The satellite picture on the OAFS homepage was showing
lots of white, cloud-like pixels when it's been clear. Annoying.
So, I've made a couple of changes to the code that generates it.
I've futzed around with the image processing of the raw goes8
images until I've got something that seems to match clear skies at
night.
This is how to read the image:
The darker a patch is on the image, the clearer the sky should be.
Black means clear.
White and gray patches are clouds or fog. This image is mostly
sensitive to low clouds. I'm not yet sure it detects cirrus properly.
Please let me know how well you think it matches the real sky.
Puke green areas are clouds too. They are thick enough to make
me want to pu... (actually, my contrast stretch algorithm
just runs out of brightness bits).
Usually, a line of yellow arrows will appear pointing towards
ottawa. These represent the path that clouds approaching ottawa are
moving in. I hope to use the arrows to tell if those juicy sucker
holes are really comming our way.
If you see the arrows running
amok, or no arrows at all, it just means that the cloud-movement
agorithim is too dumb to figure out the cloud movement for
that image. In that case, I suggest a using weather report to
find the direction of the prevaling wind. The cloud-movement
algorithm usually screws up only when its clear over a very large
area -- when we are least likely to be indoors
looking at silly satellite pictures.
I hope you guys find the images useful. Feel free to let me know
when your observations of the sky dont match the image. I might
be able to tweak it further.
-ad
ps. Three shall be the Hicksons. And the Hicksons shall be thee.
Hey Mike, thanks for great observing friday night.
Care to post a report?
