We were inspired by the idea of reducing carbon emissions. However we thought that a significant problem with motivating people to do so is that it is an extremely abstract goal with no obvious short-term effects. Moreover, climate models are generally complex and their predictions are usually locked up in scientific papers and obscure datasets. We hoped that by democratising the process and enabling people to interactively see the effects of different scenarios on the UK weather - what is closest and dearest to them, might be motivating and ultimately help the UK to reduce its carbon-footprint.
What it does
It uses UK Climate Projections data to provide an interactive map of the UK with the projected sea level rises and temperature projections overlain. It allows people to investigate themselves the projected effect of climate change upon the UK under three different scenarios of greenhouse gas emissions. Hopefully this will make the normally complex climate models and the assumptions they depend on more accessible to the general public and ultimately, by doing so, hope to spur some action on climate change when people can directly and interactivelt see the effect of reducing greenhouse gas emissions on the climate of the UK.
How we built it
We processed several large datasets from the UK Met-Office containing projected sea-level and temperature rises in the period 1990-2100. We preprocessed to remove missing data and other abnormalities and then parsed it into JSON to send to the front-end. On the front-end we process the data into the GeoJSON format and use it to create an overlay data-layer in Google Maps. We also can swap out the data dynamically to allow the user to observe the climactic effects under various future emission scenarios and choose different climate variables to study.
Challenges we ran into
We had significant problems with the obscure and strange data format that the UK Climate Projections office used for their datasets. A significant amount of work went into understanding and parsing these formats into a more usable form. Additionally, much of the data was inconsistent with other datasets for instance in measurement frequency, and we performed smoothing and other preprocessing to deal with this. Also a large amount of data was missing, which required additional thought into how our algorithms could process it. To create the frontend it also required learning the google-maps API from scratch. There is also surprising complexity and non-uniformity in the coordinate systems different datasets used to represent locations - it's not just as simple as a global latitude and longitude!
Accomplishments that we're proud of
Wrangling the data into a usable format and generally dealing with large climate datasets was a rewarding experience and something out of both our comfort-zones. Also the feeling when google maps finally rendered our data-overlays was immensely satisfying.
What we learned
What's next for UK Climate Projections - Interactive Map
Fully parsing and integrating the rest of the climate variables - such as precipitation and humidity would be an obvious first step. Also general usability improvements and making a bigger point about how an individual can help shift the balance of probabilities towards a lower emission scenario. Also, there are other datasets and projections than the UK Climate Projections and including these other models and ultimate global instead of just UK projections would be worthwhile!