3 ms·

This is an interesting use of the barometer/GPS hardware but I guess I'd appreciate a little help on understanding its greater utility? Atmospheric pressure do

This is an interesting use of the barometer/GPS hardware but I guess I'd appreciate a little help on understanding its greater utility? Atmospheric pressure doesn't change that much. It's pretty much the same where I am as it is 20 miles away at my nearest airport. What is the need I'm not seeing for a sensor net of higher granularity than already exists? Is it for tracking weather related to very intense pressure gradients like tornadoes? I thought we had radar that did a pretty good job of that. Also isn't the accuracy of the barometer really, really poor when it is in motion? I thought they just used it to get a rough sense of altitude and speed when GPS was too costly/unavailable? Cool work.

cryptoz 14y ago
Weather prediction is pretty miserable right now. If you turn on the TV or go to a website to get a weather forecast, you will be told that "it might rain later, somewhere near you, maybe". I think that's unfortunate. For many people, that's much worse than unfortunate, it's terrible. An extreme example would be farmers in places that are being affected by changing weather patterns due to climate change. Knowing with high accuracy on a local scale when it will rain could be very very useful. I'll address some of your points. > Atmospheric pressure doesn't change that much. It's pretty much the same where I am as it is 20 miles away at my nearest airport. Well, yes. The changes are subtle. But that just means I have a tough job ahead of me. Small changes != useless data. Also, is the weather the same at the airport 20 miles away as it as your house right now? It might be. Is it always the same? Most definitely not. > Also isn't the accuracy of the barometer really, really poor when it is in motion? The accuracy is very high. The barometers are very sensitive, so sensitive that if you move your phone from your feet to your head you will see the change in pressure due to the change in altitude. This adds noise to the data, lots of noise in fact, but it's something that we are working to overcome. In building the visualization tool, we've found that at larger scales the network makes it obvious which changes are weather-related and which are noise. At smaller scales, we must do some work to filter it out. > I thought they just used it to get a rough sense of altitude and speed when GPS was too costly/unavailable? That's why they are included in the devices, yes. But I think there are many greater uses for them, such as building a global network of barometers. Did you click through to the actual visualization tool, at http://pndv.cumulonimbus.ca http://pndv.cumulonimbus.ca? You will see that we have collected some really incredible data that nobody else seems to have. -- The main point here is that this is an experiment. I think it is very likely that having higher resolution data will increase the accuracy of small scale, short-term, local weather prediction. I'm working some some atmospheric scientists (Cliff Mass, for one) who are very excited about the project. We'll see where it goes!
- shawn-butler 14y ago
  Yes, that's what I meant by poor. The sensor noise makes it a difficult instrument when in motion. Yes I looked at the visualization. Very cool. I wasn't trying to be negative just wondering about how it differs from data sets currently available and what kind of value I could add on top of it. Thanks for info, it is kind of exciting. What lept into my head was some way of getting a measure of wind speed in addition to pressure but I don't see how that would work with the hardware we have.

cryptoz 14y ago

Weather prediction is pretty miserable right now. If you turn on the TV or go to a website to get a weather forecast, you will be told that "it might rain later, somewhere near you, maybe". I think that's unfortunate. For many people, that's much worse than unfortunate, it's terrible. An extreme example would be farmers in places that are being affected by changing weather patterns due to climate change. Knowing with high accuracy on a local scale when it will rain could be very very useful. I'll address some of your points. > Atmospheric pressure doesn't change that much. It's pretty much the same where I am as it is 20 miles away at my nearest airport. Well, yes. The changes are subtle. But that just means I have a tough job ahead of me. Small changes != useless data. Also, is the weather the same at the airport 20 miles away as it as your house right now? It might be. Is it always the same? Most definitely not. > Also isn't the accuracy of the barometer really, really poor when it is in motion? The accuracy is very high. The barometers are very sensitive, so sensitive that if you move your phone from your feet to your head you will see the change in pressure due to the change in altitude. This adds noise to the data, lots of noise in fact, but it's something that we are working to overcome. In building the visualization tool, we've found that at larger scales the network makes it obvious which changes are weather-related and which are noise. At smaller scales, we must do some work to filter it out. > I thought they just used it to get a rough sense of altitude and speed when GPS was too costly/unavailable? That's why they are included in the devices, yes. But I think there are many greater uses for them, such as building a global network of barometers. Did you click through to the actual visualization tool, at http://pndv.cumulonimbus.ca http://pndv.cumulonimbus.ca? You will see that we have collected some really incredible data that nobody else seems to have. -- The main point here is that this is an experiment. I think it is very likely that having higher resolution data will increase the accuracy of small scale, short-term, local weather prediction. I'm working some some atmospheric scientists (Cliff Mass, for one) who are very excited about the project. We'll see where it goes!

shawn-butler 14y ago
Yes, that's what I meant by poor. The sensor noise makes it a difficult instrument when in motion. Yes I looked at the visualization. Very cool. I wasn't trying to be negative just wondering about how it differs from data sets currently available and what kind of value I could add on top of it. Thanks for info, it is kind of exciting. What lept into my head was some way of getting a measure of wind speed in addition to pressure but I don't see how that would work with the hardware we have.