bug message reminder

When adressing a model crash or bug, please remember to include an entire model setup in your post that reproduces the crash or exposes the bug. Also add the XBlog.txt file. This is necessary information for people that are trying to help you. Including your model setup can be achieved by adding the zipped run directory (excluding output) as an attachment to the post.

Forum

RE: Vegetation

DJ
Drandreb Earl Juanico, modified 4 Years ago.

Vegetation

Capillary Posts: 2 Join Date: 2/15/15 Recent Posts
Recently updated XBeach installation via svn. I noticed some changes in the parameter designations related to the vegetation module.
I am hoping that the following issues would be clarified.
I tried seeking for answers through this forum and in the manuals, but I could not find any explanation at all.

1.) First up, the parameter 'ah' seems to have been replaced by 'zv'. I notice this in the XBlog.txt after starting a simulation. Could anyone please confirm this change? Is it correct to assume that 'ah' was merely renamed to 'zv'? If not, what is 'zv', which I could not relate to any variable symbols in the relevant equations.

2.) I also noticed in the XBlog.txt the following new variables: 'veguntow,' 'vegnonlin,' and 'vegcanflo'. Could anyone please explain what these "new" parameters mean. How is each defined and what aspect of the model for the interaction between waves and vegetation do they describe?
Arnold van Rooijen, modified 4 Years ago.

RE: Vegetation

Swell Posts: 239 Join Date: 11/7/11 Recent Posts
Hi Drandreb,

You are right, we have been working on further developing the vegetation module in the past months. These changes (which are not in the Kingsday Release version yet, but are slowly being merged into the trunk version) will be part of the next official release, and are therefore not included in the manual yet.

1) Yes we adjusted the specification to better match the Delft3D standard, and for future developments regarding coupling with sediment transport and morphology. In the new version you will have to specify vegetation characteristics per point rather than section. That means you will always have to specify at least two points, in between XBeach will interpolate the properties if they vary over the vertical (e.g. with mangroves). Zv is the vertical coordinate [m]. So in case of seagrass of 0.2 m high this will be specified as zv = 0 0.2. You will then have to specify the properties for both vertical points (e.g. N = 1000 1000).

2) the other variables you noticed are related to the way XBeach computes the effect of flow and IG waves through vegetation (resp. effect of the undertow, nonlinear waves and canopy flow). They need still some testing so I would recommend you to leave them default (=0) for now.

Hope this helps, let me know if you have any more questions.

Cheers,
Arnold
Arnold van Rooijen, modified 4 Years ago.

RE: Vegetation

Swell Posts: 239 Join Date: 11/7/11 Recent Posts
Hi Drandreb

To make things a little easier I decided to add some code to make the vegetation input backwards compatible. So if you use the latest trunk version you should now be able to specify the vegetation also as described in the Kingsday manual.

Cheers,
Arnold
AG
Alireza Gh, modified 3 Years ago.

RE: Vegetation

Capillary Posts: 11 Join Date: 3/2/13 Recent Posts
Hi Arnold,

I am trying to run a test with two classes of vegetation over the domain in MPI. However, the runs fail immediately after the reading of the vegetation parameters from the input files and give an error related to MPI.
I've tried to run with both Groundhog and Kingsday version:

Ground_Hog error:

XBeach reading from seagrass.txt
nsec = 1 (no record found, default value used)
ah = .2000
Cd = 1.0000
bv = .0200 Warning: value < recommended value of .0500
Fatal error in PMPI_Bcast: Other MPI error, error stack:
PMPI_Bcast(2434)........: MPI_Bcast(buf=0x7f01458dc2e0, count=861624, MPI_BYTE, root=0, MPI_COMM_WORLD) failed
MPIR_Bcast_impl(1807)...:
MPIR_Bcast(1835)........:
I_MPIR_Bcast_intra(2016): Failure during collective
MPIR_Bcast_intra(1596)..:
MPIR_Bcast_binomial(256): message sizes do not match across processes in the collective routine: Received 4 but expected 861624
bv = .0200 Warning: value < recommended value of .0500
N = 1200.0000 Warning: value > recommended value of 20.0000
N = 1200.0000 Warning: value > recommended value of 20.0000
XBeach reading from mangrove.txt
nsec = 3
ah = .5000
ah = .8000
ah = 1.3000
Cd = 2.0000
Cd = 1.0000
Cd = 2.0000
bv = .0500
bv = .1500
bv = .1000
forrtl: error (69): process interrupted (SIGINT)



KingsDay error:

XBeach reading from seagrass.txt
nsec = 1 (no record found, default value used)
ah = .2000
Cd = 1.0000
bv = .0200 Warning: value < recommended value of .0500
bv = .0200 Warning: value < recommended value of .0500
N = 1200.0000 Warning: value > recommended value of 20.0000
N = 1200.0000 Warning: value > recommended value of 20.0000
MPI process # 0 in xmpi_ocomm generated an error: 606696719
Other MPI error, error stack:
PMPI_Bcast(2434)........: MPI_Bcast(buf=0x7f76ed0
dbbe0, count=74272, MPI_BYTE, root=1, MPI_COMM_WORLD) failed
MPIR_Bcast_impl(18
07)...:
MPIR_Bcast(1835)........:
I_MPIR_Bcast_intra(2016): Failure during co
llective
MPIR_Bcast_intra(1596)..:
MPIR_Bcast_binomial(256): message sizes do
not match across processes in the collective routine: Received 8 but expected 7
4272
halt_program called by process 0



Have you seen this problem before? any idea?


Thanks
Arnold van Rooijen, modified 3 Years ago.

RE: Vegetation

Swell Posts: 239 Join Date: 11/7/11 Recent Posts
Hi Alireza,

I think you find a bug related to MPI. It's not something I can fix right away, so I will have a look into this.

In the meantime, can you test if your model works without MPI?

Thanks
Arnold

PS: please use the latest trunk version (under daily builds) when using vegetation as some minor improvements have been made after the last official release version (Kingsday)
You do not have permission to access the requested resource.