Subset MODIS L2 files for Arctic include data over dateline

Use this Forum to find information on, or ask a question about, NASA Earth Science data.
Post Reply
dbotis
Posts: 38
Joined: Mon Nov 21, 2011 7:44 pm America/New_York
Answers: 0

Subset MODIS L2 files for Arctic include data over dateline

by dbotis » Fri Sep 18, 2020 5:40 pm America/New_York

Hi,

I am trying to get MODISA L2 data over the Chukchi Sea and then mosaic by day. In my orders (extracted) I make sure to get things cut off east of the dateline, to avoid reprojection/mosaic issues that occur when trying to map data from both sides of the dateline. However, even though I set the west boundary at -179 or even -177, some of the files I receive bleed over into positive longitudes. This creates problems and gives weird results during my subsequent mosaic routine (via gpt). I can cut the files myself using the gpt subset operator, but when I do, the mosaic routine takes at least 10 times as long (~10-12 min/day), but the results do turn out better.

What is the best solution here? How can I avoid getting files in my orders with positive longitudes? Do I need to set the west bound of my orders farther east (-175?).

Any ideas are appreciated.

Dan

Tags:

OB.DAAC - SeanBailey
User Services
User Services
Posts: 1464
Joined: Wed Sep 18, 2019 6:15 pm America/New_York
Answers: 1
Been thanked: 5 times

Subset MODIS L2 files for Arctic include data over dateline

by OB.DAAC - SeanBailey » Mon Sep 21, 2020 4:26 pm America/New_York

Dan,

The L1 extraction will never eliminate all data outside the defined region, as it must retain the same length for all scan lines.  Additionally, there is a minimum size for a MODIS extract (so that it contains enough scans to be able to be calibrated).  So, you're stuck there unless you can move your region enough away from the dateline :cry:

Are you stuck with GPT for doing the mosaic?  If you use l2bin/l3mapgen you will avoid many of the the projection pitfalls you're encountering.

Sean

Post Reply