Melt ponds over Arctic sea ice

By Daniela Flocco

Melt ponds develop over Arctic sea ice during the melting season from the accumulation of melt water from ice and snow. These have become increasingly important over the last few decades because they have been more prevalent and absorb much more solar energy due to their dark colour compared to the highly reflective white sea ice (Perovich et al., 2002). Where ponds form, the ice beneath becomes thinner due to increased melting. Towards the end of the summer, the air temperature drops and a thin layer of ice forms over melt ponds. The ponds’ melt water trapped in the ice acts as a heat store and does not allow the underlying ice to start thickening until all the pond’s water is frozen. Ponds are up to 1.5 m deep and it can take over two months to freeze their volume of water. Considering that ponds cover up to 50% of the sea ice extent their impact cannot be neglected (Flocco et al., 2015).

Credit: Donald Perovich

A strong negative correlation exists between the change in successive mean winter ice thicknesses and the length of the intervening melt season, suggesting that summer melt processes play a dominant role in determining mean Arctic sea ice thickness for the following winter (Laxon et al., 2003). Another indication of the importance of melt ponds in explaining thinning of sea ice is that melt ponds are present in the Arctic more than in the Antarctic, where the sea ice thinning is less striking.

Ponds are rather irregular in shape but occur at a higher percentage over thin young ice: since the area of young ice is increasing (relatively to the total amount of ice which is instead decreasing), the impact of melt ponds will also become increasingly important. This will lead to a positive feedback effect in which thin ice will start thickening later in winter and will possibly be a preferential area for the formation of melt ponds in the following spring. Furthermore, corresponding to where melt ponds form, specular lenses of fresh water form under the sea ice cover, impacting the freezing point of water at the ice–ocean interface. At the beginning of the season sea ice is impermeable, so once ponds form they can be above sea level. When they start melting the ice, it becomes more permeable and when the ponds are fully developed they are in hydrostatic balance with the ocean so they drain to sea level.

Schemes handling melt ponds have only recently been included in global circulation models and are rather crude: the melt water was assumed to be flushed into the ocean without dwelling on the sea ice. Recent studies have shown that the lack of a melt pond parameterization can give an overestimation of sea ice thickness of up to 40% during summer (Flocco et al 2010, 2012). Model results have shown a good ability to forecast the minimum September ice extent, relating it to the melt pond area calculated by the model in May (Schröder et al 2014). This is one demonstration of how we have used the principles of physics to understand the changes we have observed in the cryosphere.


Flocco, D., D. L. Feltham, and A. K. Turner, 2010. Incorporation of a physically based melt pond scheme into the sea ice component of a climate model. J. Geophys. Res., 115, C08012, doi:10.1029/2009JC005568.

Flocco, D., D. Schröder, D. L. Feltham, and E. C. Hunke, 2012. Impact of melt ponds on Arctic sea ice simulations from 1990 to 2007. J. Geophys. Res., doi:10.1029/2012JC008195.

Flocco, D., D. L. Feltham, E. Bailey, and D. Schröder, 2015. The refreezing of melt ponds on Arctic sea ice. J. Geophys. Res. Oceans, 120, 647–659

Laxon, S., N. Peacock and D. Smith, 2003. High interannual variability of sea-ice thickness in the Arctic region. Nature, (425) October 30, 947-950.

Perovich, D.K., W.B. Tucker III, and K.A. Ligett, 2002. Aerial observations of the evolution of ice surface conditions during summer, J. Geophys. Res., 107 (C10), 8048, doi:10.1029/2000JC000449.

Schröder D., D. L. Feltham, D. Flocco, M. Tsamados, 2014. September Arctic sea-ice minimum predicted by spring melt-pond fraction. Nature Clim. Change, DOI: 10.1038/NCLIMATE2203.

