Ordering Planet Data - porder (older)

Derived from: https://samapriya.github.io/projects/porder/ 

To install:

• pip install planet

• planet init

  • log in with usename/password credentials

• pip install porder

Ensure you have your API KEY set up (may not be necessary if logged in as above).

• Set this as a system variable

  • In python:

    • os.environ['PL_API_KEY']='12345'

  • In BASH: 

    • PL_API_KEY = '12345'

Confirm quota

• porder quota

Make list of scenes to order

• porder idlist --input /Users/dan.sousa/Downloads/map.geojson --start 2010-01-01 --end 2019-01-01 --item PSOrthoTile --cmin 0 --cmax 10 --overlap 100 --asset analytic_sr --outfile /Users/dan.sousa/Downloads/PlanetBocaRio_2010_2019.csv

  • --input specifies bounding extent (vector)

  • --item specifies the dataset type (PSScene4Band or orthotile; 3 or 4 or 8 band)

  • --asset specifies the processing type (analytic_sr is surface reflectance; analytic is TOA)

  • --outfile is the output of the scene list

  • --cmin and --cmax are cloud cover (% of scene)

  • --overlap  (% of polygon that is covered by each image)

Split the list if it's too long

• porder idsplit --idlist "/Users/dan.sousa/Downloads/PlanetBocaRio_2010_2019.csv" --lines "500" --local "/Users/dan.sousa/Downloads/PlanetBocaRio_2010_2019_Split"

  • --local is a directory in which to write the sublists

Order the data

• porder order --name "BocaRio_2010_2019_1" --idlist "/Users/dan.sousa/Downloads/PlanetBocaRio_2010_2019_Split/PlanetBocaRio_2010_2019_500.csv " --item "PSOrthoTile" --bundle "analytic_sr_udm2" --boundary "/Users/dan.sousa/Downloads/map.geojson" --op clip evi2 ndwi email

  • --name is the Order Name

  • --idlist is the path to the scene list

  • --item is the dataset type

  • --bundle is the set of files sent with each image

  • --boundary is the polygon with the spatial extent (if being clipped)

  • --op specifies the operations happening to the data before download

    • 'clip' clips them to the spatial extent of the 'boundary'

    • 'email' sends you an email notification once the data are ready for download

    • Other options include spectral indices: ndvi|gndvi|bndvi|ndwi|tvi|osavi|evi2|msavi2|sr

• Or in a loop:

  • for i in /Users/dan.sousa/Downloads/BrianaPlanet/PM3km_B1*.geojson; do porder idlist --input ${i} --start 2016-01-01 --end 2017-12-31 --item PSOrthoTile --cmin 0 --cmax 30 --overlap 60 --asset analytic_sr --outfile ${i%????????}_list.csv; done

• • cd to directory

  • for i in *.geojson; do porder order --name "${i}" --idlist /Users/dan.sousa/Downloads/BrianaPlanet/${i%????????}_list.csv --item PSOrthoTile --bundle analytic_sr --boundary /Users/dan.sousa/Downloads/BrianaPlanet/${i} --op clip evi2 zip email; done

Once done, get the order URL

• porder ostate --state success --start 2023-01-01 --end 2024-01-01

• porder ostate --state partial --start 2023-01-01 --end 2024-01-01

Download the files

• porder multipart --url https://api.planet.com/compute/ops/orders/v2/xxxxx --local /path/to/download/folder

Convert to common (WGS84) projection and ENVI format

• for i in *math.tif; do gdalwarp -t_srs EPSG:4326 -of ENVI ${i} ${i}_e; done

  • Assumes files of interest end in "math.tif"

  • WGS-84 (lat/lon) is set using the EPSG code. Pick another if you want a different projection.

  • Assumes ENVI-formatted output 

Pull out 1 of the bands

• for i in *_e; do gdal_translate -b 1 -of ENVI ${i} ${i}EVI2; done

  • Assumes you want band 1. Change the number after -b to change the band

Stack all the single-band images into a single multiband timeseries 

• gdal_merge.py -o 2020_2021_PlanetSedg_EVI2stack -separate -of ENVI *_eEVI2