How to Implement a Sequential Simulation for Event-Based Modeling using Warm-Up Period?

[zaishang gu]

Dear SWAT user,I am attempting to set up an event-based rainfall-runoff simulation using SWAT2012. My goal is to simulate specific storm events accurately.My challenge is that I do not have long-term continuous hourly rainfall data to run a standard continuous simulation with a warm-up period. However, I do have sufficient long-term daily data.My proposed strategy is as follows:

1. Warm-up Simulation: Run a long-term daily simulation to establish realistic initial conditions (soil moisture, groundwater levels, etc.). This simulation would use my available daily weather data.
2. Save State: Save the final state of this warm-up run to an initial conditions file.
3. Event Simulation: Start a new, separate event-scale simulation (using hourly data for the storm event) by reading the initial conditions from the file generated in step 2.

I believe this approach would provide a more physically realistic starting point for my event simulation compared to using default initial conditions.My specific questions are:

1. Is this sequential method supported in SWAT? If so, how do I implement it? Specifically:
2. • How do I instruct SWAT to write the initial condition file at the end of the warm-up (daily) simulation?
   • How do I then tell the event-scale (hourly) simulation to read from that initial condition file at the start?
   • In which menus or input files are these settings located?

Thank you in advance for your time and expertise.Best regards,Guzaishang

[Omid Mohammadi]

Hi Zaishang,

We modified SWAT to print and get all required state variables as the initial conditions. Here are the papers:

https://www.sciencedirect.com/science/article/pii/S002216942200587X

https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2022WR032397

--
You received this message because you are subscribed to the Google Groups "SWAT-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swatuser+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/swatuser/a8e14548-985b-427a-992d-44f403f9eb27n%40googlegroups.com.

[zaishang gu]

Dear Omid Mohammadi,
Thank you very much for sharing the paper. It was extremely helpful for understanding the concept and confirmed that the method of saving and loading model states is viable for my research.
As I mentioned earlier, my coding skills are limited. I understand that the approach described in the paper requires significant modifications to the source code. However, my goal is quite straightforward: I would like to use the final state of a long-term daily simulation to initialize a short-term, event-scale hourly simulation.
Could you please provide some beginner-friendly, step-by-step guidance on how to use this feature? Specifically, I would like to know:
How to SAVE the initial conditions at the end of the daily simulation?
How to READ the initial conditions at the beginning of the hourly event simulation?
Any details you can share—such as the specific input file names and parameters to modify—would be immensely helpful. I suspect the process may be simpler than I imagine if the executable already supports this functionality.
Thank you again for your time and support.
Best regards,
Guzaishang