GCloudl: Arctic clouds (Research Collaboration)

Significant information from fragmented Russian polar expeditions and stations, need to be processed to improve our capacity to simulate, predict and communicate the vital environmental knowledge.


The project aims to:

  • Extend statistical processing and publication of the historical station data, involving the data from the eastern arctic area into the analysis. This work will follow the methodology and rigor of the Chernokulsky et al. (2017) study for the Norwegian and Barents seas.
  • Investigate the cloud type’s statistics and the cloud type’s transformation in response to the sea ice, atmospheric circulation and temperature changes. This work is connected to the wider Year of the Polar Prediction (YOPP) activity, as a lack of historical context and understanding of the cloud field transformation mechanisms impedes the advance of the polar prediction.

Our hypothesis emphasizes the cloud spatial self-organization, which is maximizing efficiency of the atmospheric vertical mixing. Restructuring of the cloud types have a significant effect on the air-sea-land interaction, precipitation, radiation processes, cloud feedback mechanisms and extreme weather events.

More specifically, we propose to:
A. Process the historical observational cloud data from the Kara, Laptev, East-Siberian, Chukchi seas. The archive data are available for A. Chernokulsky. The processing technology, including conversion between observational standards and correction on moonlight criterion, has been developed and demonstrated in our previous publications.

B. Prepare the combined climatology of the cloud cover and cloud types over the Eurasian Arctic east of the Norwegian Sea. The climatology will include intercomparison with the satellite remote sensing and the reanalysis data. The main attention will be paid to the historical early warming period 1930-1950, known only from stations’ observations, and the recent transition from the phase of the atmospheric dimming (due to anthropogenic aerosol impact) to brightening (due to reduction of aerosol emission in Europe and elsewhere) during 1970-1990.
C. Improve interpretation of the observed cloud type changes in the polar region. We will combine the IAP cloud climatology and the NERSC modelling studies. The convective cloud development and self-organization will be simulated with the turbulence-resolving model PALM.
D. Explore the wider impact of the polar cloud types and the cloud cover changes. The century long cloud climatology will be analysed in connection with the published data on the meridional heat and moisture transport, multi-decadal arctic variability, polar lows, cold air outbreaks and arctic precipitation.

Norsk sammendrag

Forskningssamarbeid om studier av skydekke i Arktis

Arktis er ett av de områder i verden hvor det er hyppigst skydekke. Skyer i Arktis har en betydelig påvirkning på dagliglivet i Arktis og det globale klima. I store deler av året fører skydekket til at det dannes et lag med tåke nær bakken i Arktis, som fører til nær 100% relativ luftfuktighet ved bakken. Oppvarming av Arktis vil endre på denne situasjonen. Sterk konveksjon vil bidra til at skyene i større grad vil bli organisert i adskilte celler som igjen vil føre til at solinnstrålingen vil varme luften i Arktis ytterligere opp. Slike endringer i skydekket vil bli studert over en lengre historisk tidsperioder for å kartlegge variabilitet på ti-års skalaer og mulige tilbakeførings mekanismer fra endringer i skydekket.

Betydelige mengder observasjoner fra Russiske polarforsknings ekspedisjoner, institutter og data arkiv vil ble bearbeidet og publisert i dette prosjektet. Resultatene vil styrke norsk klimaforskning og spesielt ha betydning for klimamodellering og -varsling.

Dr. Alexander Chernokulsky fra A.M. Obukhov Institute for the Atmospheric Physics i Moskva er en verdens ledende forsker innen studier av skyer i Arktis. Forskningssamarbeidet vil bidra til å betydelig øke norsk tilgang til observasjons data av skyer fra hans og en rekke andre Russiske institusjoner. Videre vil det bli gjennomført studier i de fysiske egenskaper og prosesser som påvirker til skydekket i Arktis.

Forskningssamarbeidet vil gi svar på hvordan lang-tids endringer av skydekket har påvirket vekselvirkningene mellom luft-hav-land, nedbør, strålingsprosesser, tilbakeføringsmekanismer som følge av endringer i skydekke og ekstremvær hendelser. Studiene vil bli basert på analyse av lange-tidsserier av observasjoner av skytyper. Prosjektet vil kombinere analyse av historiske observasjonsdata med høy-oppløselige modell simuleringer av restrukturering av skydekke og interne mekanismer som bidrar til endring av skyene i Arktis.

Project Summary

Project final report [pdf]

Project work flow

The project will be completed in three steps:
Step 1 [completed]. Archive preparation and control by A. Chernokulsky in IAP (October-December, 2017), followed by the joint statistical analysis and description by him and I. Esau in Bergen (NERSC, 15 days in December, 2017). At this stage, the tasks A and B will be completed.


