[GRASSLIST:646] Re: running r.topmodel

GRASS GRASS user
1

Hi Jaime and the rest of the list,

Just to say that I'm now one of the topmodel developers at Lancaster. We are doing a major cleanup of the code and intend to release some new versions soon, although in R rather than in GRASS (for several reasons...) incorporating Monte Carlo and other stuff.

BTW, there is a bug in the r.topmodel infiltration module which Huidae Cho fixed some weeks ago but probably didn't make it to the official version yet.

Anyway, if you have more questions about topmodel, feel free to ask via the list or directly to me.

cheers
wouter

On Thu, 30 Mar 2006, Thomas Adams wrote:

Jaime,

I believe I can help you, although:

(1) I have never run TOPMODEL in GRASS (just the Fortran standalone version)

(2) the link to Lancaster Univ. is old, in fact I'm not sure Kieth Bevan
Lancaster Univ. even make the original source available now, from what I can
see at: http://www.es.lancs.ac.uk/hfdg/freeware/hfdg_freeware_top.htm

My guesses are these, since they conform to the GRASS r.topmodel manual page
and the original files for the standalone version of TOPMODEL:

parameters=3Dstring
(i) TOPMODEL Parameters file, where the standalone file looked like this:

SUBCATCHMENT 1
0.032 5.0 50. 3600.0 3600.0 0.05 0.0000328 0.002 0 1.0 0.02 0.1

I can not remember what's what; I'll have to dig this up, plus I'll have to
look at the r.topmodel source code more =97 I believe the data are read in this
order from what I can see in the source=85

input=3Dstring
(i) Rainfall and potential evapotranspiration data file {probably a space
delimited file with 2 columns}
output=3Dstring
(o) Output file {looks like flow is cu meters/timestep}
Qobs=3Dstring
(i) OPTIONAL Observed flow file {single column file}

My recollection is that the inputs are in terms of meters/time-step (flow
would be cu. meters/delta-t) =97 but I MUST CONFIRM THIS!

The original input file in the standalone version combined the GRASS input
Qobs files and looked like this:

950 1.0 <=3D=3D=3D=3D=3D=3D Number of timesteps & Number of hours per time step (I
believe)
.0000000 .0000619 .0000328
.0000000 .0000647 .0000325
.0010000 .0000619 .0000330
.0010000 .0000560 .0000333
.0010000 .0000457 .0000348
.0005000 .0000323 .0000356
.0005000 .0000156 .0000356
.0000000 .0000000 .0000356
.0005000 .0000000 .0000351

Notice that the values are very small, which makes me believe that I think my
guess about the units of the input are correct; this could be different in
the GRASS r.topmodel code, of course. The columns are, respectively,
precipitation, PE, and ObsQ.

At least this can get you going. I'll try to get more on this in the

next few

days. With a little help of some of the more experienced GRASS module=20
contributors, I don't mind taking this on.

Regards,
Tom

Jaime Carrera wrote:

Hi,

Has any one used the r.topmodel module? If so, I have the following
questions:
1) What are the instructions that are required in the "topmodel paramet=

ers

file"?
2) How should the topographic index statistics file be written?

Cheers,

Jaime

2

Hi Wouter,

Good to hear from TOPMODEL developers :).

May I use this occasion to ask you what is the current location of
"Main algorithm sources are rewritten in C based on TMOD9502.FOR" as
described on
http://grass.itc.it/grass61/manuals/html61_user/r.topmodel.html?

Currently it is linked under
http://www.es.lancs.ac.uk/hfdg/topmodel.html there, but this one is
dead. I'd like to update this information in r.topmodel manual.

Thanks,
Maciek

--------------------
W polskim Internecie s? setki milion?w stron. My przekazujemy Tobie tylko najlepsze z nich!
http://katalog.panoramainternetu.pl/

3

Wouter,

That's excellent; the choice of going to R is pretty interesting. Does this mean the Glue algorithms will be ported there as well?

