intro story D-Flow FM


D-Flow Flexible Mesh

D-Flow Flexible Mesh (D-Flow FM) is the new software engine for hydrodynamical simulations on unstructured grids in 1D-2D-3D. Together with the familiar curvilinear meshes from Delft3D 4, the unstructured grid can consist of triangles, pentagons (etc.) and 1D channel networks, all in one single mesh. It combines proven technology from the hydrodynamic engines of Delft3D 4 and SOBEK 2 and adds flexible administration, resulting in:

  • Easier 1D-2D-3D model coupling, intuitive setup of boundary conditions and meteorological forcings (amongst others).
  • More flexible 2D gridding in delta regions, river junctions, harbours, intertidal flats and more.
  • High performance by smart use of multicore architectures, and grid computing clusters.
An overview of the current developments can be found here.
The D-Flow FM - team would be delighted if you would participate in discussions on the generation of meshes, the specification of boundary conditions, the running of computations, and all kinds of other relevant topics. Feel free to share your smart questions and/or brilliant solutions! 


We have launched a new website (still under construction so expect continuous improvements) and a new forum dedicated to Delft3D Flexible Mesh.

Please follow this link to the new forum: 

Post your questions, issues, suggestions, difficulties related to our Delft3D Flexible Mesh Suite on the new forum.





Sub groups
D-Flow Flexible Mesh
Cohesive sediments & muddy systems


Message Boards

Tidal water level variation inside an estuary

Saeed Shaeri, modified 5 Years ago.

Tidal water level variation inside an estuary

Youngling Posts: 6 Join Date: 10/7/12 Recent Posts
Dear forum members,

My question is about tidal water level variation inside an estuary.

To make my question simple, at a location some kilometres upstream of a tidal inlet, water level is needed. The flow domain is forced by a boundary condition which is a variable water level, some kilometres offshore of the inlet.

My understanding is that the changes of the boundary, make the water level variation all across the modelling domain. Therefore, it should take some time steps which the very far cells from the variable boundary are influenced by such changes. I believe that is the case for a wave modelling; however, in a flow model I experienced different.

Hence, my question/confusion is that if I should observe water level variation at all cells exactly from the initial time step. This would also show no time-lag between any two cells of the domain.

Looking forward to your kind comments,
Phil Shepperd, modified 5 Years ago.

RE: Tidal water level variation inside an estuary

Padawan Posts: 33 Join Date: 5/31/11 Recent Posts
If you think of your model not as a computational one, but as a physical one, made of fibreglass, filled with water ,sitting in a laboratory. You can visualise that it will take some time for a water level change at the boundary to propagate up through the estuary to your point of interest.
Your FLOW model works in essentially the same way - a water change at the boundary will move through the model grids cells, one timestep at a time, until it gets to the point of interest. So if you were to look at some trim* output, you would see the tidal wave propagating just like it would in the physical model, or the real world.
Phil Shepperd, modified 5 Years ago.

RE: Tidal water level variation inside an estuary

Padawan Posts: 33 Join Date: 5/31/11 Recent Posts

As a quick check, the speed of a wave (tidal or otherwise) in shallow water is sqrt (gd) where g is the acceleration due to gravity (9.81 m/s) and d is the local water depth (m).
Giordano Lipari, modified 5 Years ago.

RE: Tidal water level variation inside an estuary

Youngling Posts: 12 Join Date: 3/23/11 Recent Posts
Hi Saeed.

The intuition that there's a propagating long wave into the estuary is correct. For this to be an effective start of a numerical simulation, you need to consider what your initial water level is and the phase of the tide at simulated time zero.

You can keep in mind the common distinction in numerical modelling between a 'cold start' and a 'hot start'.

In the cold-start case, you begin from fictitious conditions such as a quiescent basin that, then, is acted upon by the relevant forcing (tide, discharge wind and all that applies). A cold state hints at no motion here. The motion in the water body, then, develops until reaching a dynamical equilibrium between the applied forcing and the resisting forces (bottom and internal friction). You recognize such equilibrium because it's either a steady state or an unsteady process that has lost memory of the fictitious initial conditions (e.g. a proper tide in the estuary).

The hot start is when you start, actually restart, a simulation from a snapshot situation taken from a previous (correct) simulation. So, at simulated time zero, velocity and water levels are variable all over the domain, but the balance of forces already holds in each cell of the domain.

By contrast, at time zero in the cold start, you are unlikely to have balance between the initial condition in the inner domain and the boundary condition. That's I guess what was puzzling you. Well, in that case you run the simulation until you observe regular tidal oscillations everywhere in the basin. This is the so-called spin-up time. As long as your domain's description is realistic, the simulation will take the right course (see caveats below though). To judge what a regular tidal oscillation is, it's better to refer to measurements but common sense will certainly guide you ruling out oddities. There's some more physics specific to estuarine circulation to take into account here but, as a first pass, it'd be alright.

The final touch is a good match of initial condition and boundary conditions. At the boundary you have the tide which is principally synchronised with the calendar. Start then a simulation at a calendar time when the predicted water level at the boundary is not largely different from the water level set in your initial condition (or the other way round). It seems preferable to me that you also start at flood so that the water comes into the estuary and fills it (but I may be overlooking other aspects now, so happy to be integrated).
Either way, there's one more feature in the numerical parameters that has been devised to keep issues at bay. You can ask the solver not to apply the boundary condition at full value from the outset, but to increase the forcing gradually within a limited period (that you choose) so as to avoid a jump start (which, numerically speaking, can be obnoxious in cases).

I don't know how large your estuary and tidal range are, which determines the order of magnitude of it all. As a guess out of my hips, I would say that with a ramp-up time of 1-2 day and discarding the first week of simulated time you should be sorted or getting there. With the cold start, you always need to take into account a buffer time of sorts though.

Happy to be corrected/integrated. The no-slip condition applies to water, not to water experts. Hope this helps.