NACLIM in short
Our partners in Europe

§ North Atlantic Climate
FP7 Collaborative Project

§ Duration: 51  Months + final reporting, Oct. 2012 – Jan. 2017 (Mar. 2017)

§ Research Focus: Assessment of decadal climate forecasts

§ Partner: 18 participating institutions (+1 Third party) from 10 European countries

  • 4 universities
  • 12  research institutions
  • 2 national operational institutes

§ 5 Core themes, 12 work packages

§ Total project volume: 11 Mio. Euro

EU contribution: 8.6 Mio M€

Member of the European Climate Observations, Modelling and Services (ECOMS) Initiative
(Opens external link in new windowwebsite)

What is it about?

The North Atlantic Ocean is one of the most important drivers for the global ocean circulation and its variability on time scales beyond inter-annual. Global climate variability is to a large extent triggered by changes in the North Atlantic sea surface state. The quality and skill of climate predictions depends crucially on a good knowledge of the northern sea surface temperatures (SST) and sea ice distributions. On a regional scale, these parameters strongly impact on weather and climate in Europe, determining precipitation patterns and strengths, as well as changes in temperature and wind patterns. Knowledge of these factors, and of their development in the years to come, is of paramount importance for society and key economic sectors, which have to base their planning and decisions on robust climate information. NACLIM will contribute to this goal.

Understanding global climate change
Natural variability and human induced trends

The difference between weather and climate is the timescale: weather is the atmospheric conditions[1] over a short period of time (minutes to months), while climate is typically an average of the weather conditions taken over 30 years in a particular region.

Consequently when scientist talk about climate change they are referring to changes in the long-term averages of the daily weather patterns (Gutro, 2005).

Climate has a natural pace of variability, shorter-term variations, which can be deciphered amidst longer-term trends. Natural climate variability can be the result of forcing’s external to the Earth, such as changes in solar radiation, causing extreme climatic events. Forcing’s internal to the Earth’s system, (such as volcanic eruptions and fluctuations in the oceanic circulation), also cause natural climatic variability (Zachos et al., 2001). Internal variability occurs because different aspects of the climate system have non-linear interactions and respond to changes at different speeds. Phenomena such as El Nino and La Nina are examples of internal variability of the Earth’s climate system where the oceans and atmosphere respond at different speeds changes in temperature and pressure resulting in natural variations in the climate (Solomon et al., 2007).

Superimposed on top of natural climate variability are human induced trends, known as anthropogenic or man made changes. Anthropogenic climate change is predominantly driven by man made emissions of carbon dioxide (CO2) into the Earth’s atmosphere (Hofmann et al., 2011). Carbon dioxide is a greenhouse gas. Greenhouse gasses are able to trap heat reflected from the Earth’s surface, thereby warming the atmosphere. Greenhouse gasses occur naturally in the atmosphere, and without them the planet would be significantly cooler (EPA, 2013). Since the industrial revolution humans have been releasing additional sources of CO2 into the atmosphere from the burning of fossil fuels. This is causing CO2 concentrations in the atmosphere to increase at a rate much faster than at any other point in human history (Keeling et al., 2005). [2]

Increasing anthropogenic CO2 concentrations is warming the temperature of the atmosphere, causing a climatic feedback. With sufficient warming a tipping point may be reached, where the Earth’s climate moves abruptly between different states (e.g. changes between ice and non ice age conditions; (Jenkins, 2013)).

 


[1] Temperature, humidity, precipitation, cloudiness, brightness, visibility, wind and atmospheric pressure

[2] Historically natural climate variability typically causes atmospheric CO2 concentrations to vary between 200 ppm (parts per million; during ice ages) and 300 ppm (during warmer periods between ice ages; Fischer et al. 1999). At the start of the industrial revolution atmospheric CO2 concentrations were 280 ppm (Solomon, et al. 2007). Today the current atmospheric CO2 concentration is ~400 ppm Scripps Institute of Oceanograpy, 2013).

 

What is the role of the oceans in climate change?

The oceans play a role in controlling global climatic tipping points, through ocean circulation distributing heat around the planet. In the North Atlantic, ocean currents originating in the tropics bring warm, salty water northwards into the Arctic. This is known as the Gulf Stream. Once in the high latitudes of the Arctic Ocean this warm water cools, increasing in density, and sinks into the deeper parts of the ocean. This process of deep-water formation drives much of the oceans global circulation patterns and is known as thermohaline circulation (Rahmstorf, 2006). As global temperatures increase Arctic ice melts, increasing the freshwater input to the ocean and lowering the overall density of the Arctic surface ocean waters. If density is reduced sufficiently, deep water formation may slow down or cease altogether, altering global ocean circulation patterns and heat distribution (Liu et al., 2013, Broecker, 1997), ultimately impacting global climate (Jenkins, 2013).

The input of freshwater from ice melt in the high latitudes into the oceans, which may ultimately disrupt the thermohaline circulation, is one example of a feedback and tipping point in the Earth’s climate system.

