For those of you wondering, as I was, which one is Eddington Crater...

I had to look it up as well And S_Walker - the answer is about 2hrs, 40 minutes. The post following yours qualifies I think.

Just rendering a near-side 300 frame animation of dawn to dusk with very exaggerated terrain

Thanks. What about those who don't own an iPhone?

There's Google Earth/Moon/Mars

Nah... Google Moon is nice but it doesn't use a topographic map for lighting as far as I remember? And I'm not sure if you can set it to show the view from your backyard.

Can Celestia be set to view the Moon from a position on Earth? At least two of us have already produced normal maps; it's a small step from that to the amazing screenshots that volcanopele was posting.

I imagine what S_Walker really wants is something that does Bjorn- or Doug- type rendering to show the Moon from a location and time on Earth. Don't know about that. But Celestia or Moon Globe would probably be good enough for a lot of purposes.

Similar; I'd like something like Virtual Moon Atlas or RITI Lunar Map Pro so that I can better plot interesting features to observe under favorable librations each month in Chuck Wood's Explore the Moon column in S&T (I edit it each month).

I see. Well as Doug points out, Moon Globe may well be first out the door with those features, for anyone who has an iPhone or iPod Touch. And it's free. I'll be submitting the update using LALT data quite soon. (End promotion.)

I don't own it, but RITI Lunar Map Pro looks like they definitely *should* upgrade to the LALT data for lighting. (Adding: Ironically, the LALT data is actually less sharp than the airbrushed topographic map they seem to be using - I know that because I was using it too. So they may have a problem with that. But even though it's low resolution, the global LALT map we've seen is vastly more accurate.)

Virtual Moon Atlas looks like it doesn't do topopgraphic lighting, so that would be more of a reach.

Celestia can do this, but the UI for doing so isn't as convenient as one would like. The Celestia UI was designed around moving through space, and features for viewing from the Earth's surface have never gotten the attention that they deserve. But, see question #5 here:

http://shatters.net/forum/viewtopic.php?p=118149#p118149

We're in the process of making the Kaguya map the default in Celestia, and a UI overall is coming in the not-too-distant future.

--Chris

(I've got an iPhone and love the Moon and Mars globe apps, btw. Great work!)

Thanks. I'm also a Celestia fan.

mhoward
there will be a hires add on for celestia soon .As soon as i replace my hand made map on the motherload

Exagerated rendering : http://www.youtube.com/watch?v=A3MxCvjQDzk and once processed by vimeo http://vimeo.com/7479346

Also uploading to http://www.dougellison.com/dougstuff/kaguy...render_1080.mov - About 10 meg.

Oooh, exaggerated is right! Looks great. And kind of creepy, too.

My thanks to everyone who made the 16-bit grayscale map possible. I don't have the time right now to play with applying the map to a sphere as I would in Maya or Lightwave, but I have been able to trim out pieces of the map to drop into Terragen, and am very encouraged by the results for use in future artwork. As Bjorn has shown, the data does become somewhat blob-like close up, but Terragen's textures help a bit. This also gets me very excited over the possibilities for using the LOLA data when it becomes available in the coming months.

Rick

Click to view attachment

Joining in the spinning Moon video party... Here's the Moon rendered with a specular highlight to emphasize topography:

http://www.youtube.com/watch?v=agMcK77p3Hc

The overall effect is of looking at a plastic globe.

EDIT: Uploaded the wrong video the first time. As pointed out by FordPrefect, there was a 180 degree offset between the normal and albedo maps.

--Chris

For the past 1 1/2 days i have been trying to get yout???? to except my moon vid ( like the deimos and phobos ) but youtub keeps mangling and f'ing up the ogg file .

http://www.zshare.net/download/68108952dcb3c6a7/

http://www.youtube.com/watch?v=lqxLrig6tQc

Fantastic Work everyone! I've been hoping for such a DEM since 2003. Spectacular, this global Moon DEM rocks!



Chris, it seems your DEM map is shifted by 180 degrees in relation to your albedo texture there.

This must change!
A new version is in progress and I have added normal map lightning last summer even I was not very happy with the map I can build before this data release.
I have downloaded the Kaguya data and I just start to play with them, it look great!
This must be released before the end of the year.

