With NetCDF support, it seems that Raven can go for hourly time step easily for most, if not all, hydrologic models such as HBV and others. However, technically sounds does not necessarily mean we should. Based on my knowledge, HBV was developed for a daily time step because it is a conceptual model, not physically based. Therefore, I am not sure we should use the HBV type model in an hourly timestep. Am I right?
Kangsheng
The hourly time step for HBV type model
-
hongren_shen
- Posts: 1
- Joined: Wed Apr 20, 2022 9:06 pm
Re: The hourly time step for HBV type model
Hello Kangseng,
Thanks for your question.
I would say you were partially correct.
When we apply our models (either conceptual or physically-based) to a finer resolution (either spatial or temporal), we need to think about scaling issues, e.g., whether some processes represented in the model could be compromised in the finer resolution. What I don't agree with you is that this scaling issue is not because of whether the model is conceptual or physically based; it is because of the applicability of the an algorithm the model is using to represent a specific hydrologic process. This being said physically-based models could also suffer with incompatiable scaling issues.
Going back to your problem, you can use HBV for sub-daily scale modeling for sure. If you do a quick lit search on HBV modeling (or other bucket-type models) in sub-daily scale, there will be a bunch of such cases. But I suggest you check the processes you are using in your HBV model, since there are many different versions of HBV and I'm not sure which algrithms you employed in your version of HBV. If you want to see the details of an algrithm, say snow routine, you can search it in the Raven manual document and there will be equations and descriptions available. If you find an algrithm not applicable for sub-daily scales, either simply replace it with another alternative that supports sub-daily modeling or contact us for further support.
Let us know if you have further questions.
Cheers,
Hongren
Hongren Shen
PhD Candidate
Department of Civil and Environmental Engineering
University of Waterloo
Thanks for your question.
I would say you were partially correct.
When we apply our models (either conceptual or physically-based) to a finer resolution (either spatial or temporal), we need to think about scaling issues, e.g., whether some processes represented in the model could be compromised in the finer resolution. What I don't agree with you is that this scaling issue is not because of whether the model is conceptual or physically based; it is because of the applicability of the an algorithm the model is using to represent a specific hydrologic process. This being said physically-based models could also suffer with incompatiable scaling issues.
Going back to your problem, you can use HBV for sub-daily scale modeling for sure. If you do a quick lit search on HBV modeling (or other bucket-type models) in sub-daily scale, there will be a bunch of such cases. But I suggest you check the processes you are using in your HBV model, since there are many different versions of HBV and I'm not sure which algrithms you employed in your version of HBV. If you want to see the details of an algrithm, say snow routine, you can search it in the Raven manual document and there will be equations and descriptions available. If you find an algrithm not applicable for sub-daily scales, either simply replace it with another alternative that supports sub-daily modeling or contact us for further support.
Let us know if you have further questions.
Cheers,
Hongren
Hongren Shen
PhD Candidate
Department of Civil and Environmental Engineering
University of Waterloo