Regards,
Tom

Wouter Buytaert wrote:

Hi Jaime and the rest of the list,

Just to say that I'm now one of the topmodel developers at Lancaster. We are doing a major cleanup of the code and intend to release some new versions soon, although in R rather than in GRASS (for several reasons...) incorporating Monte Carlo and other stuff.

BTW, there is a bug in the r.topmodel infiltration module which Huidae Cho fixed some weeks ago but probably didn't make it to the official version yet.

Anyway, if you have more questions about topmodel, feel free to ask via the list or directly to me.

cheers
wouter

On Thu, 30 Mar 2006, Thomas Adams wrote:

Jaime,

I believe I can help you, although:

(1) I have never run TOPMODEL in GRASS (just the Fortran standalone version)

(2) the link to Lancaster Univ. is old, in fact I'm not sure Kieth Bevan
Lancaster Univ. even make the original source available now, from what I can
see at: http://www.es.lancs.ac.uk/hfdg/freeware/hfdg_freeware_top.htm

My guesses are these, since they conform to the GRASS r.topmodel manual page
and the original files for the standalone version of TOPMODEL:

parameters=3Dstring
(i) TOPMODEL Parameters file, where the standalone file looked like this:

SUBCATCHMENT 1
0.032 5.0 50. 3600.0 3600.0 0.05 0.0000328 0.002 0 1.0 0.02 0.1

I can not remember what's what; I'll have to dig this up, plus I'll have to
look at the r.topmodel source code more =97 I believe the data are read in this
order from what I can see in the source=85

input=3Dstring
(i) Rainfall and potential evapotranspiration data file {probably a space
delimited file with 2 columns}
output=3Dstring
(o) Output file {looks like flow is cu meters/timestep}
Qobs=3Dstring
(i) OPTIONAL Observed flow file {single column file}

My recollection is that the inputs are in terms of meters/time-step (flow
would be cu. meters/delta-t) =97 but I MUST CONFIRM THIS!

The original input file in the standalone version combined the GRASS input
Qobs files and looked like this:

950 1.0 <=3D=3D=3D=3D=3D=3D Number of timesteps & Number of hours per time step (I
believe)
.0000000 .0000619 .0000328
.0000000 .0000647 .0000325
.0010000 .0000619 .0000330
.0010000 .0000560 .0000333
.0010000 .0000457 .0000348
.0005000 .0000323 .0000356
.0005000 .0000156 .0000356
.0000000 .0000000 .0000356
.0005000 .0000000 .0000351

Notice that the values are very small, which makes me believe that I think my
guess about the units of the input are correct; this could be different in
the GRASS r.topmodel code, of course. The columns are, respectively,
precipitation, PE, and ObsQ.

At least this can get you going. I'll try to get more on this in the

next few

days. With a little help of some of the more experienced GRASS module=20
contributors, I don't mind taking this on.

Regards,
Tom

Jaime Carrera wrote:

Hi,

Has any one used the r.topmodel module? If so, I have the following
questions:
1) What are the instructions that are required in the "topmodel paramet=

ers

file"?
2) How should the topographic index statistics file be written?

Cheers,

Jaime

--
Thomas E Adams
National Weather Service
Ohio River Forecast Center
1901 South State Route 134
Wilmington, OH 45177

EMAIL: thomas.adams@noaa.gov

VOICE: 937-383-0528
FAX: 937-383-0033

4

Hi Tom,

Moving to R is a personal choice, rather than one of the group. But as well, there's not one topmodel but a whole bunch of concepts, of which a few happened to make it to the internet...

I chose R primarily for uncertainty analysis: in R it is very easy to pull parameters from whatever probability distribution you want, construct a parameter matrix and do a Monte Carlo run over the whole matrix. Similarly, you can feed the modelling results directly to whatever likelihood function that is available (or that can be programmed) in R.

