SWAT check "NA" landuse?

227 views
Skip to first unread message

Jim Monchak

unread,
Apr 20, 2023, 3:55:03 PM4/20/23
to SWAT+ Editor
After running SWAT Check, in the Land Use Summary section I see an "NA" classification (3rd row in attached image). I'm wondering where this category came from, since I have no "NA" class in my CSV that was used to define the land uses?  I also verified the land uses of all the HRUs (both within SWAT+ Editor and the SQLite db), and they all have the proper assignments (there are no NA or null values).

swatcheck_LUsummary.jpg

Jaclyn Tech

unread,
May 4, 2023, 3:22:48 PM5/4/23
to SWAT+ Editor
Apologies, this is just a bug in the checker. The "NA" is probably referring to an urban landuse.

Zelalem

unread,
Aug 8, 2024, 2:10:44 PM8/8/24
to SWAT+ Editor
Hello All,

I am encountering an issue similar to the one described in your previous correspondence regarding the SWAT+ Check tool. Specifically, in the Land Use Summary section, I am seeing an additional land uses and "NA" classification in the attached image. I am puzzled by this category as my CSV file used to define the land uses does not include an "NA" class.

I have also thoroughly verified the land uses both within the SWAT+ Editor and the SQLite database. According to my checks, there are no "NA" or null values present.

Could you please advise on the possible cause of this "NA" classification and if there is a known solution or workaround? Any insights or guidance would be greatly appreciated.

Regards,

Zelalem
NA_LANDUSE.png
LANDUSES.png

oumaima oubella

unread,
Aug 14, 2024, 1:56:57 PM8/14/24
to SWAT+ Editor
Hello Zelalem,

I have the same problem can you give me the solution if you managed to solve the problem,

Regards,

Oumaima,


celray....@gmail.com

unread,
Aug 14, 2024, 2:44:11 PM8/14/24
to SWAT+ Editor
It is a bug refering to urban area, this will be fixed in the next release.

C. James

Jaclyn Tech

unread,
Aug 26, 2024, 1:20:58 PM8/26/24
to SWAT+ Editor

Rishabh Srikar

unread,
Sep 3, 2024, 9:04:01 AM9/3/24
to SWAT+ Editor
Hi Dr Jaclyn and Dr James,

I have recently upgraded my SWAT+ editor version from 1.2.2 to the one mentioned here (v3.0.4). 
I am still facing the bug with SWAT+ picking up the urban land use as NA. Furthermore, I couldn't help but notice that some changes have been made to the plant table in the SQLite database, for example, the WATR and BARR land uses are now missing leading to my following questions:
  1. Does the URBAN bug just affect SWAT+ Check, i.e., it still picks up the correct LULC from the urban table and corresponding properties?
  2. How may we address the removal of certain LULCs like the ones mentioned above, i.e., can we find alternatives? I've been trying to find a substitute for plain water bodies and plain barren land (no vegetation) but haven't found any appropriate classes that may fit.
Best,
Rishabh

Andreé Hidalgo Ravelo

unread,
Sep 3, 2024, 9:48:10 AM9/3/24
to Rishabh Srikar, SWAT+ Editor
SWAT+ no longer treats water bodies as LULC types. This assimilation caused a lot of problems in the past.
I replace "WATR" type with "WETN" type in the LULC map, and then, in the QSWAT+ 1st step, I import lakes shp with WETW attribute (i.e. RES = 4, corresponding to PLAYAS). 
For me, that is the only way to get the same effect, but with a higher level of compatibility with the software.

Take into account that the complete structure of swat+ has changed, and now the water bodies are not part of LULC maps.

On the other hand, URBN is a very generic type. 
Normally the software should be let you run your model anyway, but the tren is that you now need to specify more clearly the exact type of your city landuse (very dense population, low-population, transports, etc, etc.) with the regarding new codes (URHD, URML, UTRN, etc, etc.)

Regards.

--
You received this message because you are subscribed to the Google Groups "SWAT+ Editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to swatplus-edit...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/swatplus-editor/dea85adb-5aa7-42b2-b446-c8272f3e0799n%40googlegroups.com.

