with coupling capabilities using the OASIS coupler that can be enabled with the compiling Option -o ON (cf. readme_hdmodel.txt). Then, the HD model expects to receive input fields of surface runoff and drainage/subsurface runoff on the respective HD model resolution/domain. This means that an interpolation of these fields from the atmosphere/land resolution to the HD model domain has to be done by OASIS before.

The interpolation of the discharge at the ocean inflow points (i.e. the river mouths) onto the ocean model grid is conducted within the HD model as OASIS is not able to thoroughly conduct such an interpolation of point-related fluxes (state of March 2021). In this respect, a coupling file is required for the HD model that is not part of this distribution. The program to generate such a coupling file is quite complicated and currently not in a very user friendly shape (this is work in progress). Hence, this is not part of this distribution.However, please feel free to contact Stefan Hagemann at Hereon (stefan.hagemann@hereon.de) who is open to cooperate and generate such a file for you.

Copyright and Licenses

Copyright 2021: Institute of Coastal Systems - Analysis and Modelling, Helmholtz-Zentrum Hereon

HD model code is licensed under the Apache License, Version 2.0 (http://www.apache.org/licenses/LICENSE-2.0)
Authors: Stefan Hagemann, Ha T.M. Ho-Hagemann  

Main Contact: Stefan Hagemann, Helmholtz-Zentrum Hereon, Email: stefan.hagemann@hereon.de

 

Acknowledgements

 

We acknowledge contributions of colleagues (see code headers) from the Max Planck Institute of Meteorology (MPI-M) and collaborating institutions who developed various infrastructure routines and who contributed to the implementation of the HD model into the coupled system ECHAM5/MPIOM and its successor MPI-ESM. We especially appreciated the help of Otto Böhringer (MPI-M) who separated the HD code from the fixed association with the MPI-ESM infrastructure. We are also grateful to Veronika Gayler (MPI-M) for her adapatations to Fortran90 and the implementation of various diagnostics.

References

Main publication for 5 Min. version of HD model (Vs. 4+):

  • Hagemann, S., T. Stacke and H. Ho-Hagemann (2020) High resolution discharge simulations over Europe and the Baltic Sea catchment. Front. Earth Sci., 8:12. doi: 10.3389/feart.2020.00012.

Further basic HD publications (Vs. 1, 0.5 degree)

  • Hagemann, S., L. Dümenil (1998) A parameterization of the lateral waterflow for the global scale. Clim. Dyn. 14 (1), 17-31
  • Hagemann, S., L. Dümenil Gates (2001) Validation of the hydrological cycle of ECMWF and NCEP reanalyses using the MPI hydrological discharge model, J. Geophys. Res. 106, 1503-1510

First publication using coupled 5 Min. version of HD model over Europe

  • Ho-Hagemann, H.T.M., Hagemann, S., Grayek, S., Petrik, R., Rockel, B., Staneva, J., Feser, F., and Schrum, C. (2020) Internal variability in the regional coupled system model GCOAST-AHOI, Atmos., 11, 227, doi: 10.3390/atmos11030227

 

Notes

# For some reasons, the preview does not work on history.md even though the file was tested with GitLab Markdown viewer. Hence, the changes are repeated here, too: ## HD Model Version 5.1.0, 17 Nov. 2021 ### Added -------------------------------------------------- - Implement Netcdf conform output -------------------------------------------------- - Implemented dependency of flow velocity on discharge For our initial 5 Min. setup over Europe, reasonable results were yielded with the HD model for many European rivers (with KGE often larger than 0.4). However, a global simulation showed that for very large rivers, the simulated discharge was lagging behind. This is something that can also be seen for the Danube, the second largest river of Europe. For small or medium-sized rivers, the present HD5 parameter settings seems to be appropriate, while for larger river it seems that a dependence of the flow velocity v on the discharge q must be regarded. Leopold and Maddock (1953) found an empirical dependency v = k Q^m. We defined a reference discharge qref = 1000 m³/s below which no correction is necessary. For q=qref, vref = k * qref^m v = vref * (Q/Qref)^m Consequently, for q > qref, we correct the flow velocity vref determined by the HD model river flow parameters by a correction factor (Q/Qref)^m. Leopold and Maddock (1953) found an average value of m=0.34 for 20 semi-arid rivers at stations in the Great Plains, and an average value of m=0.1 was obtained from various US rivers downstream. For our setup we chose an intermediate m=0.25. -------------------------------------------------- *** Fixed - Removal of two bugs in the parameter file for Vs. 5.0 Accidentally, there was no wetland impact on riverflow. For the wetland fraction at 1 km derived from GLWD wetlands, GLWD type 10 (50-100%) was erroneously associated with a fraction of 0.875 instead of 0.75, and GLWD type 12 was erroneously omitted. The latter is now associated with 0.125. -------------------------------------------------- ** *Changed Corrected usage of GLWD wetlands Removal of unspecified wetland areas (GLWD types 10-12) for the impact on riverflow that mainly only occur over North America. Now, GLWD wetland types 4-9 affect river flow. Removal of GLWD rivers from ESA waterbodies. For the calculation of lake fraction from ESA water bodies, the fraction of GLWD rivers (type 3) was substracted from the ESA waterbody fraction. -------------------------------------------------- Scaling factors on 5 Min. model parameters for overland flow (2) and baseflow OLF (4) were implemented into the HD parameter file and taken out of the run script.

Files

history.md

Files (208.4 kB)

Name Size Download all
md5:e7137a65ce1b2183d8c5efdab22c0259
205.8 kB Download
md5:281fda35931fdc88fdf5668a0e82e321
2.6 kB Preview Download

Additional details

Related works

References
Journal article: 10.3389/feart.2020.00012. (DOI)

References

  • Hagemann, S., T. Stacke and H. Ho-Hagemann (2020) High resolution discharge simulations over Europe and the Baltic Sea catchment. Front. Earth Sci., 8:12. doi: 10.3389/feart.2020.00012.
  • Hagemann, S., L. Dümenil (1998) A parameterization of the lateral waterflow for the global scale. Clim. Dyn. 14 (1), 17-31
  • Hagemann, S., L. Dümenil Gates (2001) Validation of the hydrological cycle of ECMWF and NCEP reanalyses using the MPI hydrological discharge model, J. Geophys. Res. 106, 1503-1510
  • Ho-Hagemann, H.T.M., Hagemann, S., Grayek, S., Petrik, R., Rockel, B., Staneva, J., Feser, F., and Schrum, C. (2020) Internal variability in the regional coupled system model GCOAST-AHOI, Atmos., 11, 227, doi: 10.3390/atmos11030227