Spatial and temporal characteristics of atmospheric rivers in Scandinavia and their influence on the regional precipitation

EGU 2025

Erik Holmgren

Hans Chen

Chalmers University of Technology

Online version

Atmospheric Rivers and their impacts on local climate

Examples of well studied regions:

• North America (e.g. Dettinger et al. 2011; Rutz, Steenburgh, and Ralph 2014; Guan and Waliser 2019; Collow et al. 2022.)

• Flooding in the Yangtze basin (Pan and Lu 2019).

• Europe (e.g. Stohl, Forster, and Sodemann 2008; Lavers and Villarini 2013, 2015; Ionita, Nagavciuc, and Guan 2020)

• Polar regions (e.g. Gorodetskaya et al. 2014; Lauer et al. 2023)

Missing: Detailed analysis of ARs over Scandinavia

• Atmospheric Rivers are as the names suggest kind of a river in the atmosphere.
• They are long and narrow and transport unusually high amounts of moisture over long distances.
• But unlike a regular river they are not stationary in space.

• The moisture transported by the ARs contributes to the regional precipitation.

• One example of a region where ARs are particularly well studied are the west coast of North America, where previous research has found that up to around 50% of the annual precipitation occurs during ARs.

• Similarly, other studies found that ARs are affecting floods in both the Yangtze river basin in Europe.

Atmospheric Rivers and their influence on precipitation over Scandinavia

• When and where do ARs intersect Scandinavia?
• What influence does ARs over Scandinavia have on precipitation?

Scandinavia as defined in the study: the combination Denmark, Norway, and Sweden.

• In our research we wanted to produce a detailed climatology of AR activity over Scandinavia, and how it affects the local precipitation.
• This can be split into two somewhat smaller questions
• Firstly, we want to know where and when ARs intersect Scandinavia. Are there any local hotspots, and how do these differ over time? How do these ARs vary with the large scale circulation, such as the NAO?
• Secondly, we want to evaluate the influence ARs have on precipitation over Scandinavia.

Analysis

Data

• ARTMIP Tier 2 (Collow et al. 2022)
  • Four algorithms (ARDTs)
  • Applied to ERA5
• ERA5 precipitation
• 1980-2019

Method

• Select ARs that intersect Scandinavia

• Track ARs over time

• Cluster AR tracks using K-means

• AR-Precipitation time steps

• In our analysis, we have been using four different algorithms from the ARTMIP project.

  • The algorithms have been applied to ERA5 data.

• And ERA5 is an reanalysis product, meaning it is a gridded dataset.

• Because of this we also used ERA5 precipitation data to investigate the AR influence on precipiation.

• From the ARTMIP ensemble we have filtered out and tracked the ARs that intersect Scandinavia.

• The

• To these filtered ARs we applied a K-means cluster to find if there are any common AR patterns.

• We could also group the ARs based on NAO.

• Before we move on to our results, I just briefly want to explain how the tracking and clustering were done.

Annual average AR frequency and precipitation

Frequencies

• Maximum over Denmark
  • Up to 5%
• > 2% most of domain

Influence on precipitation

• Up to 1300 mm during ARs
  → 45% of annual total

• This brings us to the first results.

• Explain what an ensemble is.

• In this figure we see the annual average AR frequency in the top panel, the absolute annual average precipitation during ARs in the middle, and the AR precipitation fraction of the total precipitation in the bottom.

• Some things to note is the regional maximum AR frequency that is located over Denmark, and is around 20%. While most of the region sees an AR frequency above 10%.

• Regarding the precipitation, we see that up to 36% of the toal annual precipitation occurs during ARs, in the Southern west coast of Norway.

• And finally, the field correlation between the frequency and precipitation pattern is 0.66, moderate.

• so it appears that it is raining more during AR than otherwise.

AR clusters over Scandinavia

• Southern coast of Norway (1)
  • Up to 1% of the time, 9% of annual precipitation
  • South to North
• The northern coast of Norway (2)
  • 2.5% of the time, 15% precipitation
• “Central” Sweden (3)
  • Relatively high frequencies, 16% of annual precipitation
  • West to East
• Southern Denmark (4)
  • High frequencies
  • 22.5% of annual precipitation

• But, as I’ve mentioned I we wanted to break this down further by using clustering the ARs. And through this we found four main AR patterns over Scandinavia which are shown in the current figure.

• There are two southern patterns, cluster 1 and 4, panel a and g which are centered over Denmark. These are the most frequent ARs, with annual frequencies up to 10%.

• And the ARs belonging to these patterns are responsible for 16 and 13 percent of the annual precipitation respectively.

• But what is interesting about these is that they represent ARs with different orientations - as you can see pattern 1, in panel a, is more oriented from north to south, whereas pattern 4 is oriented in a more west-easterly direction.

• The field correlation between the ARs and precipitation is also different, with pattern showing a relatively high correlation.

• The second most frequent ARs are found in pattern 3 (looking at panel e), which is centred over the more northern west coast of norway. And they show relatively high frequencies and influence on precipitation, with a strong correlation between the AR and precipitation pattern.

• ARs in pattern 2 are the least frequent, and are located over the more central parts of the domain, and also show less precipitation.

• So this is great, we have an idea of where ARs intersect Scandinavia and if they influence precipitation in these locations, but if you remember, I also said that I want to know when. And this brings us to our next slide.

AR clusters: seasonal variation

• Autumn (Sept-Oct-Nov) most frequent
  • Least: Spring
• Northern & Central cluster (2 & 3)
  • Summer & Autumn
• Less variation in the southern cluster (4)

• Here we are breaking down the AR patterns from the previous slide, looking at how they vary throughout the year.

