SLIDE 1
AGAT: Automatic Generation of Text Weather Forecast with - - PowerPoint PPT Presentation
AGAT: Automatic Generation of Text Weather Forecast with - - PowerPoint PPT Presentation
AGAT: Automatic Generation of Text Weather Forecast with Spatio-Temporal Aggregation Lior Perez 99th AMS annual meeting 100 French districts Situation of October 26 2018 evening 11:00 PM 2:00 AM 5:00 AM 8:00 AM 21:00 UTC 00:00 UTC 03:00
SLIDE 2
SLIDE 3
Situation of October 26 2018 evening
11:00 PM
21:00 UTC
Rain/snow limit (min) 2000m Rain/snow limit (min) 1850m Rain/snow limit (min) 1500m Max district elevation: 1682m
Causses Aubrac
2:00 AM
00:00 UTC
5:00 AM
03:00 UTC
8:00 AM
06:00 UTC
SLIDE 4
Text generated by AGAT
Original AGAT text (in French):
■
Le ciel est bien chargé. Les nuages donnent rapidement quelques pluies éparses. Ces précipitations tombent d'abord des Causses à l'Aubrac en soirée, puis partout ailleurs en seconde partie de nuit. Il pleut, puis la limite pluie-neige s'abaisse progressivement à 1550 mètres au lever du jour.
Translation (Google Translate with minor adaptations):
■
The sky is very cloudy. Clouds quickly give scattered rains. This rainfall falls first from Causses to Aubrac in the evening, then everywhere else in the second half of the night. It rains, then the rain-snow limit drops gradually to 1500 meters at sunrise.
SLIDE 5
How does it work?
- 1. Sort
areas
- 3. Post-process
template
- Reduce complexity
- Find a template written
by a human forecaster for a similar situation
- Add name of places
- Add information on
precipitation type (rain / snow)
- Add time information
- 2. Get template
from closest « reference situation »
SLIDE 6
Areas numbers and weather matrix
1 2 3 4 5 6 7
21 UTC 00 UTC 03 UTC 06 UTC
Area 1 Area 2 Area 3 Area 4 Area 5 Area 6 Area 7
Weather matrix of October 26 2018 evening
SLIDE 7
- 1. Sort areas
- 1. Sort
areas
- 3. Post-process
template
- 2. Get template
from closest « reference situation »
SLIDE 8
- 1. Sort areas to reduce complexity
21 UTC 00 UTC 03 UTC 06 UTC
Area 1 Area 2 Area 3 Area 4 Area 5 Area 6 Area 7
21 UTC 00 UTC 03 UTC 06 UTC
Area 1 Area 5 Area 7 Area 2 Area 3 Area 4 Area 6 Areas with « worst » weather first Weather matrix unsorted
SLIDE 9
- 2. Get template from closest « reference situation »
- 1. Sort
areas
- 3. Post-process
template
- 2. Get template
from closest « reference situation »
SLIDE 10
- 2. Find closest « reference situation »
Our situation Closest reference situation We have defined an euclidian distance between pictograms => we can compute distance between weather matrices
SLIDE 11
- 2. Find closest « reference situation »
21 UTC 00 UTC 03 UTC 06 UTC
Our situation Closest reference situation
21 UTC 00 UTC 03 UTC 06 UTC
Area 1 Area 5 Area 7 Area 2 Area 3 Area 4 Area 6
SLIDE 12
- 2. Template text for reference situation
Original template text (in French):
■
Le ciel est bien chargé. Les nuages donnent rapidement quelques <précipitations < pluies | chutes de neige > 00 06> éparses. Ces précipitations tombent d’abord <région précipitations 00> en soirée, puis partout en seconde partie de nuit. <LPN 00 06>.
Translation (Google Translate with minor adaptations):
■
The sky is very cloudy. Clouds quickly give scattered <precipitations < rains | snow > 00 06>. This rainfall falls first <area precipitation 00> in the evening, then everywhere else in the second half of the night. <RAIN_SNOW_LIMIT 00 06>
SLIDE 13
How did we build the « reference situations » database?
■
Data: 2 years of forecast history
■
Unsupervised learning: k-means clustering
■
For each cluster, a human forecaster has written a template text
1500 clusters
homogeneous weather situations
SLIDE 14
- 3. Post-process template
- 1. Sort
areas
- 3. Post-process
template
- 2. Get template
from closest « reference situation »
SLIDE 15
- 3. Post-process tags
■
Rain / snow <precipitations < rains | snow > 00 06> From 00 UTC to 06 UTC, only rain, no snow. => « rains »
■
<area precipitation 00> At 00 UTC, precipitations on areas 1, 5 and 7 In meta-data dictionary: 1+5+7 = « from Causses to Aubrac »
■
<RAIN_SNOW_LIMIT 00 06> => « It rains, then the rain-snow limit drops gradually to 1500 meters at sunrise. »
21 UTC 00 UTC 03 UTC 06 UTC
Area 1 Area 5 Area 7 Area 2 Area 3 Area 4 Area 6
SLIDE 16
- 3. Final text
Original AGAT text (in French):
■
Le ciel est bien chargé. Les nuages donnent rapidement quelques pluies éparses. Ces précipitations tombent d'abord des Causses à l'Aubrac en soirée, puis partout ailleurs en seconde partie de nuit. Il pleut, puis la limite pluie-neige s'abaisse progressivement à 1550 mètres au lever du jour.
Translation (Google Translate with minor adaptations):
■
The sky is very cloudy. Clouds quickly give scattered rains. This rainfall falls first from Causses to Aubrac in the evening, then everywhere else in the second half of the night. It rains, then the rain-snow limit drops gradually to 1500 meters at sunrise.
SLIDE 17
Conclusion
■
AGAT is in production since 2018
■
100 text forecast generated every 15 minutes
■
A lot of time spent on validation
―
Run AGAT on real weather situations
―
Check the generated texts
―
Correct the reference situations database
■
Easy to maintain: adding new reference situations is simple
Lior.perez@meteo.fr