Processing of historical observational cloud data (completed)

Data analysis and sharing (completed)

Research visit of A. Chernokulsky to NERSC 3-16 December 2017 (completed)

Step 2 [completed]. Uploading the data archive to BCDC (the meta-data description and consultations), drafting the publications and preparation of PALM output data in IAP and NERSC separately (January-February 2018), followed by the joint work on the cloud type change interpretation in Bergen (NERSC, 15 – 25 days in March 2018). At this stage, the task C will be completed.


Preparing the combined climatology of the cloud cover and cloud types over the Eurasian Arctic east of the Norwegian Sea (completed, see Figures below).

Improving interpretation of the observed cloud type changes in the polar region (completed).

Combining the IAP cloud climatology and the NERSC modelling studies (completed).

Step 3 [completed]. Dissemination and publication of the project data analysis, communication to the media and submission to the impactful disciplinary journals (e.g. Nature Geosciences, Journal of Climate, International Journal of Climatology).



Exploring the wider impact of the polar cloud types and the cloud cover changes (completed).

Analysing the century long cloud climatology will be analysed in connection with the published data (completed)

Preparation and submission of the final project publication (completed; the publication is submitted to International Journal of Climatology)

Research visit of A. Chernokulsky to NERSC (completed). A. Chernokulsky worked at NERSC over 2 - 15 December 2018


Climatology of precipitationsClimatology of precipitationsChernokulsky, Alexander  A. Kozlov, F G. Zolina, O N. Bulygina, O A. Semenov, V. (2018). Climatology of Precipitation of Different Genesis in Northern Eurasia. Russian Meteorology and Hydrology, 43, 425-435. doi:10.3103/S1068373918070014.


Abstract. A method for discriminating among different types of precipitation is presented. The method is based on surface observations of precipitation, present and past weather, and the morphological types of clouds. The climatology of showery, nonshowery, and drizzle precipitation in Northern Eurasia is studied using the data of 529 Russian weather stations for the period of 1966–2014. Showery precipitation dominates in Northern Eurasia. In general, showery precipitation has greater temporal (monthly and diurnal) and spatial variability than nonshowery precipitation. The majority of showers are registered in summer (the maximum is in July), whereas the high est total monthly nonshowery precipitation is observed in autumn (the maximum is in October). The daily intensity values of showery and nonshowery precipitation are generally close, the maximum intensity is recorded in July–August. For three-hour in tervals, the shower in tensity is by 1.1–1.5 times higher. The drawbacks of the presented methodology are discussed.



Fig. 2. The climatology of different characteristics of (a) showery and compound, (b) nonshowery, and (c) drizzle precipitation rate: annual precipitation rate (colored), the frequency of days with the corresponding precipitation type (the white dash line), and the contribution to total precipitation (%) (the black dotted line). The dots mark the location of weather stations. The discrete spline interpolation was utilized.




Total Arctic cloudiness correlation with selected circulation indicesTotal Arctic cloudiness correlation with selected circulation indicesChernokulsky, Alexander A., Esau Igor (2019). Variability of the observed cloud cover and cloud types in the Eurasian Arctic in 1936–2012. International Journal of Climatology, submitted

Abstract. Over the Arctic, clouds and historical variations of the cloud cover have a significant impact on climate. A set of cloud-albedo-radiation feedbacks have been found significant for the observed changes in the region. Our understanding and our ability to project the regional effects of the global warming critically depend on the correct reproduction of the cloud cover structure in climate models. Although we have gained significant knowledge of Arctic cloudiness from satellite observations over the last four decades, several drawbacks and inconsistencies of this knowledge have been also reported. At the same time, our knowledge of the historical cloud cover, and in particular of the cloud types, as observed at meteorological stations has been limited. This study presents the cloud/cloud type climatology from the observations at meteorological stations in the Eurasian Arctic. The observational records from 86 stations have been processed for the historical period 1936-2012. This period includes the early warming, the recent cooling and the recent warming periods in the Arctic temperature change. The cloud observations from the stations confirmed that the Western Eurasian Area (WEA) is one of the cloudiest regions on Earth. The annual average cloud fraction (CF) in the WEA exceeds 0.8, where the cloud fraction in the Central Eurasian Area (CEA) and the Eastern Eurasian Area (EEA) are only 0.6 and 0.7 correspondingly. The maximum CF in all areas is observed during the extended summer months from June to October when the CF in CEA reaches 0.9. The WEA shows much larger fraction of the convective clouds (up to 0.2) in October, which is related to the cold air outbreaks over open water in that area. The convective cloud fraction has increased in all regions. The largest relative increase is found in the EEA, which is connected to the prominent autumnal sea ice retreat in the Chukchi Sea. The total cloud cover has been decreasing before 1980s; and reveals significant increasing trends since then.