Patrick

Not just a movie party - a lunar software convention!

Here's a comparison that may be interesting. The first screenshot is Moon Globe with its original, lame pre-Kayuga normal map, and the second is Moon Globe 1.1 with the same view, using my version of the LALT normal map. I've just submitted Moon Globe 1.1 to the App Store, so with any luck it should be showing up in two weeks or so.

Has anyone else noticed a slight misalignment between the Kaguya LALT map and existing albedo maps, specifically those from Clementine?

--Chris

Map control from altimeters is much better than the old control - MOLA is the standard for Mars, for instance. In this case the Kaguya control will be far superior to the old Unified Lunar Control Network, which is a mix of control points from Clementine and older missions, basically tens of thousands of points measured on overlapping images. Ultimately we can expect to see older datasets like the Clementine and Lunar Orbiter mosaics, plus the LRO wide angle mosaic, all converted to the altimetry control. LOLA will be the ultimate standard just because of its its high density.

Phil

Thanks for your answer, Phil. It sounds like there's nothing to be done about the misalignment for now, but it's good to understand the source of it.

--Chris

I've had no luck with this yet - how are you using the four coordinates, are you ignoring the time fields, and which instrument sets, specifically are you using.

I may be mis-remembering here, but if memory serves, there have been two different versions of the Clementine albedo map. I think I'm using the later one - it had some new holes in it that I had to patch up. Whatever I'm using, it seems to me to agree quite well with the LALT data.

Again I may be way off here, but the later-model Clementine albedo map might have been re-projected after the Clementine laser altimeter results?

Clem laser altimetry was such low resolution that it doesn't help very much here. It's absorbed into the Unified Lunar Control Network. The point really is that, if the Kaguya team did their work properly, which they presumably did, their result will be better than anything older. If there's a mismatch it's more likely to be with the older dataset.

Doug:

"I've had no luck with this yet - how are you using the four coordinates, are you ignoring the time fields, and which instrument sets, specifically are you using."

First, from the data search page, click product selection - it opens a new window. In it I choose LISM and click 'add all'. The datasets appear in the righthand window.

Second, choose (for example) LISM/MAP/TC-Morning-Map, then click 'determination'.

Back to the first window - now I select data from that dataset. I ignore time, I use coordinates that bracket an area I'm interested in. For instance, Tycho, lat range -40 to -50, long range 340 to 359. Then click 'search execution'.

That takes me to a new page, 'list of search result'. It shows nothing until I again choose the TC morning map line from the product box. But then it lists lots of products, 28 tiles of the big mosaic map. Choose one - click on the file name to see metadata and a thumbnail. Click the order box to order it, and it's put on an FTP server for you. An email tells you how to get it.

I have one - it has to be unzipped. I'm still trying to open it, but too busy to do much yet.

Phil

Thanks for the correction. Still, I believe there was a version 1 of the Clementine albedo map that was pre-Lunar Control Network, even. If you happen to be using that, I imagine it doesn't match up with the LALT data very well at all.

I downloaded the Kaguya LALT_GGT_NUM data set from JAXA and built a digital terrain model with ArcGIS software. I then applied some hypsometric rendering techniques and made these maps. Here's the first map, a global view of the moon built from the surface model and no albedo data.

Another lunar hypsometric map, the near side of the moon

The far side of the moon.

Here's the north pole perspective

And the perspective above the south pole of the moon.

If you are curious about my mapping method, I applied the cartographic tools described at http://blogs.esri.com/Support/blogs/mappin...de-toolbox.aspx

This is a modern implementation of the cartographic relief methods developed by the Swiss cartographer Eduard Imhof in the 1930's. Originally designed for maps of the Alps, this mapping method shows a new "perspective" of the moon.

It's a bit weird looking at the "naked" moon without it's ray systems and albedo differences between the mare and highlands, but it's still pretty recognizable. In term of hypsometric relief, the far side is much more interesting then the near side.

regards, Michael Zeiler, ESRI

I can confirm that newer (or perhaps just 'other') versions of the Clementine albedo map do match up better with the LALT data. Unfortunately, the Clementine map that does align well also suffers from the gaps that you mentioned.

