SWAT+ Editor not applying calibrated parameters to TxtInOut (hydrology.hyd/cn2.pst) — how to force-write correct values?

28 views
Skip to first unread message

Namraj Jaishi

unread,
Mar 7, 2026, 10:23:20 AM (13 days ago) Mar 7
to SWAT+ Editor

Hi SWAT+ Editor User Group,

I’m having an issue where parameter changes made in SWAT+ Editor (including calibration changes) don’t appear to be applied in the TxtInOut files that the SWAT+ engine actually reads. The run completes, but performance stays poor (NSE ~ -0.33), and the water balance suggests unrealistically high ET.

What I’m observing (the “mismatch”):

  1. hydrology.hyd values are not updating

  • In the Editor, my intent is something like:

    • esco ≈ 0.3 (or lower), lat_ttime ≈ 0.5, perco ≈ 0.01

  • But when I open TxtInOut/hydrology.hyd after writing inputs, the file still shows default-like values, e.g.:

    • esco ≈ 0.95, lat_ttime = 0.0, perco ≈ 0.05
      So the engine is still running with the “thirsty” settings (very high ET + drainage), which lines up with the poor streamflow fit.

  1. CN2 calibration / landuse confusion
    I understand landuse.lum uses labels (e.g., wood_f, pastg_f) rather than numeric CN2. My understanding is that a “hard calibration” for CN2 should update values in cn2.pst. However, I’m not seeing evidence the CN2 changes are being applied there either (or I’m checking the wrong place).

  2. Water balance indicates excessive ET
    For one year (e.g., 2025 in my summary table), I see roughly:

  • P ~ 1251 mm

  • ET ~ 984 mm (~79% of rainfall)
    This seems too high for my watershed and leaves too little runoff/baseflow, which explains the negative NSE at the gauge (Unit 88).

Questions:

  1. Has anyone seen SWAT+ Editor fail to write updated parameter values into TxtInOut (especially hydrology.hyd) even after “Write Input Files”?

  2. Is there a known setting/workflow to ensure the Editor writes the calibrated parameter set into the scenario TxtInOut that the engine uses?

  3. For CN2: is cn2.pst the correct file to verify “hard calibration” changes, and are there common reasons it wouldn’t update?

  4. Any recommended diagnostics to confirm which scenario folder/file set the engine is actually reading (to avoid editing the wrong TxtInOut)?

Thanks,

Namraj

glycine

unread,
Mar 12, 2026, 9:35:03 AM (8 days ago) Mar 12
to SWAT+ Editor
Hello! Im having the same issue with my model. I changed parameters but the flo-out had only small changes. 

Glydel Pascua

unread,
Mar 13, 2026, 9:56:39 AM (7 days ago) Mar 13
to SWAT+ Editor
I think I figured it out. Just add calibration conditions under your selected parameter,like type of condition is Region and operator is = and value is 1. It does improved my NSE. 

--
You received this message because you are subscribed to a topic in the Google Groups "SWAT+ Editor" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/swatplus-editor/E_kS3Js11bg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to swatplus-edit...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/swatplus-editor/3a715f53-f20b-4c98-9fed-8245007f6b78n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages