Kerbin Geographic And Science Society

[Rich]

Currently we are doing a mixture of placing names on a map:

http://kerbalspaceprogram.com/forum/index.php?topic=11388.msg188134#msg188134

and generating a KSP persistence map with ships named after the places:

[togfox]

I investigated gis software some time ago and couldnt find software that was free and easy to use. Your map looks pretty sweet. I intend on using quikgrid for this purpase but if you wish to own the gis part then be my guest. When you have a plan ill include links to dropbox or wherever in my OP. i suggest you maintain a csv with

Lat/long/alt/name

As well as the map.jpg. This will allow others to import map csv and name csv together, sperately, or use their own client fo their own purposes.

This does not diminish rich\'s work as a pesistence file allows those not confident with maps to fly direct to many locations for fun and challenge.

[8556732]

@Rich

Seems I\'m late to the party! Ha, well without encroaching on what you currently have, I could offer GIS if you wish.

Effectively you are building one into the game, by adding the place names physically (an idea which I love, as you can easily mechjeb yourself to a feature!). However, if you want alternatives I can provide these.

I\'m interested to know what method of gridding/interpolation quikgrid uses. I\'ve used nat-neighbour for mine - it doesn\'t really matter much as the spatial resolution is so low, its just a query as I have never seen that program before!

@Togfox

Yeah they have a pretty steep learning curve, but are incredibly useful, and generally massive time savers.

I would keep the data in a geodatabase, which I could export to CSV in any header format you wish.

To be honest, discovering and naming these places isn\'t really my cup of tea. I just saw the remote sensing that you guys have managed to achieve, and figured I might be able to help/offer skills. However, you seem to already have done/are doing much of it!

The elevation maps seem to provide targets for further exploration, and I can offer terrain analysis of any that are produced. This could be slope angle maps (maybe useful for identifying possible landing/base sites), Aspect Ratio Maps (for quantitatively defining mountainous regions), Volumetrics, Potential Hydrology.... etc.

In the case of defining mountainous regions, it would be a simple process to define what classifies as a mountain based upon a global relative aspect ratio, and then auto-generate a map of the mountainous areas. This could then be explored in kerpson (kerbal person lol!) to confirm if the areas are indeed what the computer says.

Just an idea for ya, and as I say its a quick 5-10 min job for me. I simply set up a workflow with the parameters we decide and bam its done.

I had a really crazy/stupid idea of running a hydrological analysis on Kerbin, and producing a prediction map of rivers. This would serve as an actual working map! Just an idea, being as we currently don\'t have rivers on the surface!

[togfox]

One thing I\'m finding challenging is defining a region. Peeps here are defining mountain ranges or a whole crater or a small island. A GIS can track area\'s/volume reasonably quickly. Quikgrid is a single point in space and doesn\'t do regions at all. If you\'re keen to own the map, points and regions then that would help - division of labour.

As for your other idea\'s on slope angle maps and aspect ratio maps, knock yourself out. I suggest you find a part that interests you and simply offer it up to the community. I can see you have much to offer in terms of training/education so just post up stuff and see if ppl go wow. I\'ve posted in this thread my favourite spot - The North Pit

http://kerbalspaceprogram.com/forum/index.php?topic=11388.msg182431#msg182431

and the terrain is challenging. Landing is hazardous despite the public and my private maps. Perhaps you can offer an analysis so I can find a good location to set up a mining colony either inside or just outside this location as an example. We can all then see the power of a GIS and drop our jaw in awe of even more real science.

[8556732]

I\'ll get on it tomorrow morning for ya as I need sleep ???. The surveys we currently have might not be high enough resolution - the roughness shown in the screenshot will probably be smoothed by any gridding action.

But I will give you a gridded map of the area you have indicated that you can use to your hearts content. What colour ramp would you like your terrain in? Greyscale? Psychedelic? Rainbow?

I\'ll give it a stab for you, I can predict that the large depressions and slopes will show up, but you will lose the finer lumps and bumps. This is pretty common, even with high resolution real satellite data.

On a side note, this is quite funny as I know a guy currently doing this sort of analysis off coast of Norway... they\'re deciding how to divide the borders of the Arctic up!

[togfox]

The surveys we currently have might not be high enough resolution

We do have the power to make this finer (for considerably larger effort!). This society is pretty resourceful. See how you go. I can offer my personal high res maps if needed (but I\'d have to find them!).

