The winter North Atlantic Oscillation (NAO)
WIND AND ENERGY OVER EUROPE
The NAO was first named by Gilbert Walker in 1924, who defined the NAO as a negative correlation or seesaw between the air pressure measured in Azores Island and in Iceland. The concept quickly gained popularity, and more complex definitions of the NAO index were coined, using more measurements. The NAO index was shown to be correlated with the weather in far-away locations in Europe.
Nowadays, the NAO is defined using modern mathematical techniques on atmospheric reanalyses, and it is still the subject of a considerable amount of research. This ‘oscillation’ is not periodic, but highly chaotic, and corresponds to a weakening and strengthening of the atmospheric pressure difference between the polar low and the subtropical high pressure systems. Changes the NAO strongly influence the large-scale circulation, including the latitude of the jet stream, which acts as a ‘guide’ for the weather systems moving across the Atlantic. This is why the NAO has significant impacts on regional climate over Europe.
NAO: phases and its impacts
The NAO index can be positive or negative. During the positive phase, both the low pressure area near Iceland and the Azores anticyclone grow stronger, bringing stronger westerly winds to northern Europe, and mild and stormy winters. In the negative phase, the opposite happens, the westerlies are weakened and cold winds from the north and the east are more likely to affect Europe, bringing cold winters. The following are some examples of impacts of extreme phases of the NAO in the energy sector.
- Positive NAO phase windy, wet and warm increase in wind-power and hydropower generation and decrease in energy demand
- Negative NAO phase cold and dry potential for strong snowstorms increased demand but decreased wind-power and hydropower generation
- Positive NAO phase cool and dry increase in demand but increased solar PV potential.
An example: The winter of 2009-2010
The winter of 2009-2010 registered a record low value for the average NAO index (Cattioux et al., 2010). This caused anomalous cold and snowy weather in large parts of Europe and it is presented as an example of how cold extremes do still happen in a warming climate. While no public study exists quantifying the impacts of this winter in the energy sector, the effects or the cold weather in increasing the energy demand are clear. Snow and ice also caused problems in power lines. In the province of Girona, in the NE of Spain, thousands of homes and companies spent more than three days without electricity after a rare heavy snow event collapsed the area the 8th of March, causing estimated looses of more than 100 million €.
Contribution from PRIMAVERA
PRIMAVERA constitutes the first inter-comparison project between high-resolution models, up to 25 km resolution. These resolution will allow to improve the representation of the physical processes related to the NAO and its regional impacts over Europe, and lead to a better understanding of the phenomena and its evolution in future decades.
European energy impact
Heatwaves are defined as persistent periods of abnormally warm conditions. While there is no consensus about a specific definition, a typical one is to consider a heatwave as 5 or more consecutive days with temperatures above the 90th percentile of the distribution. Other definitions also involve humidity, which increases the heat stress, and the spatial extent of the event. Ideally, the user should consider different climate indexes depending on its needs.
Energy sector impacts
Whereas the most publicized impact of heatwaves are related to health and excess mortality, these extreme events also pose significant risks to the energy sector.
Future climate projections from low and medium resolution models show an increase in the severity and frequency of heatwaves over Europe (Meehl and Tebaldi,. 2004). It is well known that these events have strong socio-economic impacts. Thus, it is very important to optimize the representation of heatwave dynamics in the climate models.
- Increased air temperature leads to increased electricity demand for air conditioning, but also decreased efficiency of natural gas plants, turbines and boilers.
- Increased water temperature leads to constraints for cooling fossil fuels, geothermal, biomass and nuclear power plants, which leads to decreased efficiency and security issues.
- Increased air temperature affects power transmission, as transformers capacity decreases and lines resistance increases.
- Increased water temperature makes it subject to restrictions to its release back to sources due to environmental impact management constraints.
- Increased evaporation and drought leads to decreased hydropower generation and difficulties in the fluvial transport of raw fuel to inland power stations.
- Increased solar radiation leads to an increased potential of solar photovoltaic generation, but solar panel efficiency drops due to high temperatures.
Example: The heatwave of August 2003
During the first half of August 2003, an extremely strong heatwave affected western Europe. It was the hottest August on record in the northern hemisphere at that time.
A persistent anomalous pattern on the atmospheric circulation pumped warm air from northern Africa to the European continent. Also, the persistent high pressure (anticiclonic) conditions, trapped the hot air close to the surface. Desiccation of the upper soil and thus reduced evapotranspiration also contributed to increase the intensity and duration of the event.
The Earth Policy Institute (EPI), based in Washington DC, warns that “Though heat waves rarely are given adequate attention, Heat waves claim more lives each year than floods, tornadoes, and hurricanes combined,” warns the EPI. “Heat waves are a silent killer, mostly affecting the elderly, the very young, or the chronically ill.”
- During the last decades, European heatwaves have been found to cause serious socio-economical impacts including, among others, the energy sector.
- Currently available climate change projections show an important increase in heatwave frequency and severity for the coming decades.
- Recent studies show that high resolution models are required to properly represent the dynamics of the heatwaves.
