Questions on impact of channel delineation on LSUs and using LSU shapefile

34 views
Skip to first unread message

Fereshteh Kordrostami

unread,
Jan 30, 2026, 1:14:19 PMJan 30
to QSWAT+
Hello everyone,

I have two questions:

1. Channel delineation thresholds:
If lower drainage-area thresholds are used during watershed delineation to create a denser stream/channel network, does this directly lead to an increase in the number of landscape units (LSUs), or are LSUs largely independent of channel network density?

2. Using LSU.shp as an alternative to hru2.shp:
When using the dominant HRU option (1 HRU per LSU) to filter the HRUs from hru1.shp, is it conceptually and technically acceptable to use the final LSU.shp as a spatial alternative to the created hru2.shp to solve the issue of existing holes of filtered HRUs in it?

I would appreciate any answers or insights on these points.  

Thanks,
Fereshteh

Chris George

unread,
Jan 31, 2026, 12:25:57 PMJan 31
to Fereshteh Kordrostami, QSWAT+
1. There is one LSU (or a pair if you define floodplains) per channel
reach. So if the lower threshold produces more channel reaches there
will be more LSUs.

2. You are right to think that a single HRU per LSU will avoid the
problem of holes in the hru2 shapefile. Another approach would be to
define a zero minimum HRU area when filtering by area, or to use the
maximum number in filtering by the number of HRUs: either of these will
preserve all HRUs. Another possibility might be to do your intersection
with other data on the hru1 file and extend your results to the selected
HRUs: it depends on what you are trying to achieve.

Chris

On 1/28/2026 12:38 AM, Fereshteh Kordrostami wrote:
> Hello everyone, I have two questions: 1. Channel delineation thresholds:
> If lower drainage-area thresholds are used during watershed delineation
> to create a denser stream/channel network, does this directly lead to an
> increase in the number
> ZjQcmQRYFpfptBannerStart
> This Message Is From an External Sender
> This message came from outside your organization.
> ZjQcmQRYFpfptBannerEnd
> --
> You received this message because you are subscribed to the Google
> Groups "QSWAT+" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to qswatplus+...@googlegroups.com
> <mailto:qswatplus+...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/
> qswatplus/0620e4fd-772a-4118-ac50-58681e49807dn%40googlegroups.com
> <https://urldefense.com/v3/__https://groups.google.com/d/msgid/
> qswatplus/0620e4fd-772a-4118-ac50-58681e49807dn*40googlegroups.com?
> utm_medium=email&utm_source=footer__;JQ!!KwNVnqRv!
> DsM_NurQ7y_AS8rDhmF9HulDml7VrgAkbBXy3mBEqTdYId2QFkntCgx1jh3gG-9B7AvAmSFaiFH5fVofe7wCFUY$>.

Fereshteh Kordrostami

unread,
Feb 5, 2026, 10:51:09 AMFeb 5
to QSWAT+
Thanks for the suggestions. 

My goal is that my initial generated HRU map resulted in 64,000 HRUs. I am hoping to reduce the total count to improve the runtime speed, but when I run the tool to generate an HRU2 file, but that contains substantial gaps in the watershed. I've tried each option, but each yields the large gaps (see attached images). 

I was told previously the gaps are merged into the reduced HRU2 shp file, but there is no way to find what gap is associated with the new HRUs. Is there a SWAT output file (database or textinout) or means to get from QSWAT, a map that connects the HRU1 to the HRU2 shp file. This is necessary for post-processing to produce images that include the results within the area gaps. I saw in the qswatplus-tutorial.pdf file a reference to an output file called HruLanduseSoilSlopeRepSwat.txt (Fig. 23), but was unable to find it after running the program.

A separate question is that when I filter the HRUs by the dominant HRU option, it does not include the exempting landuse option (LU to always be in use). Is there a way to enable this feature?

Thanks
Fereshteh 
Screenshot 2026-02-03 110815.png
Screenshot 2026-02-03 110727.png

Chris George

unread,
Feb 6, 2026, 8:16:23 AMFeb 6
to Fereshteh Kordrostami, QSWAT+
The gaps in the hru2 shapefile are the HRUs removed by whatever filter
method you used. These gaps are imaginary: in the model the remaining
HRUs are expanded in area to fill the model. There is no way I know of
of illustrating this in a map. If you want to not have any gaps in the
hru2 file then don't filter any HRUs, eg by setting the area threshold
to zero.

