Skip to content

springinnovate/pestcontrol_livingdatabase

Repository files navigation

pestcontrol_livingdatabase

For the USDA FACT project!

ee_sampler originally came from https://github.com/therealspring/earth_engine_scripts

lingling_scripts are also forked on https://github.com/lingling-liu/pestcontrol_livingdatabase

Process for building table:

  • Inputs

    • CSV table of points with associated years
    • list of landcover datasets to build additional masks
  • Outputs

    • CSV table, by point, of aggregated values of MODIS parameters:
      • MODIS phenology parameters: Greenup_1, MidGreenup_1, Peak_1, Maturity_1, MidGreendown_1, Senescence_1, Dormancy_1, EVI_Minimum_1, EVI_Amplitude_1, EVI_Area_1, QA_Overall_1.
      • MODIS parameters masked by landcover type for each of the landcover datasets passed as input (currently these landcover types are cultivated and natural, but dataset configuration files can easily take more

Action:

  1. Build GEE MODIS multi-band raster for each of the parameters described above
  2. For each of the bands in #1 2a) For each of the datasets listed in inputs 2aa) each of the landcover types listed above, create a new band that is masked by the landcover type from the given dataset
  3. Load points from input CSV table
  4. Use points from #3 to sample ALL the bands created by #2
  5. For each point sample in #4 write a single row to a table with headers that correspond to all the bands created in #1 and #2.