What are climate models and how do they work?

Climate models are mathematical representations of the interactions between different aspects of the Earth’s system. At a basic level, these are interactions between the land, ocean, atmosphere and cryosphere (i.e. ice cover), which are spilt into different conceptual boxes. Simple models may have only a few boxes, for example only oceans, atmosphere and land. However model complexity can be built up and the most complicated models can have thousands of boxes representing different aspects of the Earth’s climatic system (WMO, 2013). The higher the complexity of the model, i.e. the more boxes and interactions being studied, the more computer power is needed to run the models. Interactions between the model boxes (i.e. movements of air, energy and water) are of the most interest to climate modellers. Interactions are defined mathematically based on well-known principles, which come from an understanding of how the Earth’s climate system works today and in the past (National Institute of Sciences, 2011). In order to be able to model the climate successfully and forecast future changes with any confidence, an understanding of the past, and observations of the current state of the climate are both essential. Without this information climate modellers are unable to constrain model outputs and improve modelling and predictive capacity[1].

 

For any model to be of use its accuracy must be tested through simulation of past climatic conditions and comparison with model outputs. Currently, models can capture many important large scale, global features well, such as the distribution of land surface temperatures, precipitation, winds, ocean temperature and sea ice cover, on smaller scales (e.g. regional) different models often do not agree. Scientists are continually refining models as scientific knowledge increases, in order to better refine these smaller-scale processes. Despite this, a key agreement between all models is that those that only account for natural climate variability (i.e. do not include any anthropogenic forcing’s) are not able to explain the recent observed warming of the climate. However, when anthropogenic CO2 emissions are also included in models the output results and observations of increasing global temperatures match (National Institute of Sciences, 2011).



[1] For a more detailed understanding of climate models in relation to the ocean and how they work please refer to Griffies, 2004

What is "regional downscaling"?

New techniques are being used to improve model outputs on smaller regional scales. These techniques are known as regional downscaling (Wilby and Fowler, 2010, Castro et al., 2005). Regional downscaling is important for providing projections, which encompass greater detail and representation of localized extreme events (Pielke, 2012). Ultimately regional climate downscaling is essential if end users want to be able to use climate model data to inform local policy and decisions.

What's the purpose of using climate models?

Human behaviors and the choices that society makes (e.g. how much fossil fuels are burnt), ultimately affects the way the climate system will function. Climate modellers do not attempt to predict these human behaviors, but use different scenarios to produce predictions of future greenhouse gas emissions and CO2 concentrations. Each scenario includes a different assumption about future human factors (population growth, economic activity, energy conservation, technology, land use etc.). These predictions of future atmospheric CO2, based on different scenarios are input into the climate models (National Institute of Sciences, 2011).

The Intergovernmental Panel on Climate Change (IPCC) has constructed a range of plausible scenarios that are widely used in climate models to predict future climatic changes.

The model outputs can be used to inform policy makers about appropriate climate change mitigation or adaptation strategies (Nakicenovic and Swart, 2000):

  • Mitigation refers to reducing the scale of climate change through reductions in greenhouse gas emissions
  • Adaptation refers to coping with and taking advantage of climate change (National Institute of Sciences, 2011).
Which timescales for climate predictions?

Models are used to predict climate over various different timescales in the future. Within the scope of the NACLIM projects seasonal to decadal timescales are those of particular interest.

  • Seasonal prediction covers timescales month-years, with information presented as seasonal or monthly means.
  • Decadal model predictions are more experimental and are run only 10 years into the future with information presented as annual to decadal averages.

The climatic processes affecting decadal and seasonal variability are known to be different. For instance, the El Nino Southern Oscillation is the dominant driver of seasonal variability, whilst the main drivers of decadal climate variability are much more affected by changes in atmospheric composition (e.g. anthropogenic CO2 concentrations). Consequently, seasonal and decadal timescales can inform you about different aspects of the climate (Goddard et al., 2012).

Urbanization and local urban climates

In 1800 only 3% of the global population lived in urban centers. Today more than half of the global population lives in urban areas and this is predicted to rise to 77% by 2050 (PRB, 2013, WHO, 2013). Consequently understanding how climate change will affect the urban environment is critical. Urbanization of the landscape from rural land use is the most significant driver of anthropogenic climate change (Oke, 1997). Urbanization can radically change the local climate as evaporation, surface water run off, and albedo, i.e. how much heat the ground reflects or absorbs, are altered (Blake et al., 2011).

Albedo is particularly important in climate science for determining the temperature of a surface. Any surface that has a high albedo (e.g. is light coloured or shiny) is highly reflective and therefore does not absorb radiation from the sun easily. It is a more reflective surface and therefore remains cooler. Any surface with a low albedo (e.g. is dark or matt in color and texture) absorbs solar radiation more effectively. It therefore is able to heat up faster and retain heat (Wielicki et al., 1995, Wielicki et al., 2005).