The hru2 shapefile is just the hru1 shapefile with shapes for the
filtered HRUs removed.

The textual HRU output file is in the Watershed/Text directory of the
project.

If you want to choose the dominant HRU then you get the dominant HRU's
landuse, of course. If you want to make a model with every subbasin
having one HRU with the same landuse then make this landuse exempt from
elimination and filter HRUs by using the number of HRUs option and set
the number to the number of subbasins.

Chris

On 2/3/2026 8:08 PM, Fereshteh Kordrostami wrote:
> Thanks for the suggestions. My goal is that my initial generated HRU map
> resulted in 64,000 HRUs. I am hoping to reduce the total count to
> improve the runtime speed, but when I run the tool to generate an HRU2
> file, but that contains substantial
> <https://urldefense.com/v3/__https://groups.google.com/d/msgid/__;!!
> KwNVnqRv!BtKPZZccHFTI8eVHuT2ykV1BS8n-
> NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-OP2L6VdAJOxppDD8wtnZa8$>
> > qswatplus/0620e4fd-772a-4118-
> ac50-58681e49807dn%40googlegroups.com <https://urldefense.com/v3/
> __http://40googlegroups.com__;!!KwNVnqRv!
> BtKPZZccHFTI8eVHuT2ykV1BS8n-NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-
> OP2L6VdAJOxppDDmggSNHc$>
> > <https://urldefense.com/v3/__https://groups.google.com/d/msgid/
> <https://urldefense.com/v3/__https://groups.google.com/d/msgid/>
> > qswatplus/0620e4fd-772a-4118-
> ac50-58681e49807dn*40googlegroups.com <https://urldefense.com/v3/
> __http://40googlegroups.com__;!!KwNVnqRv!
> BtKPZZccHFTI8eVHuT2ykV1BS8n-NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-
> OP2L6VdAJOxppDDmggSNHc$>?
> > utm_medium=email&utm_source=footer__;JQ!!KwNVnqRv!
> >
> DsM_NurQ7y_AS8rDhmF9HulDml7VrgAkbBXy3mBEqTdYId2QFkntCgx1jh3gG-9B7AvAmSFaiFH5fVofe7wCFUY$>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "QSWAT+" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to qswatplus+...@googlegroups.com
> <mailto:qswatplus+...@googlegroups.com>.
> To view this discussion visit https://groups.google.com/d/msgid/
> qswatplus/3964ea31-c91b-49ed-848c-0faa2f9a0cf7n%40googlegroups.com
> <https://urldefense.com/v3/__https://groups.google.com/d/msgid/
> qswatplus/3964ea31-c91b-49ed-848c-0faa2f9a0cf7n*40googlegroups.com?
> utm_medium=email&utm_source=footer__;JQ!!KwNVnqRv!
> BtKPZZccHFTI8eVHuT2ykV1BS8n-NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-
> OP2L6VdAJOxppDDkwYuD1o$>.

Chris George

unread,
Feb 10, 2026, 6:56:58 AMFeb 10
to Fereshteh Kordrostami, QSWAT+
What you say would be true if that was how filtering is done, but it is
not. In your example, if there were 5 HRUs filtered down to 3, and if
the 2 filtered HRUs were together 20% of the total area, then the areas
of the three remaining HRUs would each be multiplied by 100/80, i.e.
increased by 25%, so the total area remained the same. In general the
new areas are not whole numbers of pixels.

The SWAT model is based on aggregation. It does not include every pixel
as an object: it collects all the pixels in each subbasin (or LSU in
SWAT+) that have the same landuse, soil and slope and regards them as a
unit, the HRU. The HRU does not have a precise geographical location,
only the subbasin or LSU it belongs to.