[8556732]

We do have the power to make this finer (for considerably larger effort!). This society is pretty resourceful. See how you go. I can offer my personal high res maps if needed (but I\'d have to find them!).

Don\'t worry about that, the effort seems pretty immense. Is the survey tool only limited to planetary wide swaths? Or can the data acquisition be limited to a certain region? What I\'m basically saying is can you do a high res survey of difficult regions such as yours, or do you need it to be planet-wide.

I should probably do one myself...

[togfox]

At the risk of boring others, the first image is the public map of the North Pit.

The second image is augmented with my own data collected from my personal team of surveyors. The difference is quite noticable. I\'ve attached the augmented csv file of the North Pit.

http://dl.dropbox.com/u/80858415/Northpitdatapoints.zip

[8556732]

Hey Togfox,

Can you confirm this image as your flight path over the North pit?

https://www.dropbox.com/s/udmp51h6j474y74/Screen1.png

If that\'s the case, then I have used kriging as a method of interpolation, as there are large gaps in the data. This results in a grid that differs to yours somewhat (I believe QuikGRID uses a linear interpolation, kriging is much more robust for data that isn\'t regularly sampled):

https://www.dropbox.com/s/xpnvxvkqm2xola3/tod1-DEM.jpg

Contour Map:

https://www.dropbox.com/s/q78s0oxewu5pfl7/tod1-cont.jpg

You had a number of varying altitudes recorded for some points, I have used the mean of these when gridding. I have not smoothed the DEM to remove the errors; I recommend you only use the models from within the outer wall of the pit. As you can see, the errors are typically outside your flight paths, and hence represent nodata.

I have some further analysis for you:

Slope Angle - I\'ve had to do some maths to work out the conversion factor for Munar decimal degrees to metres, if anyone else wants details on this I can illustrate them in another post. Here\'s the results though, classifications are rough (lots of useless decimal places) as I don\'t have time this afternoon to tidy the histograms up:

https://www.dropbox.com/s/qbd7z3dk5yq578t/tod1-slope.jpg

And with contours:

https://www.dropbox.com/s/ftux3mpboj2eqce/tod1-slope2.jpg

Slope Direction

https://www.dropbox.com/s/6owb8dn3e3oa4kq/tod1-aspect.jpg

And finally, where I recommend for landing/exploration (in red)

https://www.dropbox.com/s/ivx8oycvni1qxqy/tod1-lessdeg.jpg

These areas appear (at least from your survey!) to be worthy of further survey/exploration.

How\'s that for you?

I\'ve just realised the large size of the maps (oops!). This is due to my default output of 600Dpi for large poster work (I will change this for future)

[togfox]

Bravo! Very nice and professional. Now I\'m jealous. That top image is indeed my \'flight\' path ... though that is done with a crawler on the ground. Tis how I can get the extra resolution not possible on orbital surveys.

I believe I have been trumped and can no longer claim \'master cartographer\'!

[8556732]

So that survey is in-effect the same as a guy walking round with a gps and reading out data? Nice!

As I say... I have no interest in data acquisition, just the display and interpretation of it! You can keep your title for this reason

If we compile a CSV or even just an excel spreadsheet of named locations we can produce some nice looking global maps. The low resolution data should be sufficient for this

I\'m not saying yours aren\'t nice looking... but i get to play with the expensive software

What we do need though is an agreement on the relative geoid\'s of these bodies. if we want proper maps (which I\'m guessing you guys do) then someone should do the maths related to them.

I\'ve already done some sums, and I get:

Munar Degree = 3481m @equator

Minmusar Degree = 1047m @equator

Kerbal Degree = 10,444m @equator

I\'ve used these to work out the relative z-scaling factors for each planet, so that maps can be properly projected. This is using the games in-built Lat/Long, so in effect it is re-discovering the wheel

I imagine standard world-based projections will suffice, with the appropriate z-conversion

If anyone can get that data from config files it could save time! (if it\'s not in the spirit of RP, I apologise!)

[Corax]

Corax - if you zip that csv does it become drop-box able? If you cut it down to three columns and de-dupe it then there is a chance it will zip nicely.

Finally, here it is. 97M data points, cleaned up, sorted, rounded to five decimals, 2.7GB of Kerbin CSV data packed into an archive just shy of one gigabyte (note: comma separated values, not semicolon as in ISA RAM). I reply here because it\'s neither ISA_RAM format, nor was it collected using ISA_RAM.

I\'ll leave it up for a while, but will ultimately remove it again simply because of its size

[8556732]

@Corax hi

97M points is huge, even for a program such as quick grid, the interpolation of such is going to be pretty damn long.

I\'m downloading it now, and will leave it to grid. I don\'t think I\'ve ever interpolated so many points, this is pretty damn impressive stuff!

[Corax]

@Corax hi

97M points is huge, even for a program such as quick grid, the interpolation of such is going to be pretty damn long.

I\'m downloading it now, and will leave it to grid. I don\'t think I\'ve ever interpolated so many points, this is pretty damn impressive stuff!

I\'m using gnuplot to generate simple rectangular projection maps, and although it takes a while, it handles even larger files with ease. The raw data file is some 14GB, but contains a lot of information that\'s useless for mapping (pressure, temperature, time etc).

I\'m really curious to see what you can make of it

[togfox]

*****************************************

Press Release

*****************************************

The Kerbin Keographic Society is now called the Kerbin Geographic & Science Society (KGSS) and will expand it\'s scope to include all things scientific. If you haven\'t read the OP yet, all the way back on page 1, I invite you to read the new direction for the KGSS.

After reading the OP, there are two new challenges in the challenges forum (http://kerbalspaceprogram.com/forum/index.php?board=23.0) where you can help solve real scientific challenges and puzzles and expand your knowledge of the Kerbal universe. Maps and feature naming are still an important part of the KGSS, but we\'re now including scientific challenges for the whole community to partake in. Some of the missions are easy, some are hard, and for some - the answers are already known (to some). Please do not spoil the fun for others on the road to discovery.

I hope to introduce news articles as well so if you\'d like a fact-based (no RP thanks) article of yours to be published under the KGSS banner then let me know! 8)

[simplemunrockets]

Here is the mountain range I found:

I would like to propose a name for the range, and for what appears to be the tallest mountain. (There was another peak further away, but there was a very narrow ridge leading to it, and every time I tried to go across it my Kerbals fell off of it. )

Here are the pics:

Coords are 0|59|57 N, 79|11|52 W.

Mountain Name: Mt. Kurt (in honor of kurtjmac)

Range: HarvesteR Mountains

[VincentMcConnell]

When every continent has been named on Kerbin, can you release a map with every continent labeled?

[HOC]

Seconded.

[Radion]

I\'d like to request funding from the KGSS to begin the study of Kerbin\'s biosphere. As part of the study, I we will send 4 units out to each of the separate biological regions on Kerbin. One, to the middle of the ocean. Another to the poles. Another into the deserts, and one into the dense vegetation. There, each base will study the flora and fauna present, and report back with their findings.

[togfox]

Maps with named features.

[HOC]

Does each continent have a name yet? I kinda think naming them is a priority, especially when it\'s such a simple thing to do. After getting around 5 to 10 suggestions, set up a poll on this thread, and all \'members\' have, say, a week to vote on the name of a continent.

(You might also want to include a \'none of the above\' choice. If that wins, then we\'ll have to replace the least chosen suggestions and go again.)

Just a humble idea, coming from a kid who really wants those names dammit!

[mincespy]

Does each continent have a name yet? I kinda think naming them is a priority, especially when it\'s such a simple thing to do. After getting around 5 to 10 suggestions, set up a poll on this thread, and all \'members\' have, say, a week to vote on the name of a continent.

(You might also want to include a \'none of the above\' choice. If that wins, then we\'ll have to replace the least chosen suggestions and go again.)

Just a humble idea, coming from a kid who really wants those names dammit!

The RP board named pretty much everything, shame it isn\'t around anymore.

[Rich]

The fundamental difference between this project and the RP naming scheme, is that names for major locations are determined by poll, rather than on a first come first serve basis. That said, names originating from the RP board may be submitted, but will most likely undergo the poll process.

[HOC]

Exactly!

The fundamental difference between this project and the RP naming scheme, is that names for major locations are determined by poll, rather than on a first come first serve basis. That said, names originating from the RP board may be submitted, but will most likely undergo the poll process.

So... um. Start the poll? Anyone?

[Rich]

We will run the polls on a per region/feature basis when enough names have been submitted.