[GRASS-user] question on Stray light Landsat 8 data

[Gabriel Cotlier]

Hello, another question, regarding  *i.nightlights.intercalibration, *can I
run this code as python package/lbrary loading it from Spyder or Jupiter
Notebook instead of using GRASS interface, if so how is a convenient  way
to install   *i.nightlights.intercalibration *in python using Spyder?
Thanks a lot.
Gabriel

On Sat, Aug 17, 2019 at 4:54 PM Gabriel Cotlier <gabiklm01 at gmail.com> wrote:

> Dear Nikos.
> After a long time I'm trying to reproduce a routine I have for doing
> intercallibratrion of DMSP 1992-2012 but for some reason It doesn't work to
> me. I think is because the problem between the region of the layers 30 arc
> sec should resolution be from 0.008333333300000 to 0.008333333333333, i.e.
> exactly 30 arc-seconds? and the computational region be the same ? I got
> stuck on how to set it to work... from the side of the region setting.
> However in addition my routing also has a for loop which does not work ok
> as well.
> I would appreciate a lot of you can give it a look and tell me how to make
> it work...
> Thanks  a lot in advance
> Kind regards,
> Gabriel
>
> #####-----------------------------------------------------------------------------------------
> # complete routine for intercalliration of DSMP/OLS light stable product
>
> import grass.script as gscript
> import os
> import os,glob
>
> # get working directory
> print os.getcwd()
>
> # change working directory where raster files are
> os.chdir('C:\\Users\\Gabriel\\Documents\\grassdata\\lights')
>
> # see files in directory
> ls
>
> # import all raster files to grass --- here is a kind of problem...???
> for tif_file in glob.glob("*.tif"):
>     new_rast = os.path.splitext(tif_file)[0]
>     grass.run_command("r.in.gdal", flags="a", input=tif_file,
> output=new_rast)
>
> # get info of one of the imported raster
> r.info map=F121996
>
> # run intercalliration algorithm
> i.nightlights.intercalibration
> image=F101992,F101993,F101994,F121994,F121995,F121996,F121997,F121998,F121999,F141997,F141998,F141999,F142000,F142001,F142002,F142003,F152000,F152001,F152002,F152003,F152004,F152005,F152006,F152007,F162004,F162005,F162006,F162007,F162008,F162009,F182010,F182011,F182012,F182013
> suffix=c model=elvidge2014 -t
>
> # correct general region adjust to raster file --- here the region is
> exactly 30 arc for the raster as I could see....
> g.region raster=F121996
>
> # cerate a list of rasters in the mapset
> # rastlist=grass.read_command("g.list",type="rast").split()
>  rasters = grass.read_command('g.list', type='raster').splitlines()
>
> # change working directory
> os.chdir('C:\\Users\\Gabriel\\Desktop\\out')
>
> # save rasters in mapset to file
> for raster in rasters:
>     grass.run_command('r.out.gdal', input=raster, output=raster + '.tiff',
> format='GTiff')
>
> On Wed, Aug 22, 2018 at 10:06 AM Gabriel Cotlier <gabiklm01 at gmail.com>
> wrote:
>
>> Dear Nikos,
>>
>> Thanks a lot for your answer and the orientation.
>> The information and the link are very useful.
>> Kind regards,
>> Gabriel
>>
>>
>> On Wed, Aug 22, 2018 at 5:19 AM Nikos Alexandris <nik at nikosalexandris.net>
>> wrote:
>>
>>> * Gabriel Cotlier <gabiklm01 at gmail.com> [2018-08-21 12:00:24 -0300]:
>>>
>>> >Dear Nikos and GRASS users,
>>> >
>>> >I would like to ask if nonetheless the effect due to "stray light" the
>>> >*i.landsat8.swlst* code for split window is still applicable to Landsat
>>> 8
>>> >data and whether these error is specially visible on water bodies? and
>>> >whether band 10 is better than band 11 in terms of correction processing
>>> >for Level -1  data products?
>>> >
>>> >Thanks a lot.
>>> >
>>> >Kind regards,
>>> >Gabriel
>>>
>>> Dear Gabriel,
>>>
>>> for details and references, refer to
>>>
>>>
>>> https://landsat.gsfc.nasa.gov/landsat-8-thermal-data-ghost-free-after-stray-light-exorcism/
>>>
>>> Make sure you use the newest Level-1 Collection 1 Landsat 8 products.
>>>
>>> I use `i.landsat8.swlst` and plan to improve it further.
>>>
>>> However, whether to prefer a Split-Window based approach, or another
>>> Single-Channel one, depends on what you want to do. Think of spatial
>>> extent and coverage of various land (cover) types, temporal extent
>>> and more.
>>>
>>> Thermal remote sensing is hard(er) also because it's hard to get
>>> ground-truth data sets so as to validate LST estimations.
>>>
>>> Nikos
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-user/attachments/20190818/0d211feb/attachment.html>