The e-book ‘Climate Change for Astronomers’ led by Dr. Travis Rector addresses how astronomers can communicate and educate others on the science of climate change. The purpose of the book and the first chapter is summarised here. For this bite, I summarise Chapter 3, written by Dr Anna Cabre, a climate physicist and oceanographer with a cosmology and astronomy background. This chapter’s purpose is to inform us on climate models themselves, their projections, and thus how they inform our decision making regarding climate change.
Unlike short-term weather forecasts, climate models inform us on the long-term spatial average of patterns in our climate. Essentially, climate models are big simulations that include the interactions between various mechanisms that drive variations in the climate. The main components of these interactions involve couplings between the Earth’s land, its oceans, sea ice and the atmosphere. Because many of these interactions are complex, we have to take a numerical approach, in which computers calculate the solutions to the equations that describe these physical processes in an approximate manner. These include the Navier-Stokes equations, which are used to describe the behaviour of fluids (the atmosphere and ocean in this case). These equations can be solved at different points in space by dividing the system into a 3-dimensional grid. The models also must balance radiation the planet absorbs or reflects from the Sun, and the energy that is radiated away, to calculate the Earth’s temperature.
Compared to the earliest computer-based climate models in the 1960s, climate models are more complex and can also include the impacts of ice shelves (ice that is attached to land), the vegetation on the land, the topology and geography of the land and other complex biological or chemical influences. These more complex models are called Earth System Models. They are able to give us insight into climate patterns that are empirical (observed rather than based on theory), but can’t really be understood from fundamental physics due to the complexity of the interactions. In the era of a changing climate, they also let us isolate and study impacts of different mechanisms, which can help us determine impacts on the climate that are natural vs anthropogenic (man-made). There are also now ‘Integrated Model Assessments’ that take into account socioeconomic factors for climate change (such as population growth and economic growth) that are included in climate models to accurately reflect their interaction with the climate.
Limitations of climate models
As with any simulation, climate models are limited by computational power: less computational power available means lower resolution for the system (both in space and time). This impacts how well we can include the effects of geographic features like mountain ranges or valleys, or similar features that exist in the ocean. Some processes may also occur on time scales that are shorter than the resolution of the simulation. This can be compensated for by parameterizing some processes and fine-tuning the values of the parameters based on observations. Parameterization can similarly be used for other complex processes that are not understood well enough to be included in another manner. Clouds, which at any given time cover around ⅔ of the Earth’s atmosphere, are also crucial in climate models given they can cool or warm the atmosphere, but change on very short timescales so are difficult to understand and are complex to model. Additionally, the observational data that we feed into climate models limits their accuracy, but this has improved with more data available from satellites and remote sensing.
How well can we trust projections from climate models?
The most accurate projections we can draw from climate models come by drawing inference from the average of an ensemble of models. There are various ways scientists can validate the predictions from models, or clarify the influence of humans on climate change as opposed to natural variability. One way to verify the accuracy of models is to run them based on parameters from past times, then compare the predictions to the past observations corresponding to the projections. Overall, Earth System Models have become quite robust at predicting Earth’s temperatures, and the predictions of climate models have generally become stable since the 1970s (such as predicting the impact of greenhouse gas emissions on global warming). While natural causes of climate change are not always easy to distinguish from those that are anthropogenic, one can run these models while leaving the level of greenhouse gas emissions at pre-industrial levels (1860s) to study what impacts are seen on the climate (a ‘control’ simulation). Models are also ‘run’ on long timescales well before the epoch of anthropogenic influences, to ensure that they are stable prior to adding additional complexities (they ensure processes in the simulation reach equilibrium prior to adding greenhouse gas emissions).
There are also now protocols in place to ensure a particular standard for designing Earth System Models and studying climate change impacts; this is the Coupled Model Intercomparison Project (CMIP). Every 6-7 years, the Intergovernmental Panel for Climate Change (ICPP) also produces a report to summarise and review the trends in observations and projections from climate models (following the CMIP guidelines) across a range of research institutions; these reports help inform decision making regarding climate change across the globe.
