About
Im Rahmen unserer Reihe Umweltdaten - Hiveeyes und Datenmischwerk wird hier die Luftdatenpumpe für die Daten von IRCEL - CELINE: Das Feinstaubmeßnetzwerk der Belgischen Interregionalen Umweltagentur vorbereitet.
Prototypische Implementierung
Mit Luftdatenpumpe 0.14.1: luftdatenpumpe/irceline.py at 0.14.1 · earthobservations/luftdatenpumpe · GitHub.
IRCELINE stations
$ luftdatenpumpe stations --network=irceline | jq length
2019-04-24 00:44:49,669 [luftdatenpumpe.source ] INFO : Applying filter: {}
2019-04-24 00:44:49,672 [luftdatenpumpe.commands ] INFO : Acquiring list of stations from IRCELINE. source=api
2019-04-24 00:44:49,673 [luftdatenpumpe.source.irceline] INFO : Requesting IRCELINE live API at http://geo.irceline.be/sos/api/v1/stations
2019-04-24 00:44:49,720 [luftdatenpumpe.commands ] INFO : Acquired #105 stations
2019-04-24 00:44:49,720 [luftdatenpumpe.commands ] INFO : Will publish data to ['json+stream://sys.stdout']
2019-04-24 00:44:49,720 [luftdatenpumpe.engine ] INFO : Configuring data sink "json+stream://sys.stdout" with domain "stations"
2019-04-24 00:44:49,720 [luftdatenpumpe.engine ] INFO : Emitting to target data sinks, this might take some time
2019-04-24 00:44:49,744 [luftdatenpumpe.engine ] INFO : Processed 105 records
106
Ingress format
As coming from the IRCELINE SOS REST API.
Egress format
As emitted from Luftdatenpumpe.
IRCELINE readings
… kommen scheinbar nur einmal pro Stunde rein. Ist vermutlich normal.
$ luftdatenpumpe readings --network=irceline --progress | jq length
2019-05-24 22:48:12,029 [luftdatenpumpe.commands ] INFO : Acquiring readings from network "irceline" with source "api"
2019-05-24 22:48:12,138 [luftdatenpumpe.source.common ] INFO : Processing 106 items
100%|██████████████████████████████████████████████████████████████████████████████| 106/106 [00:00<00:00, 1374.45it/s]
2019-05-24 22:48:12,223 [luftdatenpumpe.source.irceline ] INFO : Requesting IRCELINE live API with timespan "PT12h/2019-05-24T22:00:00+02:00" and 529 timeseries
2019-05-24 22:48:12,223 [luftdatenpumpe.source.common ] INFO : Processing 529 items
46%|████████████████████████████████████▏ | 242/529 [00:00<00:00, 401.08it/s]
106
SOS API
http POST 'http://geo.irceline.be/sos/api/v1/timeseries/getData' timeseries:='[6693]' timespan='PT3h/2019-04-24T01:00:00Z' --print bB
API Request/Response
Request
{
"timeseries": [
6693
],
"timespan": "PT3h/2019-04-24T01:00:00Z"
}
Response
{
"6693": {
"values": [
{
"timestamp": 1556056800000,
"value": 36.1
},
{
"timestamp": 1556060400000,
"value": 34.6
},
{
"timestamp": 1556064000000,
"value": 33.1
},
{
"timestamp": 1556067600000,
"value": "NaN"
}
]
}
}
Ingress format
As coming from the IRCELINE SOS REST API.
Egress format
As emitted from Luftdatenpumpe.
Luftdatenpumpe is now able to process air quality data coming from the 52°North Sensor Observation Service instance of the Belgian Interregional Environment Agency (IRCEL - CELINE) referenced above. If this sparks your interest, you might also enjoy visiting and exploring the respective staging system at https://vmm.hiveeyes.org/.
Getting from one to two data sources here was the most important thing the whole architecture needed before thinking about adding even more sources. While things like BLUME Feinstaub-Daten API im SOS Standard might be unlocked easily now, we are also looking at other APIs like Luchtmeetnet: Dutch air quality networks, PREV'AIR - Air information Ressource in France or UK AIR: Air Information Resource.
Dieses kleine Fundstück passt hier wahrscheinlich am besten hin.