Anybody working with VIC model

[Ila Chawla]

Hi everyone,

Has anyone worked with VIC hydrology model. I want some help regarding the preparation of input files for the model. It require some global parameter file and some main data files in ascii formats and I am unable to figure out that how to make them.

[Sat Kumar Tomer]

The easy way to go with it would be to start with the sample data and then modify it according to your need. You can download sample data form this webpage:http://www.hydro.washington.edu/Lettenmaier/Models/VIC/SourceCode/Download.shtml.
There are link to many dataset, best is to download the vic.sample.stehekin.tgz, which is towards end of the webpage.

[Ila Chawla]

Thanks for the reply.

I download the sample sets and also thought of modifying the global parameter as well as header file, but wasn't sure of it.

Anyhow this model looks like a bit complicated to me and I haven't started off with it in real sense. So based on your experience what is your suggestion, that someone should start with this model or should go with SWAT and HEC-HMS (which are a bit more simple) kind of models?

[Sat Kumar Tomer]

What is the scale and resolution (spatial and temporal)?
What do you want from the model? Which hydrological variable you are looking for, do you want to study some climate scenarios. Please explain briefly about your problem. It will help me in suggesting which model might be good. 

[Ila Chawla]

Regarding spatial scale, my complete basin is quite huge somewhere around 1,00,000 sq.km., but I am planning to simulate it at sub-basin scale, but even then each sub-basin is around 7500-10,000 sq.km.

Temporal scale- I have data since 1970's.

and yes I will be dealing with land use and climate change scenarios.

and I want all the hydrological variables like infiltration, soil moisture, ET, etc.

[Sat Kumar Tomer]

I did not mean which variable you want as output. Most of the hydrological model provide all the variable in the output. I meant to ask which all variable you want to validate. Many of the hydrological model are good for the runoff, but while looking for other variables like soil moisture etc., more sophisticated models are required.

If you are looking for only runoff then SWAT and HEC-HMS might do the job. But if you want soil moisture, you should go for VIC. If I remember correctly, SWAT and HEC-HMS are not grid based, so you may face some limitations if you plan to compare any of them with gridded observations (satellite or global products).

I feel that SWAT or HEC-HMS might not be good in runoff at daily scale in Indian conditions. All the Indian studies I saw using these two models were at monthly scale.

 These were my opinion only, for a detailed comparison, you can search the article related to the models and see in what condition they were being used. Hydrological modelling is a quiet broader term, to decide about a hydrological model you should look for the data that you have for validation. Then see which of the model have been used to validate similar variables and at similar scale and resolution.

I hope I did not confuse you. If still there is some confusion, feel free to write.

[Ila Chawla]

Hi,

I have got a few more doubts in VIC model.

1. Regarding execution of VIC: On the VIC model site, they have give the command like vicNl -g and log.txt to run VIC model. But when I run it (even using the sample dataset), it just display the global parameter file, it doesn't generate the results.
Herein I was just trying to run the sample dataset that they have given, but I couldn't get the result even though in the global parameter file they have given the path for the output files. So I don't know, if I have missed some command or where am I going wrong.


2. Regrading global parameter file- the file that has been given in sample dataset is with .STEHE extension, i tried making  my file in txt format, but when I try to run it with command vicNl -g, it doesn't take it and if I modify the global parameter file given in the sample dataset (with.STEHE extension), then it reads it. So here again, I couldn't figure out that why is it not taking txt or ascii files.


Regrads,
Ila Chawla

[Sat Kumar Tomer]

Did you compile the VIC program to make the executable file? What they provide you is only C and some other auxiliary files, you need to run C codes to get the executable file.

I feel that you are trying to run the VIC model without making the binary (executable) file. If you are on Linux than you can you the make command to compile the program, if you are on Window, than you can open the code in any C editor and than run it. On windows make sure you have all the libraries.

Anyway which OS are you using? It would help in sorting out the issue.

[Ila Chawla]

I am working in window, but i have got that cygwin terminal. I have compiled the program using make command. It has been compiled completely and in directory it does show vicNl as exe file, otherwise with vinNl command it will show error. But here it is reading vicNl command.

I don't think that I have to run that exe file. I once clicked it. it showed a black box (resembling the command window) which within seconds vanished.. So is I am making some mistake here?

[Sat Kumar Tomer]

Just check if the model is correctly installed or not by running the following command:

Code: [Select]

vicNl -v
Tell me the output of this.

[Ila Chawla]

It terminal this line appears.

***** VIC Version 4.1.2.c bug fix update 2012-May-01 *****

[Sat Kumar Tomer]

Run the model with log file i.e.

Code: [Select]

vicNl -g global_parameter_filename >& log.txt
and attach the log file here.

[Ila Chawla]

This is the log file that I get.

[Sat Kumar Tomer]

The model is not able to read the file "F:/VIC/Bhagirathi/params/soil_param".
Check if the file exist at this location. You need to use the full file name along with the extension i.e. soil_param.STEHE and not only the soil_param. If this is the issue, use the proper file name for other files also.

[Ila Chawla]

Oh Okay..will do it.

Apart from this I am not sure if this the way it works?

And also as I mentioned in the 1st post, that while running this sample data also, it doesn't generate the output, it just display this global parameter file on the main window. So is there something I am missing out here?