jgehrcke / covid-19-germany-gae

COVID-19 statistics for Germany. For states and counties. With time series data. Daily updates. Official RKI numbers.
MIT License
145 stars 48 forks source link

Incorporate new Landkreis-based data sources #43

Closed jgehrcke closed 4 years ago

jgehrcke commented 4 years ago

Those data sources are relatively early in the Meldekette, and probably quite credible. Will elaborate in the comments.

jgehrcke commented 4 years ago

Die Daten stammen von den zuständigen Behörden der Landkreise und Bundesländer, die ZEIT ONLINE mehrmals am Tag abruft und hier aktualisiert.

Da es wegen der Meldekette zu Verzögerungen kommt, fragen wir regelmäßig die bestätigten Fälle der 401 Stadt- und Landkreise direkt ab. Diese spiegeln den aktuellsten Stand in Deutschland am besten wider. Diese Zahlen können daher von anderen Quellen wie dem Robert Koch-Institut (RKI) oder der Johns Hopkins Universität abweichen.

jgehrcke commented 4 years ago

Screenshot from 2020-03-24 13-46-34

jgehrcke commented 4 years ago

for the country-wide case count this was done in https://github.com/jgehrcke/covid-19-germany-gae/pull/45

/now reflects that count already

jgehrcke commented 4 years ago

Screenshot from 2020-03-24 19-09-44

jgehrcke commented 4 years ago

--> The crowd-sourced spreadsheet and ZEIT ONLINE both report ~33.400 cases at this current time (~19:15 local time). That's quite promising. I assume that both teams take the same approach, don't copy from each other, and arrive at roughly the same number. If that's true then that's good news.

CC @coezbek @nikola @TG9541 @ms82494

stefan123t commented 4 years ago

Hi Jan-Philip, I just updated Lisas blog with the following comment and URL from the ESRI server, somehow they have the RKI data per Landkreis. Maybe you could use the URL to persist the Landkreis data per day, so we have the historical data saved for posterity.

Kind regards, Stefan

_Hi Lisa, you can get the RKI_Landkreise Feature layer with the following GET request. It contains a description of the Attributes queried plus the data in a JSON document:

https://services7.arcgis.com/mOBPykOjAyBO2ZKk/arcgis/rest/services/RKI_Landkreisdaten/FeatureServer/0/query?f=json&where=1%3D1&returnGeometry=false&spatialRel=esriSpatialRelIntersects&outFields=*&orderByFields=cases%20desc&outSR=102100&resultOffset=0&resultRecordCount=1000&cacheHint=true

I do not know how often they update the data or how frequently you will be allowed to query their server for this information, nor how they get the data from RKI anyway. Maybe you could also inquire with RKI directly to get the data somewhere.

Kind regards, Stefan_

jgehrcke commented 4 years ago

Thank you @stefan123t for the message! Please keep an eye on this repo. My plan certainly is to compare history across data sources with fine-grained spatial resolution.

stefan123t commented 4 years ago

I just see you already have the ESRI RKI Landkreis Feature Service on record of #46.

If you want I have looked for the LK data on the press release sites of the states, but not all states maintain a history of these figures, as you seem to scrape the Gesundheitsamt of the counties, this would give you the numbers consolidated on state level.

Baden-Württemberg Link to XLSX below the map https://sozialministerium.baden-wuerttemberg.de/de/gesundheit-pflege/gesundheitsschutz/infektionsschutz-hygiene/informationen-zu-coronavirus/

Bayern Numbers in an HTML table https://www.lgl.bayern.de/gesundheit/infektionsschutz/infektionskrankheiten_a_z/coronavirus/karte_coronavirus/index.htm

Berlin Nur als Bild ? https://www.berlin.de/sen/gpg/service/presse/2020/pressemitteilung.906973.php

Brandenburg https://msgiv.brandenburg.de/msgiv/de/presse/pressemitteilungen/detail/~12-03-2020-30-corona-faelle-in-brandenburg

Bremen https://www.senatspressestelle.bremen.de/detail.php?gsid=bremen146.c.331858.de&asl=bremen02.c.732.de

Hamburg https://www.hamburg.de/coronavirus/

Hessen https://soziales.hessen.de/gesundheit/infektionsschutz/coronavirus-sars-cov-2/taegliche-uebersicht-der-bestaetigten-sars-cov-2-faelle-hessen

Mecklenburg-Vorpommern Bisher keine Pressemitteilungen / Fallzahlen gefunden:

Niedersachsen https://www.niedersachsen.de/Coronavirus

Nordrhein-Westfalen https://www.mags.nrw/coronavirus-fallzahlen-nrw

Rheinland-Pfalz https://msagd.rlp.de/de/unsere-themen/gesundheit-und-pflege/gesundheitliche-versorgung/oeffentlicher-gesundheitsdienst-hygiene-und-infektionsschutz/infektionsschutz/informationen-zum-coronavirus-sars-cov-2/

Saarland Bisher keine Pressemitteilungen / Fallzahlen gefunden:

Sachsen https://www.sms.sachsen.de/coronavirus.html

Sachsen-Anhalt https://verbraucherschutz.sachsen-anhalt.de/hygiene/infektionsschutz/infektionskrankheiten/coronavirus/

Schleswig-Holstein https://www.schleswig-holstein.de/DE/Landesregierung/VIII/Presse/PI/2020/200313_VIII_Corona_tabelle.html

Thüringen keine tabellarische Übersicht https://www.tmasgff.de/covid-19

Statista Übersicht für Deutschland https://de.statista.com/statistik/daten/studie/1100750/umfrage/fallzahl-des-coronavirus-covid-19-nach-bundeslaendern/

jgehrcke commented 4 years ago

Also confer https://github.com/jgehrcke/covid-19-germany-gae/issues/58

ndeuma commented 4 years ago

We have been working on a project for providing county ("Landkreis") data since the #WirVsVirus hackathon. I'm only the frontend guy, but you can find links to the other repositories on the entry page here. https://github.com/ndeuma/covid19-casedata

stefan123t commented 4 years ago

Dear Niklas, I have been able to compose a map from BKG and City of Berlin which matches the data released everyday by RKI through ARCGIS. Those maps are now on the Datawrapper River, but I would need to update the CSV source daily from ARCHIS as it is done with the other maps on that page ? Feel free to reuse the Geojson giving reference to BKG and Berlin as well as data courtesy of RKI/ArcGIS. Kind regards, Stefan

cases and cases per 100k https://river.datawrapper.de/#popup/ASPdN https://river.datawrapper.de/#popup/0AEpL

deaths and death rate https://river.datawrapper.de/#popup/YeEOY https://river.datawrapper.de/#popup/lytZ4

jgehrcke commented 4 years ago

Discussion can of course continue, but I close this issue because the underlying task tracked here is done.

Feel free to consume:

The same for deaths: