What is the Snowtweets project?

The Snowtweets Project provides a way for people interested in snow measurements to quickly broadcast their own snow depth measurements to the web. These data are then picked up by our database and mapped in near real time. We are especially interested in using web-based digital technologies to map snow data; currently, the project uses the micro-blogging site Twitter as its data broadcasting scheme.

To view the snow depth measurements (or Tweets), we have developed a data visualization tool called Snowbird that lets you explore the reported snow depths around the globe. The viewer shows where the reports are located and how much snow there is at each reported site.

The Snowtweets Project is in early stages of development and we plan to update and improve it as we go along. We rely on user participation to measure snow depth (including zero snow depth) and then send the measurements in.

Please note that this project does not seek to replicate other existing community-based projects. Rather we aim to attract as many new contributors as possible through widely used web-media, such as Twitter, who might not be aware of these other projects.

Want to contribute?

If you do not have one, you will need to register for a Twitter account at www.twitter.com. It is free and easy to do.
Measure the snow depth where you live, work or play. Ideally, you can easily do this with a simple ruler by poking it into the snow straight down (vertically) to the ground at a place you think is representative of the snow in your area (maybe in the middle of your back yard or at an undisturbed/undrafted place on the way to work, or in your school yard). There is a section below that explains in more detail some of the best ways to measure snow depth.
Then using your Twitter account, tweet this information along with your postcode/zip code/lat-lon location using the #snowtweets protocol:

#snowtweets <snow depth in cm., in. or ft.> at <postal code, ZIP code or latitude, longitude>

If you want to use your postcode or address, it should look something like this:

#snowtweets 50.0 cm. at K1A 0A2

#snowtweets 10.0 in. at 20500

#snowtweets 4 cm at Palmerston North 4414

Outside North America, it might help to specify your country with your postal code:

#snowtweets 2 cm at 102-8166 Japan

If you want to use absolute latitude/longitude coordinates, use latitude (dd.ddd) and longitude (ddd.ddd) in decimal degrees. Note, latitude should be -90.000 to +90.000 and longitude -180.000 to +180.000. For example 8.3 cm at 41 30'N: 120 45'W should be:

#snowtweets 8.3 cm at 41.500, -120.750

Not sure what to include for your location? Try your location in Yahoo Maps. If it works with Yahoo, it works with Snowbird.
Give our system a few minutes to process the data and you should see your Tweet show up in Snowbird. Your input will assist snow-related research here at the University of Waterloo.
You can tweet your snow depth from wherever you are and as many times as you want to each day:

- at home. Choose the same place each time – makes it a more reliable record. (we suggest a maximum of 5 times a day in a heavy snow storm)

or

- on the road with your wireless Blackberry/iPhone/smartphone. If you know your postcode/zip/lat-lon just enter it. We will do the rest. If you are continually travelling, enter the snow depth for as many new locations as you like.

Measuring snow depth: many ways but consistency and representativity is best

Measuring snow depth, on the face of it, seems easy. And indeed it can be. However, for really reliable measurements, there are a few things to consider:

try and locate an undisturbed patch of snow. If the snow has been disturbed (kids playing in the backyard etc) then it might not be the best place. Some observers like to set aside a small patch of ground to make their measurements where they are confident that the snow has not been disturbed;
snow depth can vary significantly over short distances as the snow gets blown around. Try not to choose the 'deepest' snow you can find but aim to measure the most representative snow you can find. You may need to take 10-20 measurements to get a sense of how it varies;
ideally, you should measure snow depth away from trees and buildings which can create windowflow patterns that artificially scour and drift snow in patterns that do not always reflect the 'general' snow patterns;
if your snow depth measurement is less than 5 cm, it was probably not easy to measure. Record it as best you can but it will probably show up as a trace on our server.

There are some excellent sources on how to measure snow. These sources range from the World Meteorological Organization protocols to community-based projects designed to get teachers and kids interested in measuring weather variables. For further information please follow the links below:

GLOBE Canada (www.globecanada.ca)
CoCoRaHS USA (www.cocorahs.org)
World Meteorological Organization precipitation protocols (WMO Guidelines)
Meteorological Assimilation Data Ingest System US (MADIS snow protocols)

The Snowtweets project aims to get as many people involved as possible and if you are not able to make a detailed measurement that is fine. We would still like you to Tweet your snow.

What will we do with the data?

We currently have several approaches to estimating snow depth or SWE at various spatial scales and with different accuracies. These include satellite remote sensing estimates and numerical simulation models. Testing these approaches needs data! This is where the Snowtweets data will come in handy!

We are developing quality control steps that allow us to determine how good the Snowtweets are. This does not mean that the data are wrong, necessarily, but we need to ensure that there is some consistency with: a) other Tweets in the area, or b) if there are no other Tweets close by, whether the temporal consistency of Tweeting is good. We will let you know how this goes as we move into the (northern hemisphere) winter season.

Snowbird: the Snowtweets viewer

A public preview of Snowbird is available for download. Visualizations of snowtweets are also available in KML (view in Google Maps) and GeoRSS (view in Google Maps) formats.

Implementation

Snowbird is being developed using Adobe Flex and Papervision3D. This platform lets us develop Snowbird as a Flash kiosk demo and an interactive Adobe AIR application using a common codebase. The Flash applet is completely non-interactive, flying through of the newest tweets automatically. The AIR application lets you browse new tweets through autopilot or you can fly to any tweet you pick. The AIR version of Snowbird also supports dynamic texture replacement, which you can use to swap out images using the combo box at the top left, just above the tweets. We'll be using this feature to visualize some interesting data in the future.

Adobe AIR is a runtime environment that works across different platforms (Mac OSX/Windows/Linux). The Adobe AIR runtime is required to run the application and will be installed automatically provided Flash 9 (or newer) is installed. Alternatively, it can be obtained for free from Adobe and installed manually.

Using the AIR application

On startup, Snowbird automatically begins to download newest tweets. Once some tweets are loaded, you can begin to browse through them. Let the autopilot take its course by waiting a few seconds or take the helm and select a tweet to fly there.

Moving up the list of tweets, the automatic fly-through pans between user posts in chronological order. To switch to manual, just select a tweet on the left pane or click a marker on the globe. Any user interaction interrupts the automatic fly-through, but it can be resumed by clicking the play button at the top-left.

Click once to fly to a tweet. Click again to fly there faster.

To resize the window, drag it by its gripper at the bottom-right.

Install the full-featured AIR application:

Download Snowbird: Snow Depth Twitter Explorer now.

This application requires Adobe® AIR™ to be installed for Mac OS or Windows.

Using the Flash demo

We have also developed a Flash demo version for you to try out without having to install the AIR application. The Flash demo is non-interactive. Just watch and enjoy...

Launch Flash demo (click outside the applet to close)

Acknowledgements

Snowbird uses Earth's City Lights, Blue Marble and Blue Marble NG images from NASA Visible Earth. GLOBE images were obtained from the National Geophysical Data Center (NGDC). The Natural Earth image was created by Tom Patterson for the US National Park Service.

The idea came from a UK snowfall mapping project developed by Ben Marsh.

Contact Us

For questions regarding the research applicability of the Snowtweets project, contact Prof. Richard Kelly (rejkelly @ uwaterloo.ca).

For developer/implementation questions, contact Raymond Cabrera at (arcabrer @ uwaterloo.ca) or Joshua King (jmking @ uwaterloo.ca).

Privacy

Please note that information posted on Twitter becomes publicly available through both Twitter and our own visualization media. While we do not collect snowtweet information directly, our system does periodically collect data publicly available through the Twitter API. As a result, your data—including (but not exclusive to) user name, location, time, and snow depth—may be stored in our database. The data is internally used solely for research purposes, but is also made available publicly (via Snowbird, KML and GeoRSS).

In addition, please see the University of Waterloo's Privacy Policy for more information, including our usage of Google Analytics tracking.

receiving and visualizing community-measured snow data from around the planet

Snowtweets

Snowbird application

Visualization