Climate change predictions and tipping points
Having some understanding of climate models, now we can start to talk about their predictions, in particular regards to rising temperatures. Observations of reduced sea ice and snow, warmer oceans at various depths, warmer land, increased water vapour in the atmosphere and rising sea levels suggest the planet is warming. It is the consensus of the ICPP that human activities have increased the concentration of greenhouse gases in the atmosphere. Due to this, since prior to the Industrial Revolution, the global surface temperature has risen on Earth by about 1.1 degrees Celsius, and all climate models and scenarios forecast a gradual temperature rise that will continue rising (and could peak after greenhouse gas concentrations peak in the atmosphere).
In the Arctic, the rise in temperature is about two times faster than in other locations, and the rising temperature is also intensified in Antarctica. This is because sea ice has high reflectivity, and when it melts the loss of the reflective surface allows for more rapid heating. Rising sea levels are largely attributed to the loss of sea ice in Antarctica and Greenland, but in part the warmer oceans expand and this also increases global sea levels. Since warmer air holds more water, increased air temperature intensifies the water cycle, allows for heavier precipitation (although this is location dependent), intensifies weather events, and increases the contrast between wet and dry seasons. Warmer oceans and melting sea ice also impact currents that are driven by gradients in temperature and salinity. Since currents deliver nutrients to phytoplankton that live at the surface of the ocean in order to photosynthesise (and produce oxygen that we breathe in the process) this can impact oxygen available for life since the phytoplankton may suffer (this may also depend on their location in the ocean). Warmer oceans also lead to coral bleaching and reduce the ability of the ocean to absorb and store CO₂ (which helps to prevent global warming). While the heating ocean tends to mitigate the impact of climate change on land temperatures to an extent, the continued rise of temperatures on land can nonetheless increase heat events leading to droughts, changes in precipitation, fires, water scarcity, and agricultural impacts. These impacts also all lead to increased social inequality.
Climate tipping points refer to points past which a temperature rise may lead to irreversible effects that are self-perpetuating; some of these are regional, while others are global in nature. As such, the Paris Agreement in 2015 led to a global initiative to prevent a temperature rise of more than 1.5 degrees Celsius relative to the preindustrial era. Some of these ‘tipping points’ may involve the melting of permafrost in the Arctic, which is essentially frozen underground land. Since this melting will release a large amount of CO₂ and methane gas, which contribute to global warming, it could be difficult to prevent the impact cascading and may have long term impacts on Earth’s climate. Since climate patterns like the ENSO and seasonal climate cycles and conditions across the globe are coupled to currents in the oceans in a complex manner, these patterns or cycles can be impacted by changes to ocean currents due to warmer waters, which may be irreversibly impacted in a worst case scenario. Likewise, the destruction of coral reefs (such as bleaching of a significant portion of the Great Barrier Reef) may have irreversible impacts on ecosystems and food supplies. Finally, changes to rainfall in the Amazon could lead to it drying out and damaging an ecosystem that would no longer be supported. The loss of the Amazon rainforest would have global impacts given its role in regulating the Earth’s temperature and carbon storage.
Future goals
The Paris Agreement’s goal to limit the temperature rise to 1.5 Celsius is ambitious; however, it has had a positive impact already on countries that are more vulnerable to climate change as many countries have started to reduce their emissions and set targets. In order to reach this ambitious goal, it will be important to continue communicating the impact of climate change across the globe and motivating transformations to our energy sectors. Energy solutions with lower carbon emissions continue to be more competitive (you can see more on these which are discussed in the Chapter 7 summaries, Part 1 and Part 2). Despite this, there is still work to be done to continue reducing greenhouse gas emissions. Part of the process to reach this goal involves communicating and understanding climate models, their limitations and their projections for the future.
Edited by Sarah Stevenson
