6 ms·

The nature of water: unveiling the most detailed view of water on Earth

74 points by DiabloD3 10y ago

daleroberts 10y ago
Don't you love those days when you want to cry as someone copies what you've done on a larger data set and gets it into Nature. http://www.sciencedirect.com/science/article/pii/S0034425715301929 http://www.sciencedirect.com/science/article/pii/S0034425715... http://www.ga.gov.au/interactive-maps/#/theme/water/map/wofs http://www.ga.gov.au/interactive-maps/#/theme/water/map/wofs
- shmageggy 10y ago
  I'm guessing you're the third author on that paper? How does what Google did differ from what you did, apart from scale? At least they cited you, right... ¯\_(ツ)_/¯
  - daleroberts 10y ago
    Yeah, at least we got a quick cite. I believe they had to sacrifice a bit on the rigour of the methodology to be able to do it at scale. For example, atmospheric correction of the data, methodology, validation, ... We've also improved a lot on our methodology since our own paper was published (using more sophisticated classification algorithms, for example). Not too fussed, I've got a lot more cool stuff in the pipeline...
- bahmboo 10y ago
  Frustrating, but your work still counts and matters.
  - daleroberts 10y ago
    Thanks. I just do the work because it's fun, that's the rewarding part for me. I just like to wander around working on random "mathy" pure and applied problems in a variety of areas...
omegaworks 10y ago
I wonder if anyone has done the same kind of analysis on groundwater / well-water reservoirs. I wish there was a simple way to combine different sources of map data to produce new visualizations. This is the 21st century, maps should be as easy to make as memes! Are there standard formats? Where would one even begin?
- Cthulhu_ 10y ago
  How would they? That's the kind of thing that does not show up on satellite images, and would need to be collected / collated through manual measurements. I'm sure the data is there, at least in some parts of the world, but it's not as accessible as satellite images, since it requires access to local / country-specific data.
yosito 10y ago
Very interesting data set! I'll have to download it when I get back to my computer and take a look. One thing that bothers me is that the language used seems to be somewhat alarmist about loss of water, when the real problem is much more complex and the surface area of water is actually growing. Maybe it's a nitpick but I think alarmist language can tend to prevent people from taking an honest look at important data. Side note, I've been looking for a map of the world's fresh water that I can plug into Google Earth. This may contain that data, I'll have to check, but is anyone aware of related datasets that I could use to find things like bodies of fresh water in deciduous forests, for example?

74 points by DiabloD3 10y ago

Don't you love those days when you want to cry as someone copies what you've done on a larger data set and gets it into Nature. http://www.sciencedirect.com/science/article/pii/S0034425715301929 http://www.sciencedirect.com/science/article/pii/S0034425715... http://www.ga.gov.au/interactive-maps/#/theme/water/map/wofs http://www.ga.gov.au/interactive-maps/#/theme/water/map/wofs

shmageggy 10y ago
I'm guessing you're the third author on that paper? How does what Google did differ from what you did, apart from scale? At least they cited you, right... ¯\_(ツ)_/¯
- daleroberts 10y ago
  Yeah, at least we got a quick cite. I believe they had to sacrifice a bit on the rigour of the methodology to be able to do it at scale. For example, atmospheric correction of the data, methodology, validation, ... We've also improved a lot on our methodology since our own paper was published (using more sophisticated classification algorithms, for example). Not too fussed, I've got a lot more cool stuff in the pipeline...
bahmboo 10y ago
Frustrating, but your work still counts and matters.
- daleroberts 10y ago
  Thanks. I just do the work because it's fun, that's the rewarding part for me. I just like to wander around working on random "mathy" pure and applied problems in a variety of areas...

omegaworks 10y ago

I wonder if anyone has done the same kind of analysis on groundwater / well-water reservoirs. I wish there was a simple way to combine different sources of map data to produce new visualizations. This is the 21st century, maps should be as easy to make as memes! Are there standard formats? Where would one even begin?

Cthulhu_ 10y ago
How would they? That's the kind of thing that does not show up on satellite images, and would need to be collected / collated through manual measurements. I'm sure the data is there, at least in some parts of the world, but it's not as accessible as satellite images, since it requires access to local / country-specific data.

yosito 10y ago

Very interesting data set! I'll have to download it when I get back to my computer and take a look. One thing that bothers me is that the language used seems to be somewhat alarmist about loss of water, when the real problem is much more complex and the surface area of water is actually growing. Maybe it's a nitpick but I think alarmist language can tend to prevent people from taking an honest look at important data. Side note, I've been looking for a map of the world's fresh water that I can plug into Google Earth. This may contain that data, I'll have to check, but is anyone aware of related datasets that I could use to find things like bodies of fresh water in deciduous forests, for example?