There really isn't that much fuel for a bushfire that large. Most fires up there last 2 mins and all the spinifex is consumed. Looks doctored to me or a real camera fail.
I wonder if it may be at least partly an algorithm fail.
My understanding is that they took pictures on lots of of nights and then used used the results that had the brightest pictures to be included in the final picture.
For the rest of the world, this means that the nights most free of clouds were used.
For Australia, it means that if there was a bushfire on any of the nights that a region was being imaged, then that nights image would be used, as it would obviously be brighter than every other night.
Even if that's the case, there still seem to be too many fires in the west - there should be relatively more in Norther Queensland surely?
The bushfires would be massive - bigger than anything we have ever had. Those lights in WA dwarf the metropolitan areas of Sydney, Brisbane and Melbourne combined - several times over. Same problem with the idea of mine sites. The areas involved are just too big.
Up in the Pilbara region i would expect some lights
as those places dont know the meaning of reducing
light pollution, but the lower middle regions are
not that bright.
Which brings to light the question,
where are places like Moomba and the Qld coal mines
which are also rather "bright" at night.
Andrew
comparing this piccy to prev "visible spectrum" sky at night piccies
its a very different result.
Certainly a lot of people here scratching their heads-and for good reason,the only plausible explanation is lightning from thunderstorms,but that seems odd-its the desert not much thunderstorm activity there.
Really does add weight to many other IIS members thoughts that APOD images are slipping (personally I did not agree with their thoughts on Tuc47),but many here did,and this image is very questionable-I've seen similar styles of images-and Sydney has much more light that desert country.
Just had a look on Google Earth, and to me, it looks like possibly moonlight being reflected off dry salt lakes and small lakes in the area...
Maybe, because the surrounds are so dark, the heat and light signatures stand out to the satellites like proverbial canine testicles, and cause a hit from the algorythm??
Do a google search for Plumridge Lakes just NE of Kalgoolrie, Ruddall river National Park (Lake Dissapointment, Lake Blanche and Lake Dora) and another big hit around Lake Mackay, all look to me to be in the approximate areas of some of the bigger light areas in WA... (I layered the NASA image over a Google map image and attached it)...
Plenty of small reflective salt deposits out there...I dont know exactly if thats it, but thats my thoughts..
Edit: Also just spotted a light hit near the area of the Bonneville Salt Flates in America.. Could be Salt Lake city, but seems pretty big and bright...
On reading the caption, I note that this image is composed of many images taken over a period between April and October, So all the bush fires and other lights are a composition, so they were not all there all at once as is portrade in that image.
Cheers
Certainly a lot of people here scratching their heads-and for good reason,the only plausible explanation is lightning from thunderstorms,but that seems odd-its the desert not much thunderstorm activity there.
Actually thunderstorms are far from rare in those areas. Look at the lightning tracker and radars on Weatherzone and BOM and you'll often see storm activity in the area.
Check out the BOM's annual thunder day map, there is a clear increase in storm days in the region which correlates almost exactly with the lights in the image.
the links an interesting read-looks like astronomers would have even more light pollution if they lived near a dictator.What a odd bit of information,certainly a well written piece.
the question,
they must be in plaque preportions out west 
