Page 1 of 1

Vertical step in PAI profile

Posted: Sun Jul 09, 2023 2:27 am America/New_York
by jakeeckersley121
Hello,

All documentation I have found for GEDI L2B v2 product (LARSE/GEDI/GEDI02_B_002 accessed via earth engine) so far, as well as the description on the product page (https://developers.google.com/earth-engine/datasets/catalog/LARSE_GEDI_GEDI02_B_002#description) refers to vertical steps (dZ) in the level 2b PAI and PAVD products as always being in 5m intervals.

Is there a script available for reducing/editing that step size? This would be incredibly useful for dryland ecosystems where vegetation rarely exceeds 5 m. I understand this can be built from equations 7, 9 + 10 from the ATBD, however i'm a little lost on where to find pv and pg grids used in the calculation.

Thanks in advance
Jake

Re: Vertical step in PAI profile

Posted: Wed Jul 19, 2023 3:49 pm America/New_York
by LP DAAC - dgolon
Hi @jakeeckersley121 I have passed your question along to our science team. We will follow up with an answer. Thanks - Danielle

Re: Vertical step in PAI profile

Posted: Thu Jan 04, 2024 8:55 am America/New_York
by LP DAAC - afriesz
@jakeeckersley121 ,

I apologize for the long delay. This notebook may provide what you are after:
https://github.com/armstonj/gedipy/blob/master/notebooks/gedi_l2b_vertical_plant_profiles.ipynb. The imports and code related the MAAP can be skipped as they are for search and discovery from the NASA MAAP (https://www.earthdata.nasa.gov/esds/maap) platform. If you have your data already downloaded, you should be able to run the script. Additionally you'll need to install gedipy. Instructions can be found here: https://github.com/armstonj/gedipy/tree/master.

Note however that the vertical resolution of 5m is for a reason. The pulse width of GEDI "blurs" the PAVD and PAI profiles at <5m vertical resolution, and errors in the separation of ground and canopy signals lead to high uncertainties in the 0-5m height bin.