Consider better ways to store and display geographical risk data

[zarino]

My instinct is that the data we’re currently importing and displaying for these metrics is too low-level:

• Flood risk data (DATASET: Flood risk data #313)
• Air pollution data (DATASET: Air pollution data #309)
• Water quality data (Sewage / water pollution data #196)

I feel like I need a qualification from Mensa to understand this…

Let alone this…

Friends of the Earth took a neat approach to this – they identified "neighbourhoods" that are well above WHO guidelines on air pollution, and then count them up, so you get a figure like "5 high pollution neighbourhoods in this local authority". Or, I guess, since their approach is based on LSOAs, you could even do "10% of this local authority is high pollution". I wonder whether we could calculate the same at a constituency level?

Comparing against the WHO guidelines is a nice touch, because it grounds the numbers, gives you a clear indication of what’s a "bad" number, versus "ok".

I wonder whether there’s a similar guideline for flood risk? Even if not, just boiling it down to "this constituency does / doesn’t contain high flood risk areas" (where "high" is defined by the source data from Defra) would be an improvement on the current DataType approach.

[zarino]

As mentioned in #180, Toby from FoE has shared their "percentage of constituency with quality below WHO standards" data (for existing constituencies) and I’ve added it to the shared drive.

[zarino]

Going to close this ticket as we’ve pretty much decided what to do (show proportion of constituency with air/water quality below some recognised standard). Now we just need to:

• Import air quality / WHO data from FoE
• Find recognised (WHO?) standard for water quality, and compare it to the data we’re collecting in Sewage / water pollution data #196.