6 ms·

It's quite amazing and simultaneously ridiculous that we have a public 3D model of a distant planet at 5 meters resolution while the best available free and pub

by q87b 3y ago

It's quite amazing and simultaneously ridiculous that we have a public 3D model of a distant planet at 5 meters resolution while the best available free and public models of Earth are at 30 meters.

TomasD 3y ago
You can fly your satellites lower in mars due to atmosphere density, as low as 20km it appears
- dewey 3y ago
  On top of that isn't the lower resolution on earth more for military reasons than a technical one or is this not the case any more?
  - zokier 3y ago
    Hasn't been really the case since french airbus started offering highres global imagery. If you have deep pockets, you can buy 30cm/px (i.e. order of magnitude better than this Mars dataset) images of pretty much anything.
    - generalizations 3y ago
      I wonder if there's any sources for those images that aren't cough restricted. Could be interesting to play with, though maybe not worth paying for.
      - zokier 3y ago
        What do you mean, there are many resources that provide access to various satellite images. In general satellite imaging is not really that sort of hushhush industry it maybe was 50 years ago.
        IntrepidWorm 3y ago
        I read parent as looking for sources that provide access to hi-res satellite imaging, without them needing to provide access to their wallet.
        atonse 3y ago
        I think the thread is referring to satellite pictures of more sensitive locations (like military bases, gov facilities, etc) But I can imagine it all depends on jurisdiction. A Russian or Chinese satellite imagery company will have no problem publishing clear images of US military bases, while an American satellite company may have to comply with US laws prohibiting that. And vice versa. Ultimately if you cobble together imagery from multiple sources, you can probably get everything you want.
- eutectic 3y ago
  And presumably the atmospheric distortion is much less as well.
  - retube 3y ago
    it's surface area also a lot smaller
    - Sharlin 3y ago
      However, the total surface area of Mars is roughly equal to the total land area of Earth.
      - dylan604 3y ago
        If I wanted to make a textured globe, you just made my oceans totally smooth. Some of the tallest mountain ranges we have are in the area you just nulled out.
        Sharlin 3y ago
        And how exactly would satellite imagery (no matter which wavelength range) help you texture the ocean floor in your globe?
        dylan604 3y ago
        nobody said optical wavelengths were the only meaning of imagery.
        swiftcoder 3y ago
        Turns out the surface of the ocean has ridges and valleys that match the seafloor, and these can be measured by radar from a satelite
  - prpl 3y ago
    Atmospheric distortion looking down is not as severe as it is looking up. This is because the distortion subtends much greater angles looking up than looking down, since the distortion is close to the observer. Looking down, it’s also true that, at least in daytime, exposures are very short as well by virtue of the earth being very bright.
- CobrastanJorji 3y ago
  Well yes, but we certainly have better resolution imagery of Earth for many places. It's just that the Mars data is free and the Earth data is owned by corporations that sell access to it for a lot of money.
MichaelZuo 3y ago
Doesn't google earth use 1 meter resolution in certain areas?
- dylan604 3y ago
  Isn't that data from airplanes and not satellites?
  - MichaelZuo 3y ago
    Airplane data would be below 30 cm resolution.
- q87b 3y ago
  "Free to access" in highly limited ToS is not free in a reasonable way when it comes to data.
  - MichaelZuo 3y ago
    Huh? Did you intend to respond to a different comment?
YetAnotherNick 3y ago
WorldView-3 claims to be 31cm resolution: https://worldview3.digitalglobe.com/ https://worldview3.digitalglobe.com/
- tejohnso 3y ago
  Not public. I don't even see a pricing page, so can assume very pricey.
  - qwertox 3y ago
    Also this is appears to be the optical resolution, unrelated to an elevation model.
    - zokier 3y ago
      All major imagining satellites provide stereo imaging suitable for 3d reconstruction. Airbus has whole suite of products around this, and probably the biggest reason why there isn't off-the-shelf products for WV3 is because its just so new. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7248834/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7248834/ https://www.intelligence-airbusds.com/imagery/reference-layers/worlddem/worlddem-thematic-layers-and-derivatives/ https://www.intelligence-airbusds.com/imagery/reference-laye...
  - zokier 3y ago
    Afaik in the ballpark of $500/25km² for archive images.
nickpeterson 3y ago
It’s easier to cleanup someone else’s house.
aannex 3y ago
We actually don't have a 5 meter global elevation for Mars. The imagery is 5 mpp but the elevation data used in the links is MOLA and is closer to 500 meters per pixel (with another merged product at 200 mpp). It is almost more weird/cool that we have hundreds (thousands?) of 1 mpp elevation models for small areas of Mars from HiRISE (available for free from the PDS) and larger swaths at ~20 meters per pixel from CTX imagery used in this global mosaic, but still not close to a global product. various sources: 1. https://astrogeology.usgs.gov/search/details/Mars/GlobalSurveyor/MOLA/Mars_MGS_MOLA_DEM_mosaic_global_463m https://astrogeology.usgs.gov/search/details/Mars/GlobalSurv... 2. https://astrogeology.usgs.gov/search/map/Mars/Topography/HRSC_MOLA_Blend/Mars_HRSC_MOLA_BlendDEM_Global_200mp https://astrogeology.usgs.gov/search/map/Mars/Topography/HRS... 3. https://www.uahirise.org/dtm/ https://www.uahirise.org/dtm/ 4. https://stac.astrogeology.usgs.gov/docs/data/mars/ctxdtms/ https://stac.astrogeology.usgs.gov/docs/data/mars/ctxdtms/

by q87b 3y ago

It's quite amazing and simultaneously ridiculous that we have a public 3D model of a distant planet at 5 meters resolution while the best available free and public models of Earth are at 30 meters.

TomasD 3y ago

You can fly your satellites lower in mars due to atmosphere density, as low as 20km it appears

dewey 3y ago
On top of that isn't the lower resolution on earth more for military reasons than a technical one or is this not the case any more?
- zokier 3y ago
  Hasn't been really the case since french airbus started offering highres global imagery. If you have deep pockets, you can buy 30cm/px (i.e. order of magnitude better than this Mars dataset) images of pretty much anything.
  - generalizations 3y ago
    I wonder if there's any sources for those images that aren't cough restricted. Could be interesting to play with, though maybe not worth paying for.
    - zokier 3y ago
      What do you mean, there are many resources that provide access to various satellite images. In general satellite imaging is not really that sort of hushhush industry it maybe was 50 years ago.
      - IntrepidWorm 3y ago
        I read parent as looking for sources that provide access to hi-res satellite imaging, without them needing to provide access to their wallet.
      - atonse 3y ago
        I think the thread is referring to satellite pictures of more sensitive locations (like military bases, gov facilities, etc) But I can imagine it all depends on jurisdiction. A Russian or Chinese satellite imagery company will have no problem publishing clear images of US military bases, while an American satellite company may have to comply with US laws prohibiting that. And vice versa. Ultimately if you cobble together imagery from multiple sources, you can probably get everything you want.
eutectic 3y ago
And presumably the atmospheric distortion is much less as well.
- retube 3y ago
  it's surface area also a lot smaller
  - Sharlin 3y ago
    However, the total surface area of Mars is roughly equal to the total land area of Earth.
    - dylan604 3y ago
      If I wanted to make a textured globe, you just made my oceans totally smooth. Some of the tallest mountain ranges we have are in the area you just nulled out.
      - Sharlin 3y ago
        And how exactly would satellite imagery (no matter which wavelength range) help you texture the ocean floor in your globe?
        dylan604 3y ago
        nobody said optical wavelengths were the only meaning of imagery.
        swiftcoder 3y ago
        Turns out the surface of the ocean has ridges and valleys that match the seafloor, and these can be measured by radar from a satelite
- prpl 3y ago
  Atmospheric distortion looking down is not as severe as it is looking up. This is because the distortion subtends much greater angles looking up than looking down, since the distortion is close to the observer. Looking down, it’s also true that, at least in daytime, exposures are very short as well by virtue of the earth being very bright.
CobrastanJorji 3y ago
Well yes, but we certainly have better resolution imagery of Earth for many places. It's just that the Mars data is free and the Earth data is owned by corporations that sell access to it for a lot of money.

MichaelZuo 3y ago

Doesn't google earth use 1 meter resolution in certain areas?

dylan604 3y ago
Isn't that data from airplanes and not satellites?
- MichaelZuo 3y ago
  Airplane data would be below 30 cm resolution.
q87b 3y ago
"Free to access" in highly limited ToS is not free in a reasonable way when it comes to data.
- MichaelZuo 3y ago
  Huh? Did you intend to respond to a different comment?

YetAnotherNick 3y ago

WorldView-3 claims to be 31cm resolution: https://worldview3.digitalglobe.com/ https://worldview3.digitalglobe.com/

tejohnso 3y ago
Not public. I don't even see a pricing page, so can assume very pricey.
- qwertox 3y ago
  Also this is appears to be the optical resolution, unrelated to an elevation model.
  - zokier 3y ago
    All major imagining satellites provide stereo imaging suitable for 3d reconstruction. Airbus has whole suite of products around this, and probably the biggest reason why there isn't off-the-shelf products for WV3 is because its just so new. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7248834/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7248834/ https://www.intelligence-airbusds.com/imagery/reference-layers/worlddem/worlddem-thematic-layers-and-derivatives/ https://www.intelligence-airbusds.com/imagery/reference-laye...
- zokier 3y ago
  Afaik in the ballpark of $500/25km² for archive images.

nickpeterson 3y ago

It’s easier to cleanup someone else’s house.

aannex 3y ago

We actually don't have a 5 meter global elevation for Mars. The imagery is 5 mpp but the elevation data used in the links is MOLA and is closer to 500 meters per pixel (with another merged product at 200 mpp). It is almost more weird/cool that we have hundreds (thousands?) of 1 mpp elevation models for small areas of Mars from HiRISE (available for free from the PDS) and larger swaths at ~20 meters per pixel from CTX imagery used in this global mosaic, but still not close to a global product. various sources: 1. https://astrogeology.usgs.gov/search/details/Mars/GlobalSurveyor/MOLA/Mars_MGS_MOLA_DEM_mosaic_global_463m https://astrogeology.usgs.gov/search/details/Mars/GlobalSurv... 2. https://astrogeology.usgs.gov/search/map/Mars/Topography/HRSC_MOLA_Blend/Mars_HRSC_MOLA_BlendDEM_Global_200mp https://astrogeology.usgs.gov/search/map/Mars/Topography/HRS... 3. https://www.uahirise.org/dtm/ https://www.uahirise.org/dtm/ 4. https://stac.astrogeology.usgs.gov/docs/data/mars/ctxdtms/ https://stac.astrogeology.usgs.gov/docs/data/mars/ctxdtms/