Posted in Climate, Climate modelling, Cryosphere, Polar | Tagged | Leave a comment

Observation uncertainty in data assimilation

By Sarah Dance

Approximately 4 million properties in the UK are at risk from surface-water flooding which occurs when heavy rainfall overwhelms the drainage capacity of the local area. Several national weather centres have been developing new numerical forecasting systems to improve prediction of such events.  Weather forecasts are based on the output from a numerical computer model. Such models are built from a mathematical description of physical laws that govern the behaviour of the atmosphere and evolve an estimate of the current state of the system forward in time.  The estimate of the current state of the system may be obtained by a sophisticated mathematical blending of information from previous forecasts with recent observations in a process known as data assimilation.

Figure 1. Forecast-Assimilation Cycle

Remote sensing and data assimilation
In data assimilation we compare model forecast predictions and observations and adjust the model state so that it is closer to the observations, bearing in mind the uncertainty in the observations. However, the quantities predicted by the model are not usually the same as those being observed by the operational observation network. For example, in weather forecasting the model may predict wind, temperature, pressure and humidity. A weather radar on the other hand sends out pulses of electromagnetic waves and measures the intensity of the returned signal as the waves bounce off raindrops in the atmosphere. Thus we require a mathematical model that describes the physical relationship between the predicted quantities and the observations.  In data assimilation, this mathematical model is often termed the observation operator.

Observation uncertainty
When we compare the model predictions to the observations using the observation operator we are typically left with a residual known as the observation uncertainty. This image shows a measure of observation uncertainty for observations from the SEVIRI instrument used in numerical weather prediction over the UK. SEVIRI is a satellite-borne instrument measuring quantities sensitive to surface temperature (channels 7, 9 and 10) and upper level water vapour (channels 5 and 6) – see Figure 2. Channels 7, 9 and 10 are not used over land (the white areas in the picture) and the high uncertainty values shown around the coastline are due to a problem in the operational system quality control. Properly accounting for observation uncertainty in data assimilation is an important research topic. Our research has already led to improvements in operational weather forecast skill for global forecasting. Experiments are currently underway to find out if further improvements can be made for local forecasting of intense rainfall events.

Figure 2. Output from SEVIRI by channel


Waller, J. A., Ballard, S. P., Dance, S. L., Kelly, G., Nichols, N. K. and Simonin, D., 2016. Diagnosing horizontal and inter-channel observation error correlations for SEVIRI observations using observation-minus-background and observation-minus-analysis statistics. Remote Sensing, 8 (7). 581. ISSN 2072-4292
doi: 10.3390/rs8070581 

Posted in data assimilation, earth observation, Numerical modelling, Remote sensing, Weather, Weather forecasting | Leave a comment

Confessions of an Admissions Tutor

By Hilary Weller

I am a postgraduate admissions tutor, so I see a lot of applications for PhD positions and I do a lot of interviewing. I would like to share some tips for applicants for PhD and post-doc positions and also some tips for interviewers. I have also done lots of interviewing for post-doc positions.

Your CV
If you are applying for a PhD position in a similar topic to your Bachelor’s or Master’s degree, then I really want to know how well you did in your degree – all the details – all the courses you did and the marks that you got in them. What did you do for your project? What mark did you get for your project? Did you do any relevant summer work? Include any other summer jobs or part time jobs – they tell me if you are hard working.

A CV for any position should document all periods of employment, unemployment and career breaks – even if you don’t want to tell me what the career break was for! I have two periods of maternity leave on my CV 🙂 which prompted positive comments from reviewers for my application for a fellowship. If you try to hide a career break I would be suspicious, but if you said “Jan 2014-Jul 2014: career break”, I would certainly be curious but I would be wary that I might be treading in a sensitive area. If you have had periods of unemployment, think of something useful that you did while unemployed. For example, did you read any relevant books, do any computer programming, spend time on any relevant hobbies or volunteering?

If you are asked for a personal statement I have some advice – tailor it to the job or PhD position that you are applying for, considering your aspirations as well as your experience, stick to the length limit, proof read it, then get someone else to proof read it.

Some very obvious interview questions to prepare for

  • Why are you interested in doing _THIS_ job/PhD?
  • What makes you well suited for _THIS_ job/PhD?
  • Tell us about a project that you have done?
  • Do you have any questions for us?

