Why wflow

• The overall wflow approach
• A framework with multiple modelling concepts
• Interaction with data and other modelling suites
• An open source and flexible tool
• Case study Colombia
• Case study Flood Early Warning System, Pakistan
• References
• More info

The overall wflow-approach

Wflow is an open source distributed hydrological modelling platform developed by Deltares and targeted to perform hydrological simulations using raster data.

Typical applications of wflow include flood forecasting and impact of climate or land use changes on runoff. Wflow has been successfully applied worldwide for flood forecasting, evaluating the impacts of climate change or (land) management alternatives and inundation modeling. Wflow is a completely distributed modelling platform which maximizes the use of satellite data.

The wflow modelling platform is a completely distributed (gridded), hydrological model that calculates all hydrological fluxes at any given point in the model at a given time step, based on physical parameters and meteorological input data. Major hydrological processes included are snow, interception, soil moisture percolation and runoff generating processes. The kinematic wave is applied for routing and includes a reservoir module.

A framework with multiple modelling concepts

The wflow platform consists of several modelling concepts that share the same structure but are different with respect to the hydrological conceptualisation. The shared software framework includes the basic maps (digital elevation model, land use, soil type etc.) and the hydrological routing via the kinematic wave. The wflow_sbm model is derived from the topog_sbm model (Vertessy et al., 1999) that has been applied in various countries, most notably in Central America. The wflow_hbv model is derived from the HBV-96 model (Bergström, 1976) but does not include the routing functions. To route the water downstream, the model uses the same kinematic wave routine as the wflow_sbm model.  Other models within the framework are the wflow_gr4 model, a distributed version of the GR4 models (Perrin et al., 2001), the wflow_W3RA global hydrological model that is based on the AWRA—L model (van Dijk 2010, van Dijk et. al 2014) and the wflow Flex-Topo model that is based on the FLEX-Topo concept (Savenije, 2010 and Euser et al., 2015).

Schematic representation of the soil processes in wflow_sbm.

Interaction with data and other modelling suites

The wflow models maximise the use of available spatial data by linking parameter values to soil or land use wflow is an open source distributed hydrological modelling platform developed by Deltares and targeted to perform hydrological simulations using raster data.types and using gridded meteorological products. The distributed nature of the model implies that the model is run on each grid cell and that water flows from one grid cell to another either through the kinematic wave routine and/or through lateral groundwater flow. The wflow framework exposes the models to the outside world via the industry standard BMI (https://csdms.colorado.edu/wiki/BMI_Description). Using this standard, the models have been successfully connected to other modelling suites such as SOBEK for hydraulic modelling, Delwaq for water quality, RIBASIM for water allocation and MODFLOW for groundwater modelling. In addition, data assimilation and calibration has been performed via the OpenDA connection. A Delft-FEWS adaptor is available.

An open source and flexible tool

The wflow software is completely open source (https://github.com/openstreams/wflow) and free online documentation is available (https://wflow.readthedocs.org/).  The models are programmed in Python using the PCRaster Python extension. As such, the structure of the model is transparent, can be changed by other modellers easily, and the system allows for rapid development. 

Case study Colombia: Land use and climate change assessment, Colombia/Adaptation to climate change in Colombia

Description

For this case study that was financed by the Dutch governmental programme Partners for Water a tipping point analysis was performed by Deltares, Future Water, SarVision and UNESCO-IHE. The project was executed in cooperation with the Department of Planning (DNP), Institute of Hydrology and Meteorology (IDEAM) and two river basin authorities CorMagdalena and Cortolima. The objective was to provide a systematic analysis of future risks of flooding and water shortages for the upper and middle Magdalena (257.000 km2) and Coello (1843 km2) river basin respectively. In this analysis plausible future projections of climate (based on IPCC fifth assessment), land use and water demand (based on expert workshops and literature) were used to explore extremes in rainfall and river discharges, using the wflow_hbv hydrological modeling platform. 

Study area Magdalena Basin.

Wflow model set-up

For this project, a wflow model was set-up for the entire Magdalena Basin at a resolution of 0.02º (approximately 2km) using a shape file of the river network to make sure rivers were correctly implemented in the model. Catchments were delineated at points where observation data was available. A land use map and a soil type map were used to derive distributed parameter values. Once the model had been calibrated, it was used to run the climate and land use change scenarios. The model was used to derive conclusions on a large (Magdalena) and a small (Coello) scale basin. 

