Citizen science to validate NASA remote sensing data

In many places the arrival of winter focuses our attention on snow – its depth, distribution and surface characteristics – so that we can plan for our next ski, snowshoe, snow machine or other recreational adventure. At the same time, environmental scientists have long been interested in tracking snow cover and its impact on ecosystem dynamics, hydrological systems, and glacier health. Knowing how much snow is on the ground at any given time allows scientists to predict trends in water supply for agriculture and hydropower, and provides crucial information for knowing where and when we might expect to see snow-related hazards such as avalanches.

Figure 1: Fiona Hill, Josie Hill, and Tessa Hill use a tape measure and ski pole to measure the snow depth in Mazama, Washington. Photo credit: David Hill

Figure 1: Fiona Hill, Josie Hill, and Tessa Hill use a tape measure and ski pole to measure the snow depth in Mazama, Washington. Photo credit: David Hill

A new NASA-funded project aims to blend the activities of both scientists and recreationalists to broaden our understanding of snow. Using a simple probe or measuring stick, citizen scientists collect snow data anytime, anywhere and submit it along with their GPS location to a smartphone application (Figure 1). The data are then used to by scientists to check the accuracy of other regional snow datasets and model simulations supported by a variety of NASA Earth Observing Programs.

The project, called Community Snow Observations (CSO), already has over 1500 observations collected by citizens across the globe. We partner with Mountain Hub, whose outdoor-oriented, community-fueled smartphone application was modified to include a ‘snow depth’ field allowing users to report on snow conditions. The data are then sent in real time to a web map service designed by a team of software developers at the University of Washington’s eScience Institute. Here users can immediately view their observations on an interactive map which can be used to inform the location of future scientific studies. Citizen scientists also visit our website to learn more about the project and watch tutorial videos explaining best practices in snow data collection.

Commercial cloud technologies have been an ideal platform for our work. Our team relies on Amazon Web Services (AWS) to deploy web applications that can scale to respond to changing user requests at different times of year. For example, our Application Programming Interface (API) uses AWS tools such as Elastic Beanstalk to handle web service deployment and load balancing of computing resources, and S3 for object storage of large gridded datasets. As we continue to develop our API it will enable collaborators to subset any range of citizen science observations and compare these with other related snow datasets. It will also allow our collaborators to pull whatever data they might need to inform their own snow science studies.

Figure 2: Modeled Snow Water Equivalent (SWE) depth over a mountainous region near Valdez, Alaska. The model output on the left is uncorrected, while the output on the right is corrected using citizen science observations.

Figure 2: Modeled Snow Water Equivalent (SWE) depth over a mountainous region near Valdez, Alaska. The model output on the left is uncorrected, while the output on the right is corrected using citizen science observations. (Click to enlarge)

CSO datasets are already proving to be extremely valuable. A team of scientists at Oregon State University has taken a subset of CSO data near Valdez, Alaska, and used these to assess the quality of output from a snow simulation model. Such models take in observations of meteorological conditions such as temperature and precipitation, use these to calculate surface mass and energy exchanges, and output estimates of snow distribution and melt rates. We found that the model significantly over-predicted snow in the test watershed, and that our CSO data could be used to correct the output so that it better represented real conditions on the ground (Figure 2).

Through the summer, our team will be busy analyzing both the data and the lessons learned from the first year of the project. There is no crowd-sourcing without the crowd, and we hope to see more citizen scientists out in the snow next year. Project participants have remarked that making repeated measurements for CSO helped them learn about the importance of snow in their environment. Lend CSO your phone and your time and turn your winter experience into data for science.  

The research team includes:

  • Katreen Wikstrom at the Alaska Division of Geological and Geophysical Surveys
  • Gabriel Wolken at the Alaska Division of Geological and Geophysical Surveys and University of Alaska Fairbanks
  • Ryan Crumley and David Hill at Oregon State University
  • Landung Setiawan (Oceanography), Jonah Joughin (Applied Physics Lab) and Anthony Arendt (Applied Physics Lab, eScience Institute) at the University of Washington