• In this figure have the pattern again as rows, and the columns shows the frequency pattern during the four seasons.

• So generally what we see here, is that the ARs are most frequent during autumn and least frequent during spring.

• and this is quite consistent for all the patterns.

• However, there are some interesting differences. In the two southern patterns, AR activity is high through most of the year, with the exception of spring.

• While ARs in the northern pattern are mostly active during summer and autumn. Here we also see a southward shift of the pattern when going from summer to autumn.

• But to better understand why we see these differences, we also need to look into how changes in the large scale circulation of the atmosphere affect these ARs.

AR clusters: variation with NAO

• Strong positive phases (NAO++)
  westerly flows →
• Northern and central cluster (2 & 3)
  • Large difference NAO-- and NAO++
• Southern cluster (4)
  • High frequencies also during negative phases

• To somewhat do this, we looked into how the AR patterns varied with the phases of the north atlantic oscillation, since the weather patterns during phases of the NAO are relatively well known.
• So in this, we again have the different patterns on the rows, and the AR grouped by NAO in the columns, which goes from negative to positive.
• And what we see here is that AR activity peaks during the strong positive phases (above 0.5). Which could then be explained by the stronger westerly flow that we generally observe during a positive NAO.
• However, it is worth noting that the influence on pattern 4 is relatively low.

Conclusions

Some highlights

• ARs are a common feature in the Scandinavian climate system.
  • AR activity peaking over Denmark, and the northern coast of Norway.
  • Summer & Autumn
  • NAO++
• ARs have a strong influence on precipitation over Scandinavia.

Manuscript to Weather and Climate Dynamics in preparation

Get in touch:

• erikholmgren.info

• erik.holmgren@chalmers.se

• So to conclude, ARs appear to be a common feature in the Scandinavia climate, with AR activity peaking over southern Sweden and Denmark, and the northern west coast of Norway.
• And these ARs also have a relatively strong influence on precipitation, where up to 36% of the annual precipitation is received during ARs.

Extra: NAO seasonal cycle

References

Collow, A. B. Marquardt, C. A. Shields, B. Guan, S. Kim, J. M. Lora, E. E. McClenny, K. Nardi, et al. 2022. “An Overview of ARTMIP’s Tier 2 Reanalysis Intercomparison: Uncertainty in the Detection of Atmospheric Rivers and Their Associated Precipitation.” Journal of Geophysical Research: Atmospheres 127 (8): e2021JD036155. https://doi.org/10.1029/2021JD036155.

Dettinger, Michael D., Fred Martin Ralph, Tapash Das, Paul J. Neiman, and Daniel R. Cayan. 2011. “Atmospheric Rivers, Floods and the Water Resources of California.” Water 3 (2): 445–78. https://doi.org/10.3390/w3020445.

Gorodetskaya, Irina V., Maria Tsukernik, Kim Claes, Martin F. Ralph, William D. Neff, and Nicole P. M. Van Lipzig. 2014. “The Role of Atmospheric Rivers in Anomalous Snow Accumulation in East Antarctica.” Geophysical Research Letters 41 (17): 6199–6206. https://doi.org/10.1002/2014GL060881.

Guan, Bin, and Duane E. Waliser. 2019. “Tracking Atmospheric Rivers Globally: Spatial Distributions and Temporal Evolution of Life Cycle Characteristics.” Journal of Geophysical Research: Atmospheres 124 (23): 12523–52. https://doi.org/10.1029/2019JD031205.

Ionita, Monica, Viorica Nagavciuc, and Bin Guan. 2020. “Rivers in the Sky, Flooding on the Ground: The Role of Atmospheric Rivers in Inland Flooding in Central Europe.” Hydrology and Earth System Sciences 24 (11): 5125–47. https://doi.org/10.5194/hess-24-5125-2020.

Lauer, Melanie, Annette Rinke, Irina Gorodetskaya, Michael Sprenger, Mario Mech, and Susanne Crewell. 2023. “Influence of Atmospheric Rivers and Associated Weather Systems on Precipitation in the Arctic.” Atmospheric Chemistry and Physics 23 (15): 8705–26. https://doi.org/10.5194/acp-23-8705-2023.

Lavers, David A., and Gabriele Villarini. 2013. “The Nexus Between Atmospheric Rivers and Extreme Precipitation Across Europe.” Geophysical Research Letters 40 (12): 3259–64.

———. 2015. “The Contribution of Atmospheric Rivers to Precipitation in Europe and the United States.” Journal of Hydrology 522 (March): 382–90. https://doi.org/10.1016/j.jhydrol.2014.12.010.

Pan, Mengxin, and Mengqian Lu. 2019. “A Novel Atmospheric River Identification Algorithm.” Water Resources Research 55 (7): 6069–87. https://doi.org/10.1029/2018WR024407.

Rutz, Jonathan J., W. James Steenburgh, and F. Martin Ralph. 2014. “Climatological Characteristics of Atmospheric Rivers and Their Inland Penetration over the Western United States.” Monthly Weather Review 142 (2): 905–21. https://doi.org/10.1175/MWR-D-13-00168.1.

Stohl, A., C. Forster, and H. Sodemann. 2008. “Remote Sources of Water Vapor Forming Precipitation on the Norwegian West Coast at 60\(^\circ\)N–a Tale of Hurricanes and an Atmospheric River.” Journal of Geophysical Research: Atmospheres 113 (D5). https://doi.org/10.1029/2007JD009006.