​​GET GEEKY - Challenge
​​Weather categories
  • ​​construct formal anthology on WMO/ICAO weather categories/symbols
  • ​​connect SVG WMO/ICAO symbols to existing taxonomy/onthology
  • ​​refine SVG/WMO ICAO weather symbols
  • ​​put SVG weather symbols into QGIS distro
​​
​​Impacts/applications
  • ​​natural energy e.g. wind speed in time series to study impact on energy sector (gas, natural)
  • ​​rostering scheduling advice for construction engineering 
  • ​​impact of extreme weather events on crops and prices of food
  • ​​flood alerts (severe, emergency) have to be sent to the people who live in the affected region: preparedness against calamities
  • ​​3D visualisation of impact of fires on air pollution/chemistry
  • ​​Frire trends: try to visualise trends in fire data with sliding scale of world map region
  • ​​based on weather condition over a specific period a suitable crop need to be cultivated (to produce suitable crops and get good yield)
  • ​​find relationship between weather and derivatives markets. Then use weather to predict the price of derivates in the future (NN)
  • ​​where should I live? current and future data of different cities/locations based on the temp/pollution/fires. Map based view on each of above features
​​
​​Services
  • ​​visualise air above you (wind temp cloud evolution over time) 
  • ​​create twitter feeds or similar to show whether yesteraday’ forecast was correct
  • ​​app that can monitor surface wind speed and give strong surface wind warning if wind speed is over threshold
  • ​​design met map stylesheet (preferably CSS)
  • ​​design a generic data tiling service (start from maptling)
​​
​​API/plugins
  • ​​develop a QGIS plugin for data discovery and download from ECMWF catalogue
  • ​​generalise OpenData THREDDS to non-met data (i.e. video, audio, DNA)
  • ​​define high level data sub-setting service (API)
​​
​​Miscellaneous
  • ​​unusual data assimilation: pressure sensors in mobile phones
  • ​​learn normalisation parameters for dashcams
  • ​​Ensemble outliers: 51 forecast- how often do statistical outliers model correctly (e.g. 1/51 members forecast a storm that occurs, 50/5 do not?
  • ​​Replace shape file with another open format.
​​
​​
​​
​​
​​