Our vision is to provide a 'one-stop' view for users from various industries to have a 'one-stop' view of the environmental and socio-economic impact caused by COVID.
There are two main objectives we set ourselves for this challenge:
After analysing a wide range of data sets provided by NASA (e.g. CO2, SO2, PM10, PM2.5 pollution data) and socio-economic indicators data (e.g. unemployment per month for each European country and consumption and development data in key industries such as transportation and energy) as well as COVID-related data (e.g. confirmed cases, recovered cases per country) we didn't identify a strong direct or indirect correlations between them. All data sets were analyzed for each European country based on monthly information or aggregate value.
However, we believe there is still a need for a data aggregated platform where all these key environmental data and socio- economic indicators across multiple industries and European countries could be displayed.
As in Europe, although most people are aware of the economic impact COVID-19 brings to their hometowns and everyday lives, the environmental impact resulted from nationwide lockdown and closures of millions of businesses and stores across Europe is not widely understood. We believe if we can make this information more easily accessible, more people will be able to understand the wider environmental impact caused by COVID-19.
As mentioned above, we set ourselves two main objectives to tackle this challenge:
1. To find whether there is any strong correlation between the NASA satellite measurements, social and economic data from both EU and UN and GitHub COVID-19 data
2. To find a way to display all these insights in an easily accessible platform so users could have a holistic view of the wider economic and environmental impact caused by COVID.
For the first objective, we analysed a wide range of open source dataset from NASA satellite, UN, and EU and identifying rich and complementary datasets. However, we face challenges when trying to identify clear correlations between economic development, quality of life, general socio and economic indicators, and incidence and evolution of COVID-19.
After partially invalidate our assumption on the first objectives due to the time constraints, we shift our effort focusing on our second objective - as we believe there is still a need for an easily accessible one-stop-shop for people who are interested in having a better understanding of the wider socio-economic and environmental impact by COVID-19 to have a holistic view of a local, regional and industry level. We've successfully created an interactive dashboard demo which we've included in the presentation video.
For data analysis, we leverage the satellite data from NASA, and registered the geographical measured values on European countries. We used Python as the main analysis and coding language, with Jupyter
Notebooks for analysis, running it with Anaconda. The app ingests the pre-processed data and prepares it to be displayed in an easily accessible interface.
From platform prototyping perspective, we developed the backend using Python with flash and pandas, and frontend with Angular, Bootstrap, d3 and charts js.
We ingest, validate, perform corrections on John Hopkins data. We are also analyzing the COVID-19 data for Europe, focusing on the epidemics evolution in each country as well as on correlations with social and economical factors extracted from UN data.
We are also investigating the correlation of covid-19 metrics with the pollution data from NASA satellite measurements.
We ingest, process and enrich data related to air and maritime transport extracted from UN open source data. Starting from the initial data sets, we added information using the lookup files and transform, where needed the quarterly data into monthly data by using the quarterly data as a average value for 3 months.
We investigated the correlation of transport data with the pollution data from NASA satellite measurements.
We investigate various factors which are related to impact of COVID-19, either directly (like correlation between morbidity and mortality to healthcare sector development in each country) or indirectly (through the measures imposed by each country) like recent months unemployment changes (spotting facts like different unemployment levels in EU countries for youngs, depending on sex) or prevalence of internet access in various countries (important due to massive shift to WFH and online education in recent months).
We were as well looking to sectors which sees now major shifts, like transportation, or might have relationship with pollution: energy and industrial sectors that relay heavily on large energy consumption.
We also looked to factors that might have an impact on COVID-19 dynamics, like education, literacy, healthcare, population age, forest percent from total land as well as GDP per capita, population density.
Here are the analysis Notebooks:
Those files contain code, label and other nomenclators information
This lookup contains airport code information associated with the municipality, country and region information. Original source: https://pkgstore.datahub.io/core/airport-codes/831/datapackage.jsonhttps://pkgstore.datahub.io/core/airport-codes/airport-codes/archive/edda13b18a6832d040c1ff19fbd4a8fd/airport-codes.csv
Data available in this lookup is useful for mapping port abbreviation with country names. Original source: https://ec.europa.eu/eurostat/cache/metadata/en/mar_esms.htm
In this lookup is needed for maritime port data. Original source: http://dd.eionet.europa.eu/vocabulary/eurostat/tra_cov/view?page=1#vocabularyConceptResults
This lookup contains information about the transport measurement abbreviation and labelling. Original source: http://dd.eionet.europa.eu/vocabulary/eurostat/tra_meas
Data available here is useful for mapping contry code information and use Latitude and Longitude coordinates. Original source: https://developers.google.com/public-data/docs/canonical/countries_csvhttps://latitudelongitude.org/https://raw.githubusercontent.com/lukes/ISO-3166-Countries-with-Regional-Codes/master/all/all.csv
The data available here contains ICAO and IATA codes associated with municipality and country information. Original source: https://airportcodes.io/en/continent/europe/
Here we are showing few insights in the data processing and analyses we performed.
For complete analysis review, navigate to each of the respective Notebook.