Some less obvious interview questions and tips for interviewers
For post-doc and PhD positions, it is important for interviewer and interviewee to find common ground. The obvious question “tell us about a project you have done” can leave the interviewer feeling uncritically impressed by the candidate. Conversely, questions along the lines of “what do you know about this aspect of my specific research area” can be unfair – a weak applicant with experience in the area could shine brighter than a strong applicant. So it is important to find areas of common interest. This requires preparation by the interviewer – you have their application and their academic transcripts – pick on a topic that you know about. Also, pick up on topics that the interviewee brings up and ask follow-up questions that you know the answer to. And interviewees – expect to be able to talk about anything mentioned in your application or transcripts! If you think that you did a relevant degree, you should be able to remember lots of what you learned.

An essential question for interviewing for a post-doc position – describe a paper that you have read recently.

Finally, remember that interviews are about candidates assessing the positions as well as vice-versa. So interviewers should be friendly and encouraging and offer plenty of information. Don’t spend too long on questions that the applicant is struggling with, move on with a smile rather than a shrug.

Posted in Academia, Teaching & Learning, University of Reading | Leave a comment

An update on the North Atlantic cold blob (January 2017)

by Pablo Ortega

One of the most remarkable climate events in the last two years has been an exceptional cooling in the eastern sub-polar North Atlantic (ESPNA, Figure 1), commonly referred to as “the cold blob”. Occurring while the planet experienced the warmest temperatures on record, this somewhat surprising cold anomaly has stirred considerable attention on the media (e.g. The Guardian, The Daily Mail, The Washington Post), as well as great interest among the scientific community.

Figure  1: Mean 2015-2016 sea surface temperature anomaly with respect to the period 1900-2016.

The first efforts focused on understanding its origin. Transient climate simulations show a “warming hole” anomaly in the same region of the cold-blob, associated with long-term decreases in the strength of the Atlantic Meridional Overturning circulation (AMOC) (Drijfhout et al., 2012; Rahmstorf et al., 2015). Additionally, analysis of long control simulations show that similar ESPNA coolings can emerge naturally, due to internal decadal fluctuations in the North Atlantic (Ortega et al., 2016; Robson et al., 2016). These cooling events are caused by weakenings of the northward ocean heat transport, following previous decreases in the AMOC strength. Likewise, analogous warming episodes tend to appear in response to AMOC strengthenings. A potential negative feedback between the AMOC and the NAO was identified in Ortega et al. (2016) and could explain these trend reversals. The analysis in Robson et al. (2016) also identifies deep Labrador Sea densities as a key proxy of the AMOC changes (for which only limited observations are available), thus extremely useful to investigate the chain of events that likely led to the observed cold blob. These are summarized in Figure 2.

Figure  2: Evolution of the anomalous NAO, deep Labrador Sea densities (averaged between 1000-2500 m) and the top 700 m mean temperature in the ESPNA. Deep Labrador Sea densities are here used as an indicator of the changes in the AMOC strength. Anomalies refer to the period 1961-1990.

During the 1980s and early 1990s, the North Atlantic Oscillation (NAO) was predominantly positive. Associated with this, strong and persistent heat fluxes over the Labrador region enhanced deep-water formation and led to a maximum of the AMOC strength in 1995, inducing a subsequent warming of the ESPNA. From 1995 to 2010, the AMOC experienced a strong decrease, mostly explained by a concomitant tendency towards more negative NAO phases. This AMOC weakening reversed the warming trend over the ESPNA, and gave rise to the record-low temperature anomalies observed in 2015 and 2016. Thus, evidence suggests that the AMOC weakening responsible for the cold blob was internally driven, and not triggered by long-term changes in the anthropogenic forcings. We cannot rule out, however, a contribution of the radiative forcings (including anthropogenic and volcanic aerosols, and solar irradiance) to these decadal changes in the North Atlantic, e.g., through a modulation of the NAO phases.

The latest observations point to a likely reversal of the cold blob in the next few years. Positive NAO phases have been predominant since 2012, and Labrador Sea densities suggest that AMOC strength has been increasing since 2014. There is even evidence of a relative ESPNA warming starting last summer. Yet, it is too soon to determine whether these changes will be sustained long enough to reverse the cold blob completely. Indeed, to this date, the cold ESPNA anomaly has only slightly weakened with respect to the previous two years (Figure 3) – a small change that could be consistent with climate noise.

Figure  3 (Left) Top 700 m mean temperature anomaly (T700, in °C) in 2015-2016. (Right) T700 mean temperature anomaly in December 2016-January 2017. Anomalies refer to the period 1961-1990.

For further reading on the cold blob and its causes, there is a special issue on US CLIVAR Variations.

Drijfhout, S., G. J. van Oldenborgh, and A. Cimatoribus, 2012. Is a Decline of AMOC Causing the Warming Hole above the North Atlantic in Observed and Modeled Warming Patterns?, J Clim 25, 8373–8379.

Ortega, P., J. I. Robson, R. T. Sutton, and A. Martins, 2016. Mechanisms of decadal variability in the Labrador Sea and the wider North Atlantic in a high-resolution climate model, Clim Dyn, Published Online.

Rahmstorf, S., J. E. Box, G. Feulner, M. E. Mann, A. Robinson, S. Rutherford, and E. J. Schaffernicht, 2015. Exceptional twentieth-century slowdown in Atlantic Ocean overturning circulation, Nat Clim Chang 5, 475–480.

Robson, J., P. Ortega, and R. Sutton, 2016. A reversal of climatic trends in the North Atlantic since 2005, Nat Geosci 9, 513–517.

Posted in Climate, Climate modelling, Numerical modelling, Oceans | Tagged | Leave a comment

HEPEX: a community of research and practice to advance hydrologic ensemble prediction

By Hannah Cloke

Although formal funded societies and projects can be very important in advancing research and improving how science is used, the unfunded voluntary community initiative of HEPEX has been one of the most important networks that I have been involved in during my career so far. HEPEX (which stands for Hydrologic Ensemble Prediction Experiment) began in 2004 just as I took up my first post as a University Lecturer. HEPEX aims to advance the science and practice of hydrological ensemble prediction and how it is used for risk-based decision making.

Participation in HEPEX is open to anyone wishing to contribute to its objectives, and so the HEPEX community thrives through organising scientific workshops and sessions at major conferences (such as the European Geosciences Union General Assembly every Spring), coordinating joint experiments, highlighting best practice in hydrologic ensemble prediction systems to help practitioners find out how ensemble prediction is being used around the world in different applications (such as for hydropower or flood forecasting), and through our online community interaction including webinars and blog discussions (; @hepexorg).  The HEPEX community are also very keen to develop serious games to help communicate best practice and to understand how we can improve forecast communication (Arnal et al, 2016).

It is not always easy to explain what you work on, especially when you have to avoid using jargon specific to your field. Yet, this is something that we all have to do. It is important to be able to explain your research simply in order to communicate effectively with scientists in other fields and, for example, businesses, policy makers and the public.  This week in HEPEX we have been thinking about this with the help of a little competition: using only the 200 most commonly used words of the English dictionary, explain “Ensemble hydrological forecasting”. Please consider having a try, you could win yourself a special mystery prize!

The next HEPEX meeting will be in Melbourne in February 2018 in the height of the gorgeous warm Australian summer. The theme for the workshop is ‘breaking the barriers’ to highlight current challenges facing ensemble forecasting researchers and practitioners and how they can (and have!) been overcome.  How can you resist such a tempting offer?

Want to know more? Want to join our community?

HEPEX website:

HEPEX twitter: @hepexorg

Arnal, L., Ramos, M.-H., Coughlan de Perez, E., Cloke, H. L., Stephens, E., Wetterhall, F., van Andel, S. J., and Pappenberger, F., 2016. Willingness-to-pay for a probabilistic flood forecast: a risk-based decision-making game, Hydrol. Earth Syst. Sci., 20, 3109-3128, doi:10.5194/hess-20-3109-2016.

