Who built Astrospheric?
My name is Daniel Fiordalis and I’ve been an amature astronomer ever since my dad introduced me to the hobby as a kid in Kalamazoo Michigan. In college I studied computer science and physics. I currently work for a software company in the Pacific Northwest during the day and code up Astrospheric in the evenings.
At first, Astrospheric was a small bit of code I was using personally to create dynamic astronomy forecasts for a trip I was planning. But, after getting in touch with Allan Rahill from the CMC I became extremely interested in solving some of the complex problems in generating visuals and searching through large amounts of data.
Can I send in feedback?
Yes, please use the feedback form
to get in touch. I don’t respond to all feedback, but I do read it.
Where does Astrospheric get its data from?
The bulk of the data shown on Astrospheric comes from the Canadian Meteorological Center (CMC).
Other data sources include the GFS model as well as the experimental RAP model which provide values
such as Smoke and Aerosol optical depth. The Seeing model is pulled from Allan Rahill’s model
(which is based off of the CMCs model). I’m constantly investigating other weather models for
interesting variables to help improve the overall forecast on Astrospheric.
How is Astrospheric different from Atilla Danko’s “Clear Sky Charts” service?
I get this question often. Users tend to think that because the primary data source is the same
between Clear Sky Charts and Astrospheric that they give the same results, which isn’t accurate.
Here are a few big differences.
Astrospheric process raw model data from the CMC 4 times a day. This is why Astrospheric updates 4 times a day instead of twice like Clear Sky Chart. Processing the raw data also allows Astrospheric to produce high quality map overlays to visualize the data.
Astrospheric dynamically generates a forecast for your exact location instead of working off a predefined set of locations. Even with thousands of locations available on Clear Sky Chart, that only represents a tiny fraction of the CMCs forecast data. By generating the forecast dynamically, Astrospheric can take advantage of the high resolution CMC data and get you the best results.
Astrospheric is upgrading its astronomy variables to better reflect current conditions. For example, Astrospheric’s Transparency forecast integrates the latest smoke and aerosol data to produce a more realistic forecast of sky transparency. Astrospheric's cloud forecast looks at trends in cloud movement in the area to produce a more accurate forecast.
There are many other functional differences between the services. Clear Sky Chart is obviously
still a valuable resource and some users may prefer it given Astrospheric's higher system and bandwidth
Why doesn't Astrospheric update more frequently?
While there are forecast models that do update very frequently (NOAA's HRRR model for example), they generally lack
the data necessary to produce high quality astronomy variables. Models are constantly improving though and I'm always
investigating new models to include. Until then, Astrospheric will update 4 times a day, which matches the frequency of
CMC's RDPS, NOAAs GFS, and many other models.
Are there plans to make Astrospheric work outside of the continental US and Canada?
Not at this time. Processing worldwide data to build accurate astronomy variables is prohibitively
expensive at this point, nor could I support that many users.
Will Apple's purchase and shut down of the Dark Sky APIs impact Astrospheric?
No. Astrospheric process raw weather model data on its own servers instead of repackaging other company's weather APIs.
However, astronomers using apps such as Clear Outside or the myriad of other weather apps that state
"Powered by Dark Sky" or "Powered by Forecast" may notice changes as those app developers find new sources
for their data.
What other weather data sources are valuable to astronomers?
There are thousands of weather apps and websites available, but it can be tough to know if they will work well
for astronomy. Here are a few tips that can help in picking them out.
Go to the source of the data and know what weather model you’re looking at.
Check more than one weather model.
Look at the data on a map since it’s still the best way to understand the weather patterns.
Weather.gov is a great resource for those in the USA since it gets you access to the primary US weather models
(GFS, NAM, HRRR, etc). If you’re feeling adventurous, you can access raw model data at https://nomads.ncep.noaa.gov/
Should I send feedback when the weather report is wrong?
Short answer - no. Every 6 hours the Candian RDPS model generate well over 200 million predictions for an area
covering most of North America. It will be wrong at times. The model is constantly ingesting new data from thousands of
automated sources in order to correct and generate the next set of predictions. Observations from individuals are not
incorporated into the model and therefore would be a waste of time to send. However, if you notice something completely out of the
ordinary (eg. a dew point of absolute zero), then it's worth letting me know, because an error may have occured in my code
which parses the data.
Can rain and other precipitation be added to the forecast?
Technically yes, but precipitation data won't be added to Astrospheric.
Why doesn’t Astrospheric work on my (insert Android device)
I like Android and use it on my daily phone. Technically I’ve targeted the build to work on Android 5.1 and higher,
which means it can run on tens of thousands of different devices. Unfortunatly it is not possible for me to test all of these configurations so I've placed a few blocks on certain devices
to ensure a better user experience.
- The device must be on Android 5.1 or higher.
- The device must have at least 1GB of RAM.
The device must be running a Google approved version of Android which passes attestation tests.
Devices that have had the base Android OS modified (rooted, custom ROM, etc) will likely not pass. If you have a device meeting the above requirements but Google Play reports the app isn't
compatable with your device then it's likely an attestation test failure.
Without these basic tests in place, I found I was spending well over 90% of my development time fielding Android issues. This cuts into my development time for other aspects of the service (iOS, browser, backend services, etc).
If you find an issue please contact me
If you’re on a Google supported version of Android (8.0 and higher) as well as a device from Google or
Samsung, then there is a good chance I can get the issue fixed.
Why is the time shown in Astrospheric wrong?
Astrospheric shows the local time for the current forecast location. This includes any adjustments for timezone as well as
daylight savings time. If you notice that the time is an hour off, please first check that your device's operating system timezone is set correctly.
If it is set correctly, then please contact me
Can I embed Astrospheric on my website?
Most likely. Check out www.astrospheric.com/embed
for more information and the rules for using it.
How can I support Astrospheric?
Financially - The best way to support Astrospheric with money is to purchase a subscription. Not only will you get weather alerts, but the money helps fund current and upcoming improvements to the service.
In the community - Tell your friends about Astrospheric! If you’re part of an astronomical society then consider using Astrospheric and Subspace!
Scientifically - If you are a meteorologist or astronomer (or in a related scientific field), and you have an idea to improve or expand the data Astrospheric is using then please get in touch with me!
Is there a French version of Astrospheric?
Not yet, but I’m working on it! My goal is to have a French version by the end of 2021.