As a result, it should be quite straightforward to use a GLUE approach within this framework...

cheers
wouter

On Mon, 10 Apr 2006, Thomas Adams wrote:

Wouter,

That's excellent; the choice of going to R is pretty interesting. Does this mean the Glue algorithms will be ported there as well?

Regards,
Tom

Wouter Buytaert wrote:

Hi Jaime and the rest of the list,

Just to say that I'm now one of the topmodel developers at Lancaster. We
are doing a major cleanup of the code and intend to release some new
versions soon, although in R rather than in GRASS (for several reasons...)
incorporating Monte Carlo and other stuff.

BTW, there is a bug in the r.topmodel infiltration module which Huidae Cho
fixed some weeks ago but probably didn't make it to the official version
yet.

Anyway, if you have more questions about topmodel, feel free to ask via
the list or directly to me.

cheers
wouter

On Thu, 30 Mar 2006, Thomas Adams wrote:

> Jaime,
> > I believe I can help you, although:
> > (1) I have never run TOPMODEL in GRASS (just the Fortran standalone > version)
> > (2) the link to Lancaster Univ. is old, in fact I'm not sure Kieth Bevan
> Lancaster Univ. even make the original source available now, from what I > can
> see at: http://www.es.lancs.ac.uk/hfdg/freeware/hfdg_freeware_top.htm
> > My guesses are these, since they conform to the GRASS r.topmodel manual > page
> and the original files for the standalone version of TOPMODEL:
> > parameters=3Dstring
> (i) TOPMODEL Parameters file, where the standalone file looked like > this:
> > SUBCATCHMENT 1
> 0.032 5.0 50. 3600.0 3600.0 0.05 0.0000328 0.002 0 1.0 0.02 0.1
> > I can not remember what's what; I'll have to dig this up, plus I'll have > to
> look at the r.topmodel source code more =97 I believe the data are read > in this
> order from what I can see in the source=85
> > input=3Dstring
> (i) Rainfall and potential evapotranspiration data file {probably a > space
> delimited file with 2 columns}
> output=3Dstring
> (o) Output file {looks like flow is cu meters/timestep}
> Qobs=3Dstring
> (i) OPTIONAL Observed flow file {single column file}
> > My recollection is that the inputs are in terms of meters/time-step > (flow
> would be cu. meters/delta-t) =97 but I MUST CONFIRM THIS!
> > The original input file in the standalone version combined the GRASS > input
> Qobs files and looked like this:
> > 950 1.0 <=3D=3D=3D=3D=3D=3D Number of timesteps & Number of hours per > time step (I
> believe)
> .0000000 .0000619 .0000328
> .0000000 .0000647 .0000325
> .0010000 .0000619 .0000330
> .0010000 .0000560 .0000333
> .0010000 .0000457 .0000348
> .0005000 .0000323 .0000356
> .0005000 .0000156 .0000356
> .0000000 .0000000 .0000356
> .0005000 .0000000 .0000351
> > Notice that the values are very small, which makes me believe that I > think my
> guess about the units of the input are correct; this could be different > in
> the GRASS r.topmodel code, of course. The columns are, respectively,
> precipitation, PE, and ObsQ.
> > At least this can get you going. I'll try to get more on this in the
next few
> days. With a little help of some of the more experienced GRASS module=20
> contributors, I don't mind taking this on.
> > Regards,
> Tom
> > Jaime Carrera wrote:
> > Hi,
> > > > Has any one used the r.topmodel module? If so, I have the following
> > questions:
> > 1) What are the instructions that are required in the "topmodel > > paramet=
ers
> > file"?
> > 2) How should the topographic index statistics file be written?
> > > > Cheers,
> > > > Jaime

--
Thomas E Adams
National Weather Service
Ohio River Forecast Center
1901 South State Route 134
Wilmington, OH 45177

EMAIL: thomas.adams@noaa.gov

VOICE: 937-383-0528
FAX: 937-383-0033