Posted in Hydrology, Numerical modelling | Tagged | Leave a comment

Mountain waves, ship waves and duck waves

By Miguel Teixeira

There is a striking resemblance between some waves generated in the atmosphere in flow over isolated mountains and wave patterns in the wakes of ships, boats, or even ducks swimming in a pond. Typically, these waves are delimited by a triangular wake with a well-defined angle (of about 39 degrees) opening downstream of the obstacle that generates them (see Figures 1, 2 and 3). The shape of this wake was first determined theoretically by Lord Kelvin in his pioneering study of ship waves (Thomson, 1887). Ship waves and “duck waves” are forced via direct piercing of the air-water interface by the body that acts as a wave source.

Figure 1: Waves in a cloud layer generated in flow over the South Sandwich islands (source:


Figure 2: Waves produced in the wake of a boat in a river (source:

Figure 3: Waves generated by a swimming duck (source:

In flow over mountains, the source obstacle typically does not touch the deformed interface. Rather, waves are generated remotely, in the same way as surface distortions in a stream are caused by stones at its bottom. Additionally the density gradient where the waves propagate is not as sharp as an air-water interface, typically corresponding to a temperature inversion (Teixeira et al., 2013). Nevertheless, all of these waves can be approximated as interfacial waves where gravity is the restoring force, and the angle, as well as the spatial structure, of the wake, can be explained using linear surface gravity wave theory (e.g. Lighthill, 1978, chapter 3). The resistance force produced by these waves influences the design and powering characteristics of ships, and has gravity wave drag as its atmospheric counterpart.

Typically, the waves depicted in Figures 1-3 are “deep-water waves”, with the wavelength not exceeding the fluid depth. This means that they are dispersive, and their group speed (at which energy is transported) is half the phase speed (at which individual crests and troughs travel). Waves within a wake are stationary with respect to their source, so they must propagate at an angle such that the projection of the mean fluid speed matches their phase speed. Therefore, they cannot be faster than the flow. The interface is only appreciably deformed in the region (defining the wake) where the energy of waves emitted at a given point (and propagating with the group speed) has travelled a distance half that of the crests that are able to keep up with the source (red dots in Figure 4).


Figure 4: Schematic diagram explaining the angle of Kelvin’s ship wake (source:

No wave energy can propagate upstream of a line joining point B in Figure 4 (where the source is located) to a point where this line is tangent to the circle defined by the red dots. The angle between this line and the direction of motion is half the Kelvin ship wake angle, arcsin(1/3) ~ 19.5º. This result only relies on the facts that the wave pattern is stationary and the group speed of the waves is half their phase speed, hence its general applicability to waves that are apparently so different. A distinct speed ratio would produce a different wake angle.

The wave crests near the middle of the wake are perpendicular to the flow and called transverse waves, while those at the edge of the wake make a smaller angle with the flow direction and are known as divergent waves. The latter are shorter than the former (see Figures 2 and 3), since they are geometrically constrained to have a lower phase speed, and the dispersion relation of deep-water gravity waves prescribes that shorter waves propagate more slowly than longer ones.


Lighthill, M. J., 1978. Waves in Fluids, Cambridge University Press, 504 pp.

Teixeira, M.A.C., Argain, J.L. and Miranda, P.M.A., 2013. Orographic drag associated with lee waves trapped at an inversion. Journal of the Atmospheric Sciences, 70, 2930-2947. Centaur listing 

Thomson, W., 1887. On ship waves. Proceedings of the Institution of Mechanical Engineers, 38, 409-434.


Posted in Boundary layer, Waves | Leave a comment

THE BRAVE PROJECT – Annual meeting, January 2017, Ghana

By Galine Yanon – Walker Institute

The overall objective of the BRAVE project is to quantify the impacts of climatic variability and change on groundwater supplies from low storage aquifers in Africa.

More than 40 institutions from Burkina Faso, Ghana and UK attended the BRAVE Project annual general meeting, held in Accra, Ghana, 24-26 January 2017. These institutions are direct and indirect partners of the Project. The meeting commenced with a project meeting and discussions around work packages, from WP1 to WP5.

After the opening session and presentation of the Agenda, Professor Rosalind Cornforth, PI of the project and Director of the Walker Institute, presented an overview of the UpGro consortium (composed of 5 projects) and the BRAVE Project in order to give a better understanding of the ongoing work to the participants. This overview was really important because some partners were not directly engaged in the different activities of the project.