Results and conclusion

For the Magdalena Basin in Colombia, the following conclusions were derived:

• Climate scenarios show a persistent increase in the occurrence of extreme rainfall events. As a consequence, extreme discharges like in 2011 are likely to increase as well. The return period of the 2011 discharge is already quite high (probably unacceptably high) under the current climate and it might increase by a factor five to once every five years under climate change.
• Changes in forest cover may have additional adverse effects on peak discharges in locations upstream in tributary rivers of the Magdalena River.
• Only one out of eight climate scenarios show decrease in precipitation in one of the growing seasons. Consequently only this one climate scenario leads to an increase of the unmet water demand in the Coello basin. This unmet demand is however also under the current climate substantial. All other scenarios lead to a decrease of the unmet demand.

Distributed soil moisture state calculated by the model.

Case study Flood Early Warning System, Pakistan

Description

Pakistan has suffered a devastating flood disaster in 2010 when a series of monsoonal deluges over Northern Pakistan resulted in catastrophic flooding, loss of life and property and an agricultural crisis that may last for years. This event motivates the need to extent the existing forecasting system of the Indus River and its major tributaries to the Swat and part of the Kabul Rivers. For this purpose, a hydrological and a hydrodynamic model were developed by Deltares and NESPAK that allow forecasting of flood events in these river basins and produce inundation maps for various return periods. The model is designed to be used for flood forecasting as an intrinsic part of the Flood Early Warning Forecast of Pakistan (FEWS-Pakistan). Moreover, Deltares organised a training session for the employees of NESPAK to enable them to work more effectively with the Deltares team and to improve their skills on distributed hydrological modelling.

Location of the Swat and Kabul Rivers in the Indus basin, Pakistan.

Wflow model set-up

A wflow model was set-up for the Swat and Kabul River Basin in Pakistan (23567 km2) at a resolution of 500 m. The developed model should be able to simulate rainfall–runoff processes and snowmelt simultaneously based on a fully distributed modelling approach to facilitate coupling with a SOBEK routing model that represents the hydrodynamic component of the total flood forecasting system. Catchments were first delineated at points where observation data was available to calibrate the model. Then, the distributed modelling approach enabled to easily delineate catchments for all lateral inflows of the hydrodynamic model (see Figure below). A land use map and a soil type map were used to derive distributed parameter values. The snow related model parameters were calibrated using satellite products for snow cover area. The calibrated model is used in FEWS-Pakistan for real time flow forecasting.

Results and conclusion

In spite of data limitations, it was possible to make use of all available data by combining local measurements and satellite products to calibrate a wflow model for the Swat and Kabul Rivers in Pakistan that is used for real time flood forecasting. 

Modelled snow cover area in January 2005.

References

Bergström, S., 1976. Development and application of a conceptual runoff model for Scandinavian catchments, SMHI Report RHO 7, Norrköping, 134 pp

Euser T., Hrachowitz M., Winsemius H. C. and Savenije H.H.G, 2015. The effect of forcing and landscape distribution on performance and consistency of model structures, Hydrological Processes 29, 3727–3743

Perrin, C., Michel, C., Andréassian, V., 2001. Does a large number of parameters enhance model performance ? Comparative assessment of common catchment model structures on 429 catchments. Journal of Hydrology 242(3-4), 275-301

Savenije, H. H. G., 2010. HESS Opinions "Topography driven conceptual modelling (FLEX-Topo)", Hydrol. Earth Syst. Sci., 14, 2681-2692, doi:10.5194/hess-14-2681-2010

Van Dijk, A. I. J. M., 2010: The Australian water resources assessment system. Version 0.5, 3. https://www.clw.csiro.au/publications/waterforahealthycountry/2010/wfhc-awras-evaluation-against-observations.pdf (Accessed November 17, 2014).

Van Dijk, A. I. J. M., L. J. Renzullo, Y. Wada, and P. Tregoning, 2014: A global water cycle reanalysis (2003–2012) merging satellite gravimetry and altimetry observations with a hydrological multi-model ensemble. Hydrol. Earth Syst. Sci., 18, 2955–2973, doi:10.5194/hess-18-2955-2014.

Vertessy, R.A. Elsenbeer, H. 1999. Distributed modeling of storm flow generation in an Amazonian rain forest catchment: Effects of model parameterization, Water Resour. Res., Vol. 35, No. 7, Pages 2173-2187

More info

For more info, please contact us.