Chris

On 2/10/2026 12:31 AM, Fereshteh Kordrostami wrote:
> Thanks, so in short, there is no way of getting what set of original
> HRUs that were merged/filtered HRUs. For example, say we had 5 original
> HRUs that are filtered down to 2 such that HRUs 3 and 4 are merged into
> 1 and HRU 5 is merged into 2,
> ZjQcmQRYFpfptBannerStart
> This Message Is From an External Sender
> This message came from outside your organization.
> ZjQcmQRYFpfptBannerEnd
> Thanks, so in short, there is no way of getting what set of original
> HRUs that were merged/filtered HRUs.
>
> For example, say we had 5 original HRUs that are filtered down to 2 such
> that HRUs 3 and 4 are merged into 1 and HRU 5 is merged into 2, yielding
> the following map:
> HRU1 -> HRU2
> 1    -> 1
> 2    -> 2
> 3    -> 1
> 4    -> 1
> 5    -> 2
>
> This mapping would allow for producing images for reports given the new
> filtered HRUs and their new spatial extents.
>
> Best,
> Fereshteh
> <mailto:qswatplus%2B...@googlegroups.com>
> >      > <mailto:qswatplus+...@googlegroups.com
> <mailto:qswatplus%2B...@googlegroups.com>>.
> >      > To view this discussion visit https://groups.google.com/d/
> msgid/ <https://urldefense.com/v3/__https://groups.google.com/d/
> msgid/__;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvUdyNQM8$>
> >     <https://urldefense.com/v3/__https://groups.google.com/d/
> msgid/__; <https://urldefense.com/v3/__https://groups.google.com/d/
> msgid/__;>!!
> >     KwNVnqRv!BtKPZZccHFTI8eVHuT2ykV1BS8n-
> >     NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-OP2L6VdAJOxppDD8wtnZa8$>
> >      > qswatplus/0620e4fd-772a-4118-
> >     ac50-58681e49807dn%40googlegroups.com <https://
> urldefense.com/v3/__http://40googlegroups.com__;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvRndB6EA$> <https://urldefense.com/v3/ <https://urldefense.com/v3/>
> >     __http://40googlegroups.com__ <https://urldefense.com/v3/
> __http://40googlegroups.com____;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApv6bwR_gs$>;!!KwNVnqRv!
> >     BtKPZZccHFTI8eVHuT2ykV1BS8n-NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-
> >     OP2L6VdAJOxppDDmggSNHc$>
> >      > <https://urldefense.com/v3/__https://groups.google.com/d/
> msgid/ <https://urldefense.com/v3/__https://groups.google.com/d/msgid/>
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvRndB6EA$> <https://urldefense.com/v3/ <https://urldefense.com/v3/>
> >     __http://40googlegroups.com__ <https://urldefense.com/v3/
> __http://40googlegroups.com____;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApv6bwR_gs$>;!!KwNVnqRv!
> >     BtKPZZccHFTI8eVHuT2ykV1BS8n-NHTX0LXDssdDl7hH1HIFT6BgitztwAiAVR9k-
> >     OP2L6VdAJOxppDDmggSNHc$>?
> >      > utm_medium=email&utm_source=footer__;JQ!!KwNVnqRv!
> >      >
> >
>  DsM_NurQ7y_AS8rDhmF9HulDml7VrgAkbBXy3mBEqTdYId2QFkntCgx1jh3gG-9B7AvAmSFaiFH5fVofe7wCFUY$>.
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "QSWAT+" group.
> > To unsubscribe from this group and stop receiving emails from it,
> send
> > an email to qswatplus+...@googlegroups.com
> <mailto:qswatplus%2Bunsu...@googlegroups.com>
> > <mailto:qswatplus+...@googlegroups.com
> <mailto:qswatplus%2Bunsu...@googlegroups.com>>.
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvUdyNQM8$>
> > qswatplus/3964ea31-
> c91b-49ed-848c-0faa2f9a0cf7n%40googlegroups.com <https://
> urldefense.com/v3/__http://40googlegroups.com__;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvRndB6EA$>
> > qswatplus/3964ea31-
> c91b-49ed-848c-0faa2f9a0cf7n*40googlegroups.com <https://
> urldefense.com/v3/__http://40googlegroups.com__;!!KwNVnqRv!
> B4t1CEW5VPMoLburtXrDjDgShcbLprUZoKKzKfZ0fielcpJycdwCgalciZoXQWQqFAv9d7M83xfJdApvRndB6EA$>?

Fereshteh Kordrostami

unread,
Feb 12, 2026, 10:28:31 PMFeb 12
to Chris George, QSWAT+
Thanks, so in short, there is no way of getting what set of original HRUs that were merged/filtered HRUs.

For example, say we had 5 original HRUs that are filtered down to 2 such that HRUs 3 and 4 are merged into 1 and HRU 5 is merged into 2, yielding the following map:
HRU1 -> HRU2
1    -> 1
2    -> 2
3    -> 1
4    -> 1
5    -> 2

This mapping would allow for producing images for reports given the new filtered HRUs and their new spatial extents.

Best,
Fereshteh

Reply all
Reply to author
Forward
0 new messages