One of the objectives of BRAVE is the capacity building of early career researchers and the benefit of the project for the communities engaged through in-country partners (CARE International in Ghana, Christian-Aid and Reseau Marp in Burkina Faso). So, after the overview, discussions focused on the different work packages of the Project:

  • WP1 presented by Dr Henny Osbahr, into the Understanding of vulnerability in the communities;
  • WP2 presented by Dr Galine Yanon, into the Understanding of Decision-making Pathways, Governance Structures and Institutional Influence;
  • WP3 and WP4, presented by David Macdonald, into the Improvement of our understanding of the hydroclimate and strategic planning and adaptive capacity; and
  • WP5 by Professor Rosalind Cornforth, Delivering Evidence and Demonstrating Resilience.

The Program Coordination Group (PCG) across all project of the Consortium was also presented to participants.

Further information on later stages in the meeting and its outcomes can be found in Galine’s reports on the Walker Institute website.

Posted in Africa, Conferences, Hydrology, Teaching & Learning | Tagged | Leave a comment

Measuring radiation with aircraft

By Peter Hill

In my career as an atmospheric scientist I’ve relied on observational data from a wide range of sources including satellite imagery, surface measurements, ground-based and satellite based radar, and aircraft measurements. Last July I had my first opportunity to contribute to the available data when I took part in the aircraft field campaign for the EU-funded DACCIWA (Dynamics-Aerosols-Chemistry-Cloud Interaction in West Africa) project.

The DACCIWA project is investigating pollution in southern West Africa (SWA) and how this affects health and the regional climate. This region is very reliant on agriculture which is highly sensitive to the amount of rainfall. Any changes in rainfall due to pollution may have important implications for the worlds’ supply of cocoa, not to mention the livelihoods of millions of people in SWA.

My role in DACCIWA is focused on atmospheric radiation; how sunlight and thermal radiation interact with the atmosphere over SWA. Radiation is important because it is a key component of the atmospheric energy budget (see Figure 1). Consequently radiation changes can lead to circulation changes which may in turn affect more obviously societally relevant processes such as precipitation.2017 01 19 Peter Hill Fig 1Figure 1: Key terms in the atmospheric energy budget for southern West Africa (defined here as 8°W – 8°E and 5 – 10°N). Units are W m-2. Values shown are June-July means for 2000-2015. Divergence of dry static energy and sensible heating are derived from ERA-Interim. Radiation values (i.e. shortwave SW heating and longwave LW cooling) are from the CERES-EBAF dataset and latent heating is from the TRMM dataset. Adapted from Hill et al, 2016.

Pollution particles (aerosols) reflect and absorb radiation directly, but may also affect radiation by changing cloud properties. Radiation measurements are important to understand the extent to which both occur. These measurements can also be used as an additional check on aerosol and cloud measurements made during the campaign by both the aircraft and by satellites. Radiative transfer is a relatively well understood process. If the measured cloud and aerosol properties are correct, we should be able to predict the measured radiation quite accurately using computer-based models.

The campaign involved three aircraft, two of which were equipped with instruments to measure radiation (in addition to many other instruments). Each had two pyranometers, which measure solar radiation, and two pyrgeometers, which measure thermal radiation. One of each was mounted above the aircraft pointing upwards to measure downwelling radiation and one of each was mounted below the aircraft pointing downwards to measure upwelling radiation. During the campaign, out of a total of 50 flights, seven were made with the primary objective of making radiation measurements. I was lucky enough to fly on the British Antarctic Survey Twin Otter (Figures 2 and 3) on one such flight, which was an exhilarating and surprisingly comfortable experience.

2017 01 19 Peter Hill Fig 2

Figure 2: The British Antarctic Survey Twin Otter aircraft outside the hangar at Gnassingbé Eyadéma airport in Lomé, Togo during the DACCIWA aircraft field campaign.

2017 01 19 Peter Hill Fig 3

Figure 3: Airborne view of Lomé from the Twin Otter aircraft. Note the haze due to pollution.

