Earthdata Forum

Question from GEDI NASA Earthdata Webinar: Why are GEDI RH metrics negative for some footprints?
https://earthdata.nasa.gov/learn/user-resources/webinars-and-tutorials

Relative Height is calculated by the following equation: elev_highestreturn - elev_lowestmode. The lower RH metrics (e.g., RH10) will often have negative values, particularly in low canopy cover conditions. This is because a relatively high fraction of the waveform energy is from the ground and below elev_lowestmode. For example, if the ground return contains 30% of the energy, then RH1 through 15 are likely to be below 0 since half of the ground energy from the ground return is below the center of the ground return, which is used to determine the mean ground elevation in the footprint (elev_lowestmode). The RH metrics are intended for vegetated surfaces. Results over bare/water surfaces are still valid but may present some confusing results. See Section 6 of the Level 2 User Guide for more detailed information.

Hello,

To add to this thread, I have been running into issues with retrieving relative height values from the Level2A release. I've subsetted the data for only data with quality flag==1, but found negative values up to rh60. The data from the first release did not have negative values above rh15. Any guidance on this would be greatly appreciated.

Cheers

@ahonda ,

I have a couple questions for you.
1. Are you downloading the source .h5 files first and then processing the files, or are you subsetting the files in Earthdata Search first?
2. Could you share the code you're using to process the data or are you using the GEDI tutorials available here: https://git.earthdata.nasa.gov/projects/LPDUR/repos/gedi-tutorials/browse

Thank you for your message.

1. I downloaded the source .h5 files first, then processed
2. The data was downloaded and processed by using the "rGEDI" package through RStudio. Attached is the code for the download, clipping and extraction of L2A data from the .h5 files.

Not mentioned in the code, but we removed files that could not be processed after clipping.

ahonda wrote:

@afriesz

Thank you for your message.
>
> 1. I downloaded the source .h5 files first, then processed
> 2. The data was downloaded and processed by using the "rGEDI"
> package through RStudio. Attached is the code for the download, clipping
> and extraction of L2A data from the .h5 files.
>
> Not mentioned in the code, but we removed files that could not be processed
> after clipping.