VIIRS HDF5 Data from NOAA's CLASS

1868
3
02-14-2014 09:02 AM
Highlighted
New Contributor
I have been downloading VIIRS HDF5 data from NOAA's CLASS repository and the files are not correctly displayed in ArcMAP 10.1. In one instance, I overlaid a top of atmosphere NDVI map of California over a polygon dataset of countries that I trust. Even though both datasets are Geographic (WGS'84), the VIIRS EDR map appeared over Turkmenistan. In more detail, a map whose most western extent was about -130 E (or 130 W) became 50 E instead. (It happened other times and always added up to just about 180. It usually isn't exactly 180, which might just be rounding/estimation.)

Have other people experienced this? Does someone have an explanation, and how about a solution too?

Thnaks.
Reply
0 Kudos
3 Replies
Highlighted
Esri Notable Contributor
I'm not sure what to do about it, but I've seen other data from NGS (part of NOAA) that uses positive west longitude values. That would fit with this data. Does the data look flipped too? (I know it's probably hard to tell).

Melita
Reply
0 Kudos
Highlighted
New Contributor
If it were a positive W issue wouldn't it appear as 130 E instead of 50 E? Am I missing something? These are swaths, which, to me at least, means that the lat/long values are separate layers in the file. Usually I don't work with swaths, so I think of the coordinate system usually being described in the metadata and Arc then calculates the lat/long for each point. Is that perception correct? What does Arc do with a swath, if I am correct in thinking that the lat/long data is only presented as separate file layers? To be more exact, at the western most extent of the map, Arc claims it is 44.45E while the longitude layer value is -135.968. On the eastern most extent, Arc = 77 and long layer = -102.876. (Keep in mind that I get these values using the Inquire tool, so they probably aren't exactly the western most and eastern most values.) It seems conspicuous that each pair of values almost add up to 180 deg, but I haven't been able to think of a reason why. One curiosity that persistently nags at me is that the values in the lat/long layers do range from -180 to 180, which is what Arc expects, so why does it think that it is seeing long values of 44.45 to 77 instead?

To answer your question, the state is correctly oriented, so I'd say that the data is not flipped. I've attached an image so that you can see as well.[ATTACH=CONFIG]31467[/ATTACH]
Reply
0 Kudos
Highlighted
Esri Notable Contributor
Very true. I'm not sure I can figure this out late on a Friday afternoon. Possibly a central meridian/longitude of origin that's at 180? But misread as 0?

I see references to ground track Mercator "projection", but it looks like this data has already been converted from that.

Melita
Reply
0 Kudos