Rain loss - Initial and Continuous loss model

CyprienBosserelle / BG_Flood

Numerical model for simulating shallow water hydrodynamics on the GPU using an Adaptive Mesh Refinment type grid. The model was designed with the goal of simulating inundation (River, Storm surge or tsunami). The model uses a Block Uniform Quadtree approach that runs on the GPU but the adaptive/multi-resolution/AMR is being implemented and not yet operational. The core SWE engine and adaptivity has been inspired and taken from St Venant solver from Basilisk and the CUDA GPU memory model has been inspired by the work from Vacondio _et al._2017)

GNU General Public License v3.0

34 stars 15 forks source link

Rain loss - Initial and Continuous loss model #70

Closed AliceHarang closed 1 year ago

AliceHarang commented 1 year ago

This objective is to add a model of Initial and Continuous Loss to the code. This model aims to add a very basic model of ground infiltration, in order to improve the surface runoff when rain on grid is used. An initial loss of water is estimated before the runoff begin, then a constant volume of rain, the continuous loss is remove when the runoff occurs. These values are defined, based on the material of the ground, at each cell location.

from https://www.mssanz.org.au/modsim2011/I8/elkafagee.pdf

Discuss:

It is actually create similarly to the roughness, do we want to be able to add a constant value too?

Steps:

[x] Create variables: ground water accumulation, IL map, CL map
[x] Read initial maps of IL - CL
[x] Write the map in outputs
[x] Modify the Equations
[x] Add testcase / validation

AliceHarang commented 1 year ago

Hi, Do you think we would need to add the possibility of reading these variables as constant? (next to the actual "map mode")

CyprienBosserelle commented 1 year ago

The new system works great.

Right now the user has to force the IL and CL as maps an although I though it may be useful to allow a constant value everywhere , I think the way it is now is preferable. It forces you to think what to do in the ocean (il and cl =0.0) and river bed.

Few things I'd like to see added (I can help do that)

[x] initialise infiltration to IL where initial h>0.0 so the ocean/estuaries do suddently drop even if IL is not 0.0 (it could make funny looking waves)
[x] Need a better output variable name for infiltration
[x] Add md file explaining equation and typical values. maybe a image nor two of expected results.

CyprienBosserelle commented 1 year ago

Sorry for the pedantic comment about adding a md file for the documentation but I think we need to do that for every PL. That way the documentation doesn't run away from us. I will add that to every pull request for now on....

AliceHarang commented 1 year ago

Hi, I do agree with the md file, we need to know what is in the code. For the initial value for the ocean/estuaries, do you want to force it over the user map? For the moment, if there is a user map, we use it; if there is no IL and CL input, there is no infiltration (everything should have a on/off switch).

Regarding the output var name, I didn't have much inspiration... (it is the (accumulated) height of water infiltrated through this cell up to this time)

CyprienBosserelle commented 1 year ago

I did the initial condition for initial loss (infiltration = il where h>eps ) I will think of better output name variable and propose a name.

AliceHarang commented 1 year ago

Regarding the documentation md file, I made a basic word document that I can quickly transform in a md file.

AliceHarang commented 1 year ago

I added a md file to explain the model. Do you think it is sufficient? (Can you clarify the figure of expected results?)

AliceHarang commented 1 year ago

A sanity check on outvars to be sure that the model/options are in phase with the requested outputs. Done for ic/cl/hgw Done for datmpdx/datmpdy