Local urban climates can be different to global climate. Microclimates are associated with the dense nature of urban developments and the high population density, resulting in elevated greenhouse gas emissions and air pollution. Urban microclimates are particularly defined by elevated temperatures, which are localized around city centers in comparison to surrounding rural areas (Blake et al., 2011, Oke and Cleugh, 1987). Elevated temperatures in cities result from the city infrastructure altering of the balance of solar radiation being absorbed and emitted (Oke and Cleugh, 1987). The albedo of a city is significantly lower than a natural surface due to the extent of dark asphalt roads and rooftops, thus cities absorb more heat (Rosenzweig et al., 2006). Impervious building surfaces reduce latent heat cooling from evapotranspiration. Furthermore the physical structure of cities with tall buildings and long streets creates urban canyons, trapping solar heat energy and reducing ventilation, preventing heat escaping back into the atmosphere. Building functioning, such as air conditioning and heating systems, as well as transport infrastructure and anthropogenic emissions add additional heat sources into the local environment (Blake et al., 2011). 

The urban heat island effect

Cities release heat, trapped by the urban heat island effect, at night, thus keeping ambient city temperatures elevated during what should be a cooler night time period (Laaidi et al., 2012).

This phenomenon of warming over city centers is known as the urban heat island effect. Some cities in the USA are warming at almost twice the global rate (Stone et al., 2012). Studies have shown that 50-68 % of observed warming since the 1950’s can be attributed to land use changes and increasing urbanization (Zhou et al., 2004, Kalnay and Cai, 2003). Greenhouse gas emissions although elevated (up to 60% greater than the global average) in urban areas, have been found to have little effect (~0.1 °C temperature increase) on local urban temperatures (Balling et al., 2001). This is because greenhouse gas emissions, although higher in urban environments are only elevated in the lower atmosphere. When looking in the upper parts of the atmosphere above cities, concentrations are close to the global average (Blake et al., 2011). In comparison the contribution of heat from the urban heat island effect may contribute between 5 and 10 °C to the local urban temperature (Balling et al., 2001).

Heat waves in urban centers

Extreme events, defined as climatic variations lasting for a limited amount of time, are a particular issue for urban centers.

Heat waves are an example of an extreme event. There is no accepted definitive definition of a heat wave (Souch and Grimmond, 2006). Simplistically they are considered as rare events with sustained elevated temperatures, which are known to affect human health (Kovats and Jendritzky, 2006). It is predicted that as future climate change continues there will be a shift towards longer lived and more frequent occurrences of heat waves (McGregor et al., In review). Furthermore it is predicted that areas that already experience intense heat waves may experience even more intense heat wave events in the future (Meehl and Tebaldi, 2004).  Urban environments are particularly vulnerable to the effects of heat waves. It is predicted that heat waves will increase deaths in urban areas by 22 % over the coming decade if strategies are not put in place to mitigate and adapt to the changing climate and heat waves.

Heat waves are also associated with increased threats of wildfires which have indirect impacts in the city as infrastructure, agriculture, ecosystems, air quality and water can be affected (Blake et al., 2011).

Cities such as Melbourne, which have to deal regularly with very hot conditions, are more prepared and adaptable to the extreme temperature conditions. In cities less well adapted, the occurrence of an extreme event heat wave can have severe consequences. In 2003 a heat wave occurred across much of northern Europe resulting in more than 30,000 deaths nearly half of which occurred in France (Rey et al., 2009). The mean summer temperatures were recorded to be 3.7 °C higher than the mean summer period recorded between 1950 and 2006 (Bessemoulin et al., 2004). In France the maximum temperature reached 35 °C in urban environments and the minimum night time temperature never fell below 20 °C (Charpentier, 2011).

  •  During such heat wave events it has been seen that mortality amongst the elderly and those with underlying medical conditions is greatest (Vandentorren et al., 2006).
  • Housing characteristics can also contribute to increased mortality with individuals top floor with a lack of thermal insulation associated with a greater mortality risk (Laaidi et al., 2012).
  • Socioeconomic status within urban environments can also have an impact on mortality risks during heat waves. The poorer parts of society cannot afford large, well-ventilated or air-conditioned spaces, thereby increasing their mortality risk (Hayhoe et al., 2010).
Climate services supporting policy makers

In the long term cities need to be more prepared for heat waves, providing information on protective and preventative measures. Long term adaptations to local infrastructure and planning of future developments should also be implemented, for example the inclusion of thermal insulation in roof spaces, which could provide protection from heat waves (Vandentorren et al., 2006).

In order to be better prepared for extreme heat wave events, policy makers need to be better informed of the occurrence, duration and intensity of heat waves. Enhancing collaborations, partnerships and communications between science communities and the policy makers is therefore a top priority.

This is where climate services come into action, acting as a middleman between the stakeholders and science communities, bridging the gap between science and policy, ensuring all needs are met.

References

Download the Word document for accessing the references mentioned above.