Skip to main content

Main water system

The model of the main water system holds the rivers, canals and lakes of Rijkswaterstaat. The scripts are in pipeline/rijkswaterstaat, numbered in order of execution, and run in three DVC stages.

Map of the Netherlands with the Basin areas of the IJsselmeer, the rivers, the delta and the canals in blue, links between them, and small markers for the structures and boundaries.

The model of the main water system on the map.

flowchart TD
  subgraph frozen["bathymetry, frozen"]
    s1["1_bathymetrie.py<br>verwerkt/bathymetrie"]
  end
  subgraph demand["hws_demand"]
    direction TB
    s2["2_basins.py<br>Basins and profiles"] --> s3["3_netwerk.py<br>verwerkt/netwerk.gpkg"]
    s3 --> s4["4_kunstwerken.py<br>verwerkt/hydamo.gpkg"]
    s4 --> s5["5_model_netwerk.py<br>modellen/hws_netwerk"]
    s5 --> s6["6_model_sturing.py<br>modellen/hws_sturing"]
    s6 --> s7["7_model_onttrekkingen.py<br>modellen/hws_demand"]
    s7 --> s8a["8a_update_state.py<br>modellen/hws_demand"]
  end
  subgraph transient["hws_transient"]
    s8b["8b_update_bc.py<br>modellen/hws_transient"]
  end
  frozen --> demand --> transient --> merge["samenvoegen"]
  rwzi["rwzi"] --> transient

Bathymetry

1_bathymetrie.py merges data from Baseline (version j23_6-v1, non-public) with grids from bathymetrie-Nederland into one 5 × 5 m GeoTIFF. Both are needed, since the Midden Limburgse en Noord Brabantse kanalen are not in Baseline, and the IJsselmeer is not completely covered in bathymetrie-Nederland. For the mask we use Top10NL waterdelen and manually filled a boolean column baseline.

This stage is frozen, so it is not run again in a reproduction, since the source data does not change.

Bathymetry mask

Bathymetry mask

Bathymetry

Bathymetry

From data to a network

  • 2_basins.py derives the Basin areas of the main water system. From the bathymetry it writes a CSV file with the profile of each Basin, verwerkt/basins_level_area.csv.
  • 3_netwerk.py derives a ribasim_nl.Network (not a Ribasim network yet!) from OpenStreetMap, with modifications mainly in the IJsselmeer, to build the model network with in 5_model_netwerk.py. It needs the Basins of 2_basins.py, since a network node goes at each intersection of a Basin area with the network. This network, verwerkt/netwerk.gpkg, also guides the links in the coupling.
  • 4_kunstwerken.py combines the structures from NIS (Netwerk Informatie Systeem) of Rijkswaterstaat, Baseline, OpenStreetMap, the structures in the primary flood defences, and the abstractions, into a HyDAMO-based format, from which the Ribasim network is derived.
  • 5_model_netwerk.py combines the results of steps 2 to 4 into a first Ribasim model, modellen/hws_netwerk/hws.toml. It includes the distribution keys (verdeelsleutels) of Lobith and Monsin as structures, but no control nodes, and no demands. The Waal at Tiel is a TabulatedRatingCurve with the Q(h) relation of a calculation with LSM, verwerkt/kunstwerken/QhTielWaal.xlsx, downstream of the branch to the Amsterdam-Rijnkanaal, with the flow downstream of it.

Control and demands

  • 6_model_sturing.py adds control, with ContinuousControl, DiscreteControl and FlowDemand nodes, into modellen/hws_sturing/hws.toml. It also adds the structures next to another one, see PARALLEL_KWK in ribasim_nl.rws_kunstwerken, such as the locks next to the weirs of the Maas, the Nederrijn and the Lek, with a FlowDemand for their losses. It splits the Amsterdam-Rijnkanaal north of the Lek in three Basins, to which the Prinses Irenesluis listens.
  • 7_model_onttrekkingen.py adds the demands of drinking water, industry and energy as UserDemands, and adds Junctions, into modellen/hws_demand/hws.toml.
  • 8a_update_state.py runs this model, and writes its final state back as its initial state.

Boundaries and forcing

8b_update_bc.py makes the model that goes into the coupling, modellen/hws_transient, for 2017 up to 2020:

What Source
Discharges of the Rhine at Lobith and the Meuse at Monsin Daily means, aangeleverd/Matroos_debieten_2017_2022_Lobith_Monsin.xlsx
Levels at Kornwerderzand and Den Oever A seasonal table of −0.4 to −0.1 m, repeated every year
Level at IJmuiden Constant −0.4 m
Precipitation and evaporation LHM, as in the forcing of the regional models
Wastewater The RWZI model
Measurements Observation nodes, see ribasim_nl.observations

The other model in this folder that is tracked, hws_2025_10_1, is a fixed earlier version, from which the fix script of Vechtstromen reads the polygon of the Twentekanaal.