Significant multi-decadal predictability is already known for sea surface temperature anomalies in the North Atlantic
and Nordic Seas. Those anomalies have an impact on typical large-scale circulation patterns of the Atlantic-
European region. More specifically, it has been suggested that frequency and persistence of the atmospheric blockings
in this region is strongly affected by the wintertime anomalies in the ocean through a certain modifications
of the eddy pumping mechanisms and inverse enstrophy cascades. Our study reveals that the large-scale circulation
patterns are strongly correlated with wintertime air pollution episodes in Bergen, Norway. Certain large-scale
circulation regimes (e.g. the West Atlantic or Greenland blockings) lead to local air quality hazards. We assessed
these circulation regimes and their predictability. We modified and applied an atmospheric circulation proxy for
the identification of air pollution episodes. Use of this proxy on data from a high-resolution atmospheric general
circulation model showed a good reproduction of the total number of potentially polluted days per month and their
inter-monthly variability.We also found a link between the persistence of the flow above the Bergen valley and the
occurrence and severity of the local air pollution episodes. Analysis of the large-scale circulation over the North
Atlantic-European region, with respect to air pollution in Bergen, revealed that the persistence in the meteorological
conditions connected to the air pollution episodes is not necessarily caused by large-scale anomalies of the
atmospheric circulation over the Norwegian west coast. It is rather connected to anomalies further upstream as far
away as Greenland. Finally, we conduct a set of very high-resolution simulations with the large-eddy simulation
model PALM to identify the local circulations and boundary layer regimes corresponding to the typical large-scale
meteorological conditions connected to the air quality hazards in Bergen.
Long term predictability of local air quality hazards
1. Long-term predictability of local air
quality hazards and periods of reduced
turbulent mixing in Scandinavia
Igor Esau, Tobias Wolf, Lasse Pettersson and Joachim Reuder
Nansen Environmental and Remote Sensing Centre
Geophysical Institute, University in Bergen
Bjerknes Centre for Climate Research
Bergen, Norway
(EMS2018-292)
04.09.2018 European Meteorological Society Assembly 1
3. Observations
04.09.2018 European Meteorological Society Assembly 3
Observations reveal:
• Atmospheric pollution accumulates
gradually over several days
• Both the peak values and duration of air
quality hazards matter
• Hazardous concentrations are related to
temperature inversion episodes (stably-
stratified atmospheric conditions) and
calm (ani-cyclonic weather)
2017
Persistence of stably stratified
atmospheric conditions
4. Air Quality Index (AQI)
04.09.2018 European Meteorological Society Assembly 4
Wolf T. and I. Esau, 2014: Air quality hazards under present and
future climate conditions in Bergen, Norway, Urban climate, 10,
801-814, doi: 10.1016/j.uclim.2014.10.006
Wolf T., Esau I. & Reuder J. The large-scale circulation during air
quality hazards in Bergen, Norway Tellus A, 2017, 69, 1406265Parallel observations
of air quality
(NOx,PM)
ERA-I meteorology
Selection weather
factors and
thresholds
Application to long-term
meteorological data,
models and reanalyses
AQI corresponds to
weather conditions
highly correlated
with the observed
air quality hazards
5. Long-term AQI
variability (ERA-I)
04.09.2018 European Meteorological Society Assembly 5
Wolf T. and I. Esau, 2014: Air quality hazards under present and
future climate conditions in Bergen, Norway, Urban climate, 10,
801-814, doi: 10.1016/j.uclim.2014.10.006
Wolf T., Esau I. & Reuder J. The large-scale circulation during air
quality hazards in Bergen, Norway Tellus A, 2017, 69, 1406265
Wintertime AQI variability shows cycles with quasi-
period of 10-14 years
Years with more stagnating episodes in winter (DJF):
2009/10, 2010/11, 2012/13
2002/03
1993/94, 1995/96, 1996/97
1984/85, 1985/86
1978/79, 1979/80
The local stagnation episodes require:
• Weather persistence – Longer periods of stable
weather conditions with low wind
• Clear sky – Negative surface heat balance and
cooling to build up stable stratification
6. Long-term wind
variability
04.09.2018 European Meteorological Society Assembly 6
Historical long-term variability of
storminess and blockings – they are not
anti-correlated
Seasonal P99 storm index series for
the pressure triangles.
Wang, X. L.; Wan, H.; Zwiers, F. W.; Swail, V. R.; Compo, G. P.; Allan, R.
J.; Vose, R. S.; Jourdain, S. & Yin, X. Trends and low-frequency
variability of storminess over western Europe, 1878–2007, Climate
Dynamics, 2011 , 37, 2355-2371
Greenland blocking index
7. Spatial pattern with the
maximal local effect
04.09.2018 European Meteorological Society Assembly 7
Greenland (West Atlantic)
blocking locally (Bergen and
Scandinavia) induces:
• Cold
• Dry
• Clear-sky
• Calm (wind from the East)
8. Long-term blocking
variability
04.09.2018 European Meteorological Society Assembly 8
Very high seasonal air pollution in Bergen (NO2)
Previous high attention to air pollution in Bergen (SO2)
Greenland blocking indexHistorical long-term variability of Greenland blocking
Hanna, E.; Cropper, T. E.; Hall, R. J. & Cappelen, J. Greenland
Blocking Index 1851-2015: A regional climate change signal,
International Journal of Climatology, 2016, 36, 4847-4861
9. AQI – Blocking links
(Persistence)
04.09.2018 European Meteorological Society Assembly 9
Wolf T., Esau I. & Reuder J. The large-scale circulation during air
quality hazards in Bergen, Norway Tellus A, 2017, 69, 1406265
• AQ hazard episodes require persistence local weather
• Persistent local weather requires geographically locked
large-scale circulation anomaly
• Large-scale circulation anomaly
• Change slowly – have certain predictability
• Better captured by Climate Prediction Models
10. AQI in Climate
Prediction Models
04.09.2018 European Meteorological Society Assembly 10
Wolf T., Esau I. & Reuder J. The large-scale circulation during air
quality hazards in Bergen, Norway Tellus A, 2017, 69, 1406265
• Large-scale circulation anomaly
• Change slowly – have certain predictability
• Better captured by Climate Prediction Models
ERA-I
MRI-GCM
11. Long-term SAT/SST
variability
04.09.2018 European Meteorological Society Assembly 11
Årthun, M.; Eldevik, T.; Viste, E.; Drange, H.; Furevik, T.; Johnson, H. L. &
Keenlyside, N. S. Skillful prediction of northern climate provided by the ocean,
Nature Communications, 2017, 8:15875
12. Long-term SAT-SST
connections (Statistics)
04.09.2018 European Meteorological Society Assembly 12
Årthun, M.; Kolstad, E. W.; Eldevik, T. & Keenlyside, N. S. Time Scales and Sources
of European Temperature Variability, Geophysical Research Letters, 2018
Dominant time scale (years) of SAT variability during the
cold season (November–April). The regions used to
produce SAT time series for northern Europe (NE; 6–20∘E,
57–70∘N), eastern Europe (EE; 20–40∘E, 51–56∘N), and
southern Europe (SE; 9–5∘W, 38–43∘N) are shown.
13. Actual AQI Prediction
04.09.2018 European Meteorological Society Assembly 13
Increase in AQ hazard episodes in 2017 – 2022
Wolf T. and I. Esau, 2014: Air quality hazards under present and
future climate conditions in Bergen, Norway, Urban climate, 10,
801-814, doi: 10.1016/j.uclim.2014.10.006
14. Added value of
modeling (PALM)
04.09.2018 European Meteorological Society Assembly 14
• Existing models DO NOT predict the highest
concentrations which are the most NEEDED
• Models DO NOT reveal the spatial distribution and
pathways of the air pollution
• Models DO NOT inform policy makers about effective
consequences of their scenario decisions
Daily maximum of hourly mean NO2 concentrations
Observations
Modelresults
Central road traffic (working days) (Saturday) (Sunday)
15. Air Pollution Pathways
04.09.2018 European Meteorological Society Assembly 15
PALM simulations:
• Spatially resolve dispersion
in given weather conditions
• Resolve contribution of
distinct sources of pollutants
• Quantify effects of policy
and management scenarios
Wolf T., Petersson L., Esau I. Spredning og konsentrasjonsdannelse
av NO2 og PM2.5 i Bergen sentrum - et studie med vekt på bidrag
fra skip i havna, NERSC report #370, 2016
16. AQ Scenario Assessment
(Harbor and Fireplaces)
04.09.2018 European Meteorological Society Assembly 16
Wolf T., Petersson L., Esau I. Spredning og konsentrasjonsdannelse
av NO2 og PM2.5 i Bergen sentrum - et studie med vekt på bidrag
fra skip i havna, NERSC report #370, 2016
17. AQ Scenario Assessment
(Tourist Cruises)
04.09.2018 European Meteorological Society Assembly 17
Summer scenario run (PALM) for air pollution
from cruise ships in Bergen harbor
Actual PALM simulations of air pollution from
cruise ships in Bergen harbor on 23 July 2018
Bergen harbor on 23 July 2018
18. Impact and Priorities
04.09.2018 European Meteorological Society Assembly 18
Øystein Hov (Nature, 552, 2017):
Priorities for weather and climate research
• Non-govermental
• Tailored (reflecting user needs)
• Easily accessible (on-line)
• Transferable (formulated as accepted by users)
• Inter-operable (available data/services from others)
• Actual (targeting the areas of interest, urban,
extreme weather conditions and phenomena)
• Seamless modeling
• Seamless model and data fusion
• Able to work with diverse quality and quantity of
information, involve public-private cooperation
• Archievable and tracable
Direct impact
• Bergen Municipality: 50MNOK to replace old wood ovens.
• Development of regular information service to Bergen Harbour.
• Significant changes of road taxes during air pollution events (225 NOK)
20. Conclusions
04.09.2018 European Meteorological Society Assembly 20
• Extreme weather events are caused by large-scale circulation anomalies
• But their most pronounced impact is usually strongly localized
• E.g. AQ hazard episodes in Scandinavia are linked to prolonged period of clear-sky clam weather
• Potential predictability:
• Such weather patterns are related to hemispheric size circulation anomalies (the Greenland blocking)
• The wintertime blocking events reveal long-term variability with a quasi-period of 10-15 years
• The blocking variability is connected with the advective SST anomalies, and hence, could be predictable
• Local pathways and scenario assessment:
• The AQ pathways and impact are localized
• The localization pattern depend of the spatial configuration of emitting sources (traffic, ships, fireplaces)
• Decision scenarios to change this configuration could be assessed in advance at small cost with models
• Perspectives:
• The turbulence-resolving modeling creates opportunity for targeted local climate services for environment-
sensitive business and public authorities
• Prototype of such a service was demonstrated by NERSC researchers in Bergen, Norway
21. Regarding the EMS Annual Meeting 2018, we are pleased to inform you that your following abstract has been
accepted and scheduled as oral, 25 min in session UP1.2, room E I on Tuesday, 04 Sep 2018, 14:15:
EMS2018-292: Long-term predictability of local air quality hazards and periods of reduced turbulent mixing in
Scandinavia by Igor Esau et al.
For details of your presentation as well as of the corresponding session programme please see:
https://meetingorganizer.copernicus.org/EMS2018/session/29295.
Detailed information on pre-registration and requesting a letter of invitation is provided on the conference
webpage https://www.ems2018.eu
20 slides needed
04.09.2018 European Meteorological Society Assembly 21