- Thanks to their improved resolution, PRIMAVERA high resolution models will provide stakeholders an improved assessment on the evolution of heatwave risks for the coming decades.
Electricity (power system)
In response to climate change, many European countries are sourcing an increasing fraction of their electricity from renewables such as solar, wind and hydro-power. Unlike in the case of the traditional model of power system operation, whereby the output from large power stations is directly controlled to meet electricity demand, here neither demand nor supply are known in advance. The physical and economic integration of “variable” renewable generators into power networks remains a major challenge in energy policy and planning.
Power systems pose several major scientific challenges in terms of climate modelling:
Spatial localisation. Renewable generation assets (e.g., wind- and solar farms) and demand centres (e.g., cities) exist in specific geographical locations. High-resolution climate data may therefore prove valuable in accurately assessing the climate-response of individual assets (e.g., the output from a particular wind-farm).
High-frequency time dependencies. The parts of a power system with controllable output (typically coal, gas and nuclear power plant) differ greatly in cost and response time (e.g., a typical coal power station requires several hours to "switch on" from a cold start). Accurately representing local meteorological properties at short (~0.5-3h) time scales is therefore an important ingredient in simulating power system responses.
Spatial connections and compound meteorological sensitivities. Transmission infrastructure connects the power system across national and continental scales. This, coupled with the requirement for near-instantaneous supply-demand balance across the network leads to complex multi-variate meteorological sensitivities spanning large geographical areas. The spatial correlations within and between meteorological variables therefore become very important for assessing impacts on power systems.
High impact – low probability events
Extreme weather and climate conditions have profound implications for the energy sector across a wide range of technologies and energy-forms. Some examples include:
- Cold winter temperatures: icing of power lines (leading to damage) and peaks in demand (e.g., for heating and electricity, often associated with price spikes or supply shortages).
- High summer temperatures: reduction of generation efficiency, curtailment of power plants (e.g., water used for cooling may exceed environmental regulations on temperatures for river discharge), peaks in demand (for air conditioning, often associated with price spikes or supply shortages).
- Extreme precipitation: flooding of infrastructure assets
- Drought: restrictions to hydropower availability, falling river levels limiting transport of raw fuel for electricity generation (e.g., the movement of coal on the River Rhine)
- Storm surge: risk to coastal plant (particularly nuclear)
- Extreme winds: infrastructure damage (e.g., power lines, wind farms, offshore oil rigs).
Correlations between meteorological variables and across spatial scales can play a major role. For example:
- Capacity Adequacy (for electricity): this is typically the maximum residual demand for power once the contribution from renewable generation has been deducted. It is therefore strongly dependent on both temperature (as a major driver of demand) and wind/insolation (as major drivers of renewable generation).
- Trading (for gas): the demand for gas in both the East Coast of the US and Europe is strongly linked to winter temperatures. Trans-Atlantic shipment of LNG (Liquefied natural gas) is therefore strongly influenced by temperature co-variability across the two regions.
A flexible energy sector, that incorporates electricity production from broad variety of resources, including wind energy, needs to be adaptable to changing trends in resources availability. Wind power resource characterisation and response of wind power resources to climate change is an important application of climate information in this sector. More accurate planning and resource modelling can make banks more comfortable with the risk profile of e.g. offshore wind projects and as a result, banks may increase their amount of lending to the wind energy projects. This can accelerate the pace of the offshore wind development.
Road, rail, metro/subway
Road and rail transport are in general affected similarly by weather/climate. Temperature extremes affect the integrity of road surfaces and railways, with high temperatures resulting in (for example) melting of road surface and buckling of rails, and low temperatures resulting in degradation of the road surface by frost heave and cracking of rails. Heavy rainfall may lead to flooding (whether directly, as a result of surface water, or indirectly, as a result of rivers bursting their banks or groundwater levels rising to the surface) which is a common issue for road and rail infrastructure as it not only damages the infrastructure but also restricts the passage of traffic along routes. The period over which heavy rainfall occurs is an important factor with intense (usually convective) rainfall generally yielding relatively rapid surface water flooding, while moderately heavy rainfall sustained over a period of days to weeks may cause river flooding. Other climate-related hazards are: Landslides, groundwater flooding, high winds and lightning. Metro/subway systems share some similarities with surface transport, but also have their own specific issues, examples being the greater need for (and complexity of) cooling of underground infrastructure and underground drainage management.
Aviation is typically associated more with the mitigation agenda than that for adaptation – that is, the effect of aviation on climate has been more of a focus than the effect of weather/climate on aviation. However, aviation can be affected by weather and climate factors. Changes to sea level are an important consideration for certain airports which are located at the coast. Jet engine performance is affected by atmospheric parameters including temperature and humidity, meaning that take-off distances can be affected by certain weather conditions. Cold weather, ice and snow all affect operations at airports (e.g. the need to de-ice aircraft and runways). The position of the Atlantic jet stream is also important for optimizing the efficiency of routing of transatlantic flights. Atmospheric turbulence also poses a hazard for aircraft in flight.