Together with observations from three highly-instrumented field sites, the aircraft campaign has provided a wealth of measurements. These measurements provide an indispensable dataset for understanding pollution, weather, and climate in this region. The measurements will facilitate lots of exciting scientific research and scientists across Europe and SWA will be working with this dataset for at least the next two years. Keep up to date with our progress via the DACCIWA newsletter, or twitter feed.


Hill, P. et al, 2016. A multisatellite climatology of clouds, radiation, and precipitation in southern West Africa and comparison to climate models  –, where further details can be found.

Posted in Aerosols, Africa, Atmospheric chemistry, Climate, Climate change, Climate modelling | Tagged | Leave a comment

Childhood white Christmases: nostalgia or reality?

By Inna Polichtchouk

Nearly every Christmas, I travel back to Finland in the hope of celebrating Christmas Eve in the well below freezing temperatures surrounded by a plethora of snow. My childhood memory of this magical day begins with a cross-country skiing trip in the forest amongst frozen sparkling trees, with low mid-day sun gently thawing the icicles formed on eyelashes by my sublimed breath. The skiing trip is followed by a heart and bone warming sauna and a plunge into a soft snowdrift to even out the body temperature. However, over the past decade this nostalgic childhood memory of a white Christmas has begun to fade and be replaced by a new one where a skiing trip is now a mere walk in the rain and the snowdrift a mud puddle. Is this childhood memory real? I decided to investigate.

Figure 1 shows snow depth at Helsinki-Vantaa airport weather station, averaged over Christmas Eve and Christmas day from 1960 onwards. Having lived in and around the capital area, this weather station is chosen to refresh/test my memory. Other weather stations in and around Helsinki show similar measurements. The snowless Christmases are circled. To be consistent with the Finnish Meteorological Institute (FMI) definition, “snowless” is defined as snow depth below 1 cm. It is clear that out of the 16 snowless Christmases shown, 50% have occurred since the turn of the millennium and following the first 15 years of my life. Before 2000, I only lived through three snowless Christmases. Since moving out of Helsinki in 2005, a snowless Christmas appears to be more of a norm than an anomaly. In particular, the 21st century snowless Christmases have mostly been wet and warm as seen in Figure 2.  Given the data, I hence preserve the right to claim that my childhood was filled with white Christmases.

2017 01 12 Inna Polichtchouk Figure 1

Figure 1. Snow depth (cm) at Helsinki-Vantaa airport weather station, averaged over 24-25 December. Snowless Christmases are circled. “Snowless” is defined as having snow depth below 1 cm. Weather station data is available from Finnish Meteorological Institute (FMI) and also at


2017 01 12 Inna Polichtchouk Figure 2a

2017 01 12 Inna Polichtchouk Figure 2b

Figure 2. (upper) Daily accumulated precipitation (mm) and (lower) Mean daily temperature (°C) at Helsinki-Vantaa airport weather station, on 24 December. Snowless Christmases are circled.

Does the lack of snow at Christmas in Helsinki reflect the lack of precipitation or is it just a harsh reality of a warming climate? Figure 3 shows the temperature and precipitation anomalies for December of all the years with snowless Christmas. Indeed, it appears that the snowless Christmases are mainly due to anomalously warm December temperatures.

2017 01 12 Inna Polichtchouk Figure 3a2017 01 12 Inna Polichtchouk Figure 3b Figure 3. December anomalies (from 1959-2015 mean for the Helsinki-Vantaa airport weather station) of (upper) Precipitation (%) and (lower) mean daily temperature (°C]. Only years with snowless Christmases are shown.

This trend for snowless Christmases in Helsinki is, of course, likely a manifestation of internal variability. However, I am seriously contemplating celebrating Christmas a month later. Since 1960, the average daily January temperatures have been below freezing and even the smallest amount of January total precipitation of 8.1 mm (in 1972) would produce at least 8 cm of snow cover* – enough to recreate that childhood memory of a white Christmas!

* An approximate rule of thumb is that 1 mm of rainfall produces 1 cm of snow at near zero temperature.

Posted in Climate | Leave a comment