[GRASS-user] i.atcorr with MSS band 5,6,7

GRASS GRASS user
1

Dear Grass users,
how can I add spectral conditions of the bands 5,6,7 of the Landsat MSS
sensor?

I have seen that these are defined in:
http://trac.osgeo.org/grass/browser/grass/branches/develbranch_6/imagery/i.atcorr/Iwave.cpp

But is is unclear to me how they are added.

As far as the manual concerned, the pre-defined spectral conditions are
only available to LS MSS band 1-4:

http://grass.itc.it/grass64/manuals/html64_user/i.atcorr.html
31 mss (landsat5) band 1 (0.475-0.640)
32 mss (landsat5) band 2 (0.580-0.750)
33 mss (landsat5) band 3 (0.655-0.855)
34 mss (landsat5) band 4 (0.785-1.100)

Thanks in advance for any hints,
Marco

2

I thought the MSS sensor only had 4 bands, and the TMS sensor sensor 7?

Best, MEH

Dear Grass users,
how can I add spectral conditions of the bands 5,6,7 of the Landsat MSS
sensor?

I have seen that these are defined in:
http://trac.osgeo.org/grass/browser/grass/branches/develbranch_6/imagery/i.atcorr/Iwave.cpp

But is is unclear to me how they are added.

As far as the manual concerned, the pre-defined spectral conditions are
only available to LS MSS band 1-4:

http://grass.itc.it/grass64/manuals/html64_user/i.atcorr.html
31 mss (landsat5) band 1 (0.475-0.640)
32 mss (landsat5) band 2 (0.580-0.750)
33 mss (landsat5) band 3 (0.655-0.855)
34 mss (landsat5) band 4 (0.785-1.100)

Thanks in advance for any hints,
Marco

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

3

I thought the MSS sensor only had 4 bands, and the TMS sensor sensor 7?

Sorry, I heven't expressed myself clearly.

I am working with old images from the 1970ies recorded by Landsat 1.
According to
http://landsat.gsfc.nasa.gov/about/mss.html
http://eros.usgs.gov/#/Guides/landsat_mss
the MSS used the bands 4-8.

I am mislead somehow or wrong?

If not I would be glad to receive some advice on how to add these
additional spectral characteristsics to the module.

Best regards,
Marco

4

i.atcorr supports the MSS sensor aboard Landsat 5 and is not applicable to Landsat 1 data (nor landsat's 2,3 or 4). In-order to atmospherically correct landsat 1 data with i.atcorr you'll need to add support for it. It's not hard and I would be willing to spare what free time I can to help you but the first step, and possibly the most difficult, is to find the spectral response curves for the RBV and MSS sensors aboard Landsat 1.

Cheers,

Mike
On 1-Dec-09, at 3:29 PM, Marco Tuckner wrote:

I thought the MSS sensor only had 4 bands, and the TMS sensor sensor 7?

Sorry, I heven't expressed myself clearly.

I am working with old images from the 1970ies recorded by Landsat 1.
According to
http://landsat.gsfc.nasa.gov/about/mss.html
http://eros.usgs.gov/#/Guides/landsat_mss
the MSS used the bands 4-8.

I am mislead somehow or wrong?

If not I would be glad to receive some advice on how to add these
additional spectral characteristsics to the module.

Best regards,
Marco

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

5

Hello,
thanks for your response.

i.atcorr supports the MSS sensor aboard Landsat 5 and is not applicable
to Landsat 1 data (nor landsat's 2,3 or 4). In-order to atmospherically
correct landsat 1 data with i.atcorr you'll need to add support for it.

Yes, I got that from the help page.

It's not hard and I would be willing to spare what free time I can to
help you but the first step,

Thanks for your offer. Very generous.

and possibly the most difficult, is to find
the spectral response curves for the RBV and MSS sensors aboard Landsat 1.

Does anyone have an idea where I can find these response curves?
Where are the current pre-defined band configurations from?

Kind regards,
Marco

6

Dear list,

and possibly the most difficult, is to find
the spectral response curves for the RBV and MSS sensors aboard Landsat 1.

Does anyone have an idea where I can find these response curves?
Where are the current pre-defined band configurations from?

Sorry, for pushing again on this.
But may someone give me a hint which parameters I need to look for?

How high is the inaccuracy, if I just use the corresponding
pre-definitions for MSS of Landsat 5 (i.e. preset 31-34)?

Thanks in advance,
Marco

7

I'm curious -- why do you need to atmospherically correct MSS? About 95% of RS applications do not benefit from atmospheric correction -- the ones that do are spectral mixture type analyses (where you are using endmembers, and with a 4-band system this isn't going to be very useful), and if the atmo correction is perfect (which it won't be), doing change detection or "recycling" classification rules. If you are planning on just doing classification on the one image, you really don't need to do atmospheric correction. Just wanted to drop this question in the mix, as I've seen many a person go down the long, arduous task of atmospherically correcting imagery only to find out it isn't helpful.

--j

Marco Tuckner wrote:

Dear list,

and possibly the most difficult, is to find
the spectral response curves for the RBV and MSS sensors aboard Landsat 1.
      

Does anyone have an idea where I can find these response curves?
Where are the current pre-defined band configurations from?
    

Sorry, for pushing again on this.
But may someone give me a hint which parameters I need to look for?

How high is the inaccuracy, if I just use the corresponding
pre-definitions for MSS of Landsat 5 (i.e. preset 31-34)?

Thanks in advance,
Marco

_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user
  
--

Jonathan A. Greenberg, PhD
Postdoctoral Scholar
Center for Spatial Technologies and Remote Sensing (CSTARS)
University of California, Davis
One Shields Avenue
The Barn, Room 250N
Davis, CA 95616
Phone: 415-763-5476
AIM: jgrn307, MSN: jgrn307@hotmail.com, Gchat: jgrn307