1

I've searched at length on previous forums but no solution seems to work for me.

I'm trying to run the i.atcorr module on GRASS (version 6.4.3 on a Macbook Pro, Mavericks OSX) with Landsat8 imagery.

Parameters didn't seem to be loading from the provided metadata file (MTL_txt), so I went back to the GRASS manuals and tried to input values by hand, here is what I used:

8 = geometrical conditions=Landsat ETM+

05 04 03.81 98.117 23.131 = month day hh.ddd longitude latitude ("hh.ddd" is in GMT decimal hours)

1 = atmospheric mode=tropical

0 = aerosols model=none (ASSUMING NO AEROSOLS)

0 = visibility [km]

0.732 = aerosol optical depth 550 nm

-1.408 = mean target elevation above sea level [km]

-1000 = sensor on board a satellite

0 = THIS IS WHAT I'M WORRIED ABOUT: There is no pre-defined coding for Landsat 8 satellites and I am unsure how to define my own spectral conditions??

I checked I was in the correct region (i.e. used g.region) and I've tried using both radiance and reflectance at sensor values (computed using i.landsat.toar, which seems to have worked nicely) but the rasters returned are empty :(

Here is what I'm running (in verbose)

i.atcorr -a --overwrite --verbose iimg=Part1toar_B1@Charlie icnd=/Users/charlottewhitham/Documents/PhD/My research/my GIS/grassdata/NGH/Charlie/.tmp/Charlie.local/819.2 oimg=P1toar_atcorrB1 * ****************************** 6s version 4.2b


  • geometrical conditions identity *
  • ------------------------------- *
  • etm+ observation * * *
  • month: 5 day: 4 *
  • solar zenith angle: 23.55 deg solar azimuthal angle: 103.24 deg *
  • view zenith angle: 0.00 deg view azimuthal angle: 0.00 deg *
  • scattering angle: 156.45 deg azimuthal angle difference: 103.24 deg * * *
  • atmospheric model description *
  • ----------------------------- *
  • atmospheric model identity : *
  • tropical (uh2o=4.12g/cm2,uo3 =.247cm-atm) * * *
  • aerosols type identity : *
  • no aerosols computed * * *
  • spectral condition *
  • ------------------ *
  • constant *
  • value of filter function : *
  • wl inf=27640754752056197120.000 mic wl sup= 0.000 mic * * *
  • target type *
  • ----------- *
  • homogeneous ground *
  • constant reflectance over the spectra 0.000 * * *
  • target elevation description *
  • ---------------------------- *
  • ground pressure [mb] 862.22 *
  • ground altitude [km] 1.408 *
  • gaseous content at target level: *
  • uh2o= 2.056 g/cm2 uo3= 0.244 cm-atm * * *
  • atmospheric correction activated *
  • -------------------------------- * Atmospheric correction... Atmospheric correction complete. (Fri Dec 6 12:16:10 2013) Command finished (22 sec)

And here is a description of the output:

| Layer: P1toar_atcorrB1@Charlie Date: Fri Dec 6 12:12:58 2013 | | Mapset: Charlie Login of Creator: charlottewhitha | | Location: NGH | | DataBase: /Users/charlottewhitham/Documents/PhD/My research/my GIS/grassda | | Title: ( P1toar_atcorrB1 ) | | Timestamp: none | |----------------------------------------------------------------------------| | | | Type of Map: raster Number of Categories: 0 | | Data Type: CELL | | Rows: 7381 | | Columns: 7551 | | Total Cells: 55733931 | | Projection: UTM (zone 47) | | N: 2666715 S: 2445285 Res: 30 | | E: 522315 W: 295785 Res: 30 | | Range of data: min = NULL max = NULL | | | | Data Description: | | generated by i.atcorr | | | | Comments: | | i.atcorr -a iimg="Part1toar_B1@Charlie" iscl=0,255 icnd="/Users/char\ | | lottewhitham/Documents/PhD/My research/my GIS/grassdata/NGH/Charlie/\ | | icnd_lsat1.rtf" oimg="P1toar_atcorrB1" oscl=0,255

As you can see the range of data is NULL to NULL... not good.

  • Have you set an appropriate region ? – gene Dec 6 '13 at 18:31
  • Yes, the region is set using the raster layers I'm trying to correct.... – Char_leopard Dec 7 '13 at 15:00
  • 1
    Landsat8 is not yet supported - someone needs to get and properly format the spectral curves, following grasswiki.osgeo.org/wiki/… – markusN Dec 8 '13 at 20:29
  • 1
    This is not true anymore. i.atcorr supports for Landsat8 imagery. – Nikos Alexandris Nov 20 '14 at 19:01
  • @NikosAlexandris I think it should be possible for you to craft an answer of a few sentences to say basically what you have commented. By providing an answer you can potentially earn reputation and also help the asker and future readers better understand the historical nature of the question and currency of your answer. – PolyGeo Feb 13 '15 at 11:02
2

When this question was submitted, there was no support for the recent sensor of the Landsat family.

However, according to http://trac.osgeo.org/grass/changeset/61314, wihch is GRASS GIS' official project management system, the i.atcorr module in GRASS 7, supports for Landsat8 imagery since end of July 2014.

  • Herewith, I kindly ask for to either accept this answer, or close this question if the original poster does not act upon within some reasonable time (given that this is an old question and the poster has no interest any more). – Nikos Alexandris Mar 19 '15 at 8:54
  • 1
    Thank you @Nikos Alexandris. I have accepted your answer and thank you for your time with this - I really appreciate it. – Char_leopard Mar 19 '15 at 13:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.