--Chris

i used the "Clem_NIR_V0.1 " 5 band pds data for my map and there is a misalignment
ftp://pdsimage2.wr.usgs.gov/cdroms/clementine/Clem_NIR_V0.1/
texture with the normal
Click to view attachment
without the normal
Click to view attachment

some of this might be from the new topo has 0.0 long in the middle on the pixel and the texture is at the left edge of the pixel
the normal was bumped up to 8192x4096 image
from the LALT_GGT_MAP.IMG header
" and from +0.03125 to +359.96875 degrees in longitude."

A decided to do a rotation movie. The "subspacecraft" latitude is 10°S, the phase angle ~70°, the field of view 25° and the distance from the Moon's center 12000 km. In a way this feels like looking at a new planet/satellite. I have not been interested in doing Moon renders until now because no DEM that I considered good enough was available.

Click to view attachment

Bit of fun turning the elevation map into a coloured gradient, followed by a lighting effect (using the elevation map) - and then polar projecting the whole thing around the south pole. the result is fairly profound!

An italian guy here is looking for somebody able to make a 3d animation of the apollo landing sites:
http://moonscape-project.blogspot.com/2009...-moonscape.html

Is anyone of you able?

---------

I would like to see Kahuya data in WorldWind, but it does not run as a webstart application on my internet-connected PC, so I'd like to download it and install it on another PC: any tips about where to download an offline version?

Beautiful image!

I've downloaded some data from the SELENE website for the MI instrument, I'm trying to open the IMG file but not having much luck. I've tried importing it with ERDAS, opening it with multispec, and also ImageJ. Can anyone give some pointers on getting the PDS3 IMG files to open in some software? I'm still abit new to the non earth data thing. Thanks.

Things I try (usually in this order)

NASAView
ImageJ with PDS Plugin
img2png
a very very large drink followed by ISIS3

i was able to open them using isis3 and the " raw2isis"
once i opened then in ghex to find out how big the header was
pds2isis and vicar2isis failed

funny , ha ha

Just to join the rotation movie party: Moon Globe rotation movie on YouTube.

I finally got my copy of ENVI running. Opening the MI data was easiest with ENVI. Thanks for the tips tho!

I don't know if this is well known or not, but I am a bit surprised to not see this discussed here. The highest point on the Kaguya data, and therefore their idea of "Mt Everest" is on the wall of a crater, which itself is on the wall of a 500km unnamed crater just north of crater Korolev on the far side

alt: 10715
image x,y: 342 1526 (from bottom right of image)
lon,lat: 158.625W 5.375N

The lowest point ("Challenger Deep") is on the floor of 140km diameter Antoniadi crater, also on the far side in the South Polar basin.

alt: -9138
image x,y: 120 314
lon,lat: 172.5W 70.375S

Both of these points are unique, in that there are no other points with the same elevation in the data.

Click to view attachment

This map has all the craters I could find a list for, the highest point marked in black in the red area near the top, and the lowest point is marked in red in the deep purple area near the bottom. If the crater has a name, it's not in my list and I would love to know it.

Edit: Emily already mentioned this in the middle of her blog entry, so no discoveries here. I didn't find any of this until I looked up "Antoniadi" on Google. That "unnamed" crater apparently is provisionally named Dirichlet-Jackson.

Here's a better Moon Globe rotation movie at an oblique angle.

as to the high and low points i have been finding "noise" in the 5760x2880.img
the low point looks like a single pix ( 16bit unsigned "0" in a crater that avg about 300 to 500 ) .I would need to find it again to double check .

i just checked and there is noise but not these two points

nip2 tells me the min is at ( 120x2565 )
max is at ( 342x1353)

min
Click to view attachment
max
Click to view attachment

Kwan,

The definitive source of lunar names is the IAU Planetary Gazetteer, the most useful version of which, for the present purpose, is Jennifer Blue's Digital Nomenclature Atlas. The high point you located is on the rim of a 43-km diameter crater called Engel'gardt in LAC 69 -- actually on the shared rim of Engel'gardt and the larger ("136 km") Engel'gardt B.

You can see a grayscale version of the Kaguya global DEM for this area, with all the current IAU names superimposed using a software called LTVT, here. The misregistration of some of the features relative to where they appear in the Kaguya data is due to inaccuracies in the IAU data, which were originally intended only to be accurate enough to unambiguously identify what feature was meant by each name. I understand the IAU coordinates and diameters may be redone next year using newer and better controlled images.

