Most of GIS users know John Snow map to fight the cholera break in Zoho in 1854. Seems that the COVID-19 pandemic is bringing back the importance of GIS: A lot of maps and dashboards have been build to monitor and help on reducing the virus spread.

But more than the map on its self, what John Snow has to inspire us on?
The cholera outbreak occurred on a miasm paradigm, that held that diseases like cholera were caused from the air pollution.

Could you imagine your self arguing with an entire society trying to prove them they are wrong? and proposing a new point of view? Seems that would be something similar with dealing with the actual negationism. And John Snow did it! Not only with a map, worth saying. So let us be inspired by John Snow, not only with the map, but by the fact that he used other evidences to fight not only the cholera outbreak but the miasm point of view.

Why is that so important?
Since 2019 Brazil, like others countries, have a negationist government. So you can have a plot: in 2019, the amazon's biome suffered from several forest fire after several years of reduction on its activities. The Brazzilian president kept saying it was "fake news".

The same happens with the COVID-19 in Brazil. In this context InfoAmazonia, an independent media press, started "Engolindo Fumaça" project, which was selected and is being financed by Big Local News from a Stanford’s university. The objective is to infer whether or not the air pollution from forest fire activities in the Brazilian Amazon biome have been playing a role in aggravating hospital admissions and length of stay during the COVID-19 pandemic. All of that using FOSS4G and open data and using the results to guide a series of journal articles with real histories of people whom is suffering the consequences.

The challenge was: 1) Identify the best data set with PM<2.5 variable; 2) Validate the data sets from point 1 with air pollution sensors; 3) Validate the correlation of PM<2.5 with forest fire occurrences and; 4) to identify if there is a relation between the exposure to PM<2.5 with hospital admissions and length of stay by respiratory syndromes caused by COVID-19.

We analyzed different data sets for particulate matter < 2.5 and compared each of them with MODIS Optical Aerosols Depth. The most suitable PM 2.5 variable was Copernicus Atmosphere Monitoring Service (CAMS). Them we compared CAMS' PM 2.5 values with Acre's Federal University laboratory, which makes part of air pollution sensors network. The fire occurrence were obtained from the satellite S-NPP from which were possible to infer that most municipalities has its PM 2.5 values correlated with fires occurrences.

We are now working on the health data where the challenge is estimating the exposure of those registered on the hospital admission considering a time lag.