Jaclyn Tech

unread,
Sep 4, 2024, 11:09:56 AM9/4/24
to SWAT+ Editor
Hi Rishabh,

In the 3.0.4 fix I removed any printing of the letters "NA" at all in the LULC column of the land use summary table. So if you are still seeing rows with NA as LULC, something was not updated correctly or something else is going on. But yes, this bug printing "NA" would have just been an issue in SWAT+ Check, and not the model itself.

Jaclyn Tech

unread,
Sep 4, 2024, 11:13:52 AM9/4/24
to SWAT+ Editor
Actually, sorry! Writing this message out just made me realize, I think I was going to quick releasing the fix and didn't push the correct api exe. I am rebuilding now and will publish to the github page at https://github.com/swat-model/swatplus-editor/releases/tag/v3.0.4 in just a few minutes. Check for a file with today's date, and reinstall from there. Hopefully that will fix it!

Thanks,
Jaclyn

Message has been deleted

Rishabh Srikar

unread,
Sep 5, 2024, 2:37:06 PM9/5/24
to SWAT+ Editor
Hi Dr Jaclyn,

Thank you very much for your prompt response. I will update my version and let the community know the status. 

Best,
Rishabh

Rishabh Srikar

unread,
Sep 5, 2024, 2:37:15 PM9/5/24
to SWAT+ Editor
Hi Andree,

Thank you very much for your reply. I truly appreciate your clarifications. 

Although I did know about the subclassifications under Urban land use (I am using URTN), I did not know that SWAT+ does not accommodate water bodies as a land use. I shall attempt the workaround you suggest here, however I have a couple of questions right away regarding its execution:

1. What should the structure of the water body shapefile with attribute field RES (1/2/3/4 depending on water body type) be? Can it be a single shapefile with one shape and one attribute record but be a multipart polygon (to accommodate disjointed water bodies spread across the watershed)?
2. After updating my Editor and reading the latest manual revision (v3.0), I came across the following line " Playas do not form separate items in delineation: they simply override the land use their pixels are assigned from the land use map, and so directly create HRUs, with land use WETW". Apologies if it's obvious but I am unable to understand this, what does a WETW attribute mean here? Do I need to have my shapefile consist of another attribute column indicating an LULC code?

Best,
Rishabh

Andreé Hidalgo Ravelo

unread,
Sep 6, 2024, 11:03:03 AM9/6/24
to Rishabh Srikar, SWAT+ Editor
Hello, I directly extracted water bodies from Corine Landcover associated types. It's not necessary to vectorize from a raster, but, indeed, the input is a polygon shapefile. 

The exact input format can be found in the QSWAT+ reference manual.

Regards. 


On Thu, Sep 5, 2024, 8:36 PM Rishabh Srikar <rishabhs...@gmail.com> wrote:
Hi Andree,

Thank you indeed for your response. I should have clarified that I did know about the subclassifications under urban (using URTN in my case) but SWAT+ editor still depicts it as NA at least in the SWAT check section as well as the home page of the interface. I hope this is just a representation issue and the model is still picking up the correct LULC from the urban table along with its properties during its run. 

I did not know that SWAT+ no longer accommodates water bodies as a LULC. I shall attempt what you suggest and hope that I can get back here in case I run into anymore hurdles. One question about the water bodies off the top of my head is; Will a simple vectorization and extraction of water bodies as a single polygon shape file (with an attribute field RES and appropriate values) do?

Best, 
Rishabh





On Tuesday, September 3, 2024 at 7:18:10 PM UTC+5:30 taki...@gmail.com wrote:

Rishabh Srikar

unread,
Sep 6, 2024, 11:03:07 AM9/6/24
to Andreé Hidalgo Ravelo, swatplu...@googlegroups.com

Hi Andreé,


Thanks again for your response. I have a large number of small water bodies as part of my LULC which are probably not covered in any global dataset, therefore I think I'll have to vectorise in my case.

Best, 
Rishabh 
Reply all
Reply to author
Forward
0 new messages