-- Jim

Really nice work Michael. Could you give a short tutorial on how to import the data and convert it into a DEM? What projections are to be used?

Sean,

You may wish to check today's LPOD.

-- Jim

Carsten,

I'm glad you like the maps. My method applies the following steps with ArcGIS software (www.esri.com):

After downloading the large LALT_GGT_NUM dataset from the JAXA, I used an ArcGIS tool called "ASCII 3D to Feature Class". This tool is designed to import an ASCII file with x,y,z values to create a multipoint feature class, a point data structure designed to scale up to billions of elevation points, such as you encounter with lidar data.

From the multipoint feature class, I created a terrain dataset. A terrain dataset is a generic GIS data structure which is essentially a multi-resolution triangulated irregular network. See http://webhelp.esri.com/arcgisdesktop/9.3/...ataset_concepts

From the terrain dataset, I exported a raster dataset.

From the raster dataset, I applied a geoprocessing model that applies two cartographic effects: hillshaded-relief plus hypsometric tints using a color ramp. This model is described at http://blogs.esri.com/Support/blogs/mappin...de-toolbox.aspx

This technique is a modern implementation of shaded relief pioneered by the Swiss cartographer, Eduard Imhof, in his classic book, Cartographic Relief Presentation, first published in 1965.

In those maps, I use two map projections, Mollweide for the full global view of the Moon, and Orthographic for the four hemispheric views.

Through my work, I have access to the ArcGIS software suite. This is professional grade GIS software that most space enthusiasts could not casually purchase, so I know that my method is not easily adapted by most on this forum. But this was a quick and fun project for me because I am presently writing a textbook on geographic data modeling and this was a good application of terrain visualization methods.

Also, when I examined the DEM built from the LALT_GGT_NUM dataset, I found a large fictitious lunar mountain at selenographic coordinates 16d 11m N, 87d 33m E

I made a slope map from the Kaguya DEM and categorized the peaks with the greatest slopes on the Moon. Some suspicious lunar peaks are at

48 19 S, 141 22 E, 63 degree slope
16 11 N, 87 33 E, 60 degree slope
20 56 N, 60 32 E, 58 degree slope
48 11 S, 8 48 W, 56 degree slope
32 26 S, 58 57 E, 55 degree slope
22 15 N, 81 53 E, 52 degree slope
59 11 S, 46 16 E, 50 degree slope
61 38 N, 8 16 W, 47 degree slope
66 29 N, 70 10 W, 46 degree slope

I don't necessarily assert each one of the peaks are false, but they are suspicious because of their high slope values.

regards, Michael Zeiler ESRI

Your maps are very nice indeed, Michael.

Apparent peaks caused by bad pixels in the original data are found in many raw DEMs and can be edited out if necessary. A good example can be seen in Google Earth's Moon map at 10.8114 N, 54.3234 E on the southern edge of Mare Crisium - pointed out by Joel Raupe. This is in the high resolution DEM on the Apollo 15 Metric camera strip in that map.

Phil

Michael,

Your 3D renderings (as featured on today's LPOD) are indeed very lovely, but are you sure your coordinate readouts are correct?

The Kaguya DEM definitely has some small-scale glitches -- including bad cells, completely missing some craters of ~10 km diameter and representing others as hills -- but I can find nothing particularly strange at the locations you mention. For example, if I am not mistaken, 16d 11m N, 87d 33m E should be sample 1401 in line 1342 of the global DEM. The elevation listed there (at least in the machine-readable, floating point LALT_GGT_MAP.IMG) is -1.490 km (relative to the 1737.4 km reference). None of the 8 nearest neighbor cells differ by more than 100 m (most by much less), which over 1/16th degree (~1.9 km) gives slopes of 3 deg or less.

The "large fictitious lunar mountain" you report should be at the exact center of the two attachments (produced with LTVT), which compare the area (with Neper in the upper left) as viewed by Lunar Orbiter IV (1967 May 11 17:30:07 UT) and as predicted for the same circumstances using the Kaguya DEM. I am unable to see anything particularly anomalous at that the center of these images.

-- Jim