unmatched tif dimensions

[perigeolee]

Hi all, I am using AppEEARS to download ECOSTRESS data for a specific shapefile, but I have encountered an issue with the dimensions of the downloaded images. For instance, the ECOSTRESS data for April 2020 in this specific region comprises seven files, with four files having dimensions of 434x667 and three files having dimensions of 434x666. Can anyone advise how to resolve such an issue? Thanks!

[LP DAACx - dgolon]

Hello @perigeolee Could you please let us know your AppEEARS order number or the email address you used to create the order? You can either reply to this post and select "Submit as private post" or email as directly at lpdaac@usgs.gov as we'd like to look into this. Thanks --Danielle

[perigeolee]

Hi Danielle, thanks a lot for the kind help! I didn't find an order number as you mentioned, but I found the ID of the request: 65446d3c-79a0-462b-897d-2e7bd1c60f9c. Let me know if you need further information.

[LP DAACx - dgolon]

Thank you for providing that request ID. Our developers are taking a look. Please note: we have a holiday coming up here. so responses may be delayed but we should have an answer by mid-week next week. Thanks -- Danielle

[LP DAACx - dgolon]

Hi @perigeolee Here is what I heard back from the team. This is expected behavior. This stems from the fact that we do not map ECOSTRESS version 1 scenes to a common grid. Each scene is mapped to its own unique grid based on the spatial parameter for that ECOSTRESS scene. This has a couple implications. First, the origin points for the ECOSTRESS outputs will very slightly cause an observable shift in the grids from scene to scene. Second, the pixels from scene to scene do not align/snap together. If your analysis relies on these data being snapped, you will first need to map the outputs to a common grid. If you need identical dimensions across all of the files, you'll likely need to clip those re-gridded scenes to your AOI. ECOSTRESS version Gridded products are supposed to address this as they are all mapped to a common grid. Version 2 data is still be processed and delivered to the LP DAAC, so the collection is not quite complete for the Gridded products, but they could be an option if they meet your spatiotemporal requirements.

[perigeolee]

Thanks for the detailed explanation!