Impact on European agriculture
Tropical cyclones are powerful low pressure systems that form over the tropical oceans, in areas with high sea surface temperatures (above 27ºC). These storms are well known for causing devastation when they pass through populated areas. European territory is seldom affected by tropical cyclones directly. However, during the September-October period, tropical cyclones formed over the North Atlantic ocean, known as hurricanes, do sometimes reach Europe just after undergoing a transition to extra-tropical storms. These storms are called post-tropical cyclones, because of their tropical origin, and bring large amounts of moisture with them, and can cause heavy rainfall over large areas, as well as hurricane-force winds.
As the September-October period coincides with the harvest, the associated extreme precipitation events may reduce crop yield and quality through waterlogging and increase crop vulnerability to pests and diseases. Also, soggy fields make it difficult for heavy machinery to harvest the crops. Trustworthy projections of post-tropical cyclones and associated heavy rainfall are important for the European agricultural sector.
An example: Post-tropical cyclone Hortensia (1984)
Hortense, or “Hortensia” in Spanish, reached hurricane status on September 25th, 475 km southwest of Bermuda. After a few days, the October 2nd, Hortensia was a weak tropical storm located west of the Azores islands. When it transitioned into an extra-tropical storm, it suddenly gained strength through a process known as explosive cyclogenesis, so a strong post-tropical cyclone suddenly developed between Azores and the Iberian Peninsula. The storm reached northwest Spain during the night between the 3-4 of October, causing winds up to 158 km/h in Monte Ventoso (Ferrol), heavy rain and coastal flooding.
Despite not having tropical cyclone characteristics, this storm was still known as hurricane Hortensia in Spain. According to the press records, Hortensia killed 6 and caused losses of 164.5M€ only in Galicia, Northwestern Spain. Of this losses, 18M€ were from the agricultural sector, which was severely affected.
The following animation shows how the cyclone suddenly deepened west of the Iberian Peninsula, hitting land with a steep pressure gradient (contours), which caused the high winds. The total columns water vapour content (blue shades), shows the moisture plume carried by the cyclone, which caused the heavy rain.
Future climate projections can help us to see if future climate will favour the occurrence of more events like Hortensia. Authors like Baatsen et al. (2015) found that, according to the EC-EARTH model, western Europe will be affected by more and more intense severe autumn storms.
Baatsen et al. (2015) looked at these cyclones over three periods: Present climate, near future climate and far future climate. They found that, as the Sea surface temperature (SST) warms, the area where tropical cyclones can form, enclosed by the 27ºC isotherm, will grow and become closer to Europe. This will in turn cause a more favourable environment for these storms to reach Europe more frequently and with more strength.
As model resolution is key to properly capture these phenomena, high resolution PRIMAVERA simulations will provide an excellent opportunity to confirm and gain insight about these storms and their future evolution.
In the following slides Baatsen et al. (2015) results are displayed: the SST and the trajectories of the cyclones simulated by an EC-EARTH ensemble for three time periods are shown.
Sea Surface Temperature for the present period (2002-2006). The 27ºC isotherm (white) encloses the area where tropical cyclones can form.
Sea Surface Temperature for the near future period (2030-2034). The 27ºC isotherm moves to the North and the East.
Sea Surface Temperature for the far future period (2094-2098). The 27ºC isotherm continues moving to the North and the East getting closer to Europe.
Simulated tropical cyclones trajectories for the present period (2002-2006).
Simulated tropical cyclones trajectories for the near future period (2030-2034). The number of post-tropical cyclones that reach Europe increases.
Simulated tropical cyclones trajectories for the far future period (2094-2098). The number of post-tropical cyclones that reach Europe continues increasing.
- Post-tropical cyclones hit Europe during the harvest season in autumn, causing serious impacts, and many of them have a tropical origin.
- Future climate projections show that warmer North Atlantic sea surface temperatures can lead to a rise of the frequency and intensity of these storms.
- High resolution climate model projections, like those of PRIMAVERA, are required in order to achieve a better understanding of these cyclones and their future evolution.
Using high resolution models to study wind storms over Europe
An example: Storm Doris
Huge damages in Europe
These events will continue to happen in future decades. Thus, their occurrence and possible changes under climate change need to be assessed with climate models.
We model the climate system using climate models. These models use grids of boxes that cover the earth’s surface and extend vertically in the atmosphere and the oceans; the models step forward in time and simulate the physical processes within these boxes.
The climate models used in the IPCC 5th Assessment Report, known as the CMIP5 models, have model grid box sizes ranging from 100—300km and are unable to capture the fine detail and extreme winds and rains associated with European wind storms, which often occur on smaller spatial scales.
The climate models for PRIMAVERA have much smaller grid boxes (typically 25km), giving much more detailed information.
- Europe is regularly struck by harmful wind storms which cause large insurance losses, disruption of livelihoods and even loss of life
- For analysing their impacts and their future evolution, climate models are required. However, currently available climate models lack resolution to properly capture these storms with enough spatial detail.
- The state-of-the-art multi-model, high-resolution PRIMAVERA project will provide new capabilities for assessing European wind storm risk in the next three decades, and support the development of next-generation climate projections and models.