Combining parameters in final map

[odranreb98]

Hi,
I'm relative new to mike wave, I'm using it for a university project wich is part of my thesis and I have a student lab-kit with no support.
I'll try to briefly explain my problem: I have various 2D and 3D modeling output for different meshes (I'm studying costal defences in MIKE 3 wave) and I want to synthesised them by creating 2D (horizontal plane maps) wich combine the outputs in terms of surface elevation, current speed and current direction (directed off-shore), then select the worse case scenario of the combined index.
I have no experience with python and I discovered by chance MIKE IO. I'm trying to understand if it's possible to do what I just explained efficiently.

Do you think I have some chance to manage this topic using MIKE IO?
Any help or suggestion will be extremely helpful to me. I got to know that while MIKE have many different outputs visualisation options it seem not possible to easily combined different parameters.

Thank you very much

[ecomodeller]

Possible, but you need to learn some Python first and then how to use MIKE IO.
I suggest you read the MIKE IO docs and take a look a the online course material from Getting started with Dfs files in Python using MIKE IO. Good luck with your project!

[odranreb98]

I have been able to get into some basics of MIKE IO following the material kindly recommended bi ecomodeller: reading files, extracting some basic informations and plot them:
code.

I didn't manage to combine in the same plot outputs of the dsf file, I'm trying to show current speed and current direction (directed off-shore). Have someone some suggestion about it?
Have some of you any idea of how to interpret the Current direction (horizontal)? I can't understand the relationship between the numbers stored and the direction in the domain.

Thank you for any suggestion and help

[ecomodeller]

Take a look at https://support.dhigroup.com/. If you are unfamiliar with Python try the Mike Zero tools, e.g. the plot composer can be used to visualize currents with directions.

[odranreb98]

I have already studyed plot composer and data manager capabilities, but didn't find a way of combining outputs in the same plot. I know about statistics tool to rapidly analyse simulation results.
That's the reason why I'm trying MIKE IO.
Thanks for your reply

[odranreb98]

I'm moving on with the project and I'm stuck in the final move. Can someone please help me?
I have worked following this path:

1. run the simulation in MIKE 3 Wave
2. clip the timesteps I need with MIKE IO generic library of functions
3. extract maximum with MIKE data manager
4. create two dsfu files, one with current speed and one with current direction (maxima)
5. I have to multiply them and export the final dsfu 2D file, is it correct to use '*' operatore after defining variables like I have done?

Thank you very much

here is my code:

# code:
import pandas
import numpy
import matplotlib.pyplot
import mikeio
import mikeio.generic

# import of MIKE wave output
fin='C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP.dfsu'
mikeio.open(fin)

# select timesteps (and create statistic in MIKE data manager)
fout = mikeio.generic.extract('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP.dfsu', 'C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz.dfsu', start=650, end=750)
mikeio.open('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz.dfsu')

# max of surface elevation (se)
mikeio.open('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz_stat_se.dfsu')
stat_se = mikeio.read('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz_stat_se.dfsu') # stat. of se: max, min mean
sem = stat_se[1]
sem

# max of current speed (cs)
mikeio.open('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz_stat_cs.dfsu')
stat_cs = mikeio.read('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_staz_stat_cs.dfsu') # stat. of cs: max, min, mean
csm = stat_cs[1]
csm

# final multiplication
final = csm * sem
final.to_dfs('C:/Users/berny/Desktop/Batimetrie/2_Scogliere/SD_E+1.5_V50/HP_final.dfsu')

[jsmariegaard]

Hi @odranreb98

It seems that you're pretty close. Well done. A few suggested improvements for your code above:

• You don't need the mikeio.open() statements above (open returns an object that can be used to inspect the header of the file and later read a part of it); I can see that you use mikeio.read() afterways and that is a better option for you if the data fits in memory.

• Note that Python use 0-based indexing so if you want the first item (in this case the max) you should use stat_cs[0] instead of stat_cs[1]. It is even better to use the name of the item to be more explicit.

• Your second last step final = csm*sem is unfortunately not yet supported - but will be soon see Support multiplication and division of datasets. #792 so you will have to do it a bit more manual:

  final_vals = csm.to_numpy()*sem.to_numpy()
  final = mikeio.Dataset([final_vals, geometry=sem.geometry, time=sem.time, zn=sem._zn])

(not tested)

Hope this gets you a step further :-)

[odranreb98]

• You are right maybe it's a repetition to use mikeio.open, I was only using the code line to get confident with the file and what is inside them;
• That's a precious correction, I'm new to Python and still get MATLAB way of seeing indexing. I correct the istruction with the name:

csm = stat_cs["Statistical maximum : Current speed"]

• I can't get your point, before appling your correction I was getting some product in output (you can see from the image that the final dsfu file is generated and in the legend there is a product that have some sense also numerically)

Adding your line of code in my script I got some error:

unfortunately I'm not familiar with the function you have used, I imagine that it's needed to take the output file back in dsf file form, after the conversion to numpy matrix to carry on the operation that is not possible with mikeio library yet.
I'll try to figure out where the error come from seeing in https://dhi.github.io/mikeio/user-guide/dataset.html

Thank you very much once again