Other communication and dissemination activity

Chernokulsky A., Esau I., Bulygina O., Mokhov I. (2018) Long-term variability of cloudiness over the Russian and Norwegian Arctic, Abstracts of SCAR & IASC conference POLAR-2018, presented at Davos, on 19 June 2018, poster

Chernokulsky A., Esau I. (2018) Cloud cover and cloud types in the Russian and Norwegian Arctic 1936–2013, Proceedings of 22nd International school-conference for young scientists «Atmospheric composition. Atmospheric electricity. Climatic processes. SATEP-2018», p.92, presented at Maikop, on 27 September 2018, oral.

Chernokulsky A.V., Kozlov F.A. (2018) Precipitation type redistribution toward convective rainfall increase over Northern Eurasia in 1965-2017. Research Activities in Atmospheric and Oceanic Modelling. E. Astakhova (ed.), WCRP Report № 15/2018, P.02.07–02.08.

Shikhov A.N., Chernokulsky A.V., Sprygin A.A., Azhigov I.O. Identification of mesoscale convective cloud systems with tornadoes using satellite data, Sovremennye Problemy Distantsionnogo Zondirovaniya Zemli iz Kosmosa, V. 15, 2018, [in press].

Guest project GCloudl in NANSEN NEWS 2/2018

Chernokulsky A.V., Esau I., Bulygina O.N., Mokhov I.I. (2018). Long-term variability of cloudiness over the Russian and Norwegian Arctic, 07.03.2018, UiB GFI seminar [announce]

Abstract. The study presents a long-term climatology of cloudiness over the Norwegian and Russian parts of the Arctic Ocean. The analysis is based on routine visual surface observations that conducted at island and coastal Russian and Norwegian stations from 1930s. Total and low cloud cover and fraction of different morphological cloud types are assessed. The climatology and inter-annual variability is evaluated separately for different seas (from the Norwegian to Chukchi seas) and for open-water and solid-ice regions. In general, total cloud cover (TCC) has higher intra- and inter-annual variability over SI than over OW. A decrease of TCC in the middle of the 20th century and an increase in the last few decades is revealed at individual stations and for the Atlantic sector. Long-term positive trend of convective and negative trend of stratiform cloud forms are found. Statistically significant relationship between cloudiness and sea-ice concentration is shown.

Project results

Location of Arctic stations: Location of Arctic stations used in the projectLocation of Arctic stations: Location of Arctic stations used in the project


Climatology of the Eurasian Arctic clouds (total cloudiness)


Climatology of the Arctic Eurasian cloudsClimatology of the Arctic Eurasian clouds

Changes in the total Arctic cloudines by stations and seasons

Arctic cloud climate changesArctic cloud climate changes


Total cloud cover correlation with selected circulation indices

Total Arctic cloudiness correlation with selected circulation indicesTotal Arctic cloudiness correlation with selected circulation indices

Data sets

Project data access [download] Data set is free for noncommercial use.

Dataset contains seasonally averaged information on cloud characteristics (such as cloud cover and amount of morphological type of low-level clouds) from 104 Russian and Norwegian island and coastal Arctic stations.

Zip-file includes the list of station (with the coordinates (lat-lon) and country belonging to (R)ussia or (N)orway), and four folders for each season (JFM - January-February-March, AMJ - April-May-June, JAS - July-August-September, OND - October-November-December).

Each folder contains 104 files (each for a station) with 163 rows (for each year for 1861–2013) and 11 columns.

Columns are:
total cloud cover (percentage)
frequency of clear sky (percentage)
frequency of scattered clouds (percentage)
frequency of broken clouds (percentage)
frequency of overcast (percentage)

low cloud cover (percentage)
amount of cumulus and cumulonimbus clouds (percentage)
amount of stratocumulus clouds (percentage)
amount of stratus and nimbostratus clouds (percentage)
number of observations

-9.9 stands for undefined values.

The processing of raw data is described in Chernokulsky et al. (2017) for cloud cover and in Esau and Chernokulsky (2015) for cloud morphological type amount. These papers should be cited when data are used.

Chernokulsky, A.V., Esau, I., Bulygina, O.N., Davy, R., Mokhov, I.I., Outten, S., Semenov, V.A., 2017: Climatology and interannual variability of cloudiness in the Atlantic Arctic from surface observations since the late 19th century. J. Climate, 30, 2103–2120, doi: 10.1175/JCLI-D-16-0329.1.

Esau, I.N., Chernokulsky, A.V., 2015: Convective Cloud Fields in the Atlantic Sector of the Arctic: Satellite and Ground-Based Observations. Izvestiya, Atmos. Oceanic Phys., 51, 1007–1020, doi: 10.1134/S000143381509008X.

Project Details
Funding Agency: 
Research Council of Norway
NERSC Principal Investigator: 
Igor Ezau
Coordinating Institute: 
Nansen Environmental and Remote Sensing Center
Project Status: