A new framework for using climate scenario data for impacts and assessment studies

No matter where the sources of greenhouses gases emissions come from, they become everyone’s problem once they reach the atmosphere. It is well known that anthropogenic emissions of greenhouse gases and aerosols induce important changes to the Earth's climate.

Recent research has shown that 1) global mean temperature is increasing proportionally to the amount cumulative CO2 emissions, and that 2) regional patterns of change for temperature and precipitation respond nearly proportionally to changes in global mean temperature.

In this context, it might be possible to establish a direct relationship between different types of anthropogenic emissions and the magnitude of local changes. Such a relationship could help to identify adaptation and mitigation needs without the constraint of assuming a particular trajectory for the future anthropogenic emissions.

• Generate patterns of annual and seasonal temperature and precipitation change scaled as a function of cumulative emissions of carbon dioxide while taking into account the uncertainty linked to the use of an ensemble of climate models;

• Use these scaled patterns to provide the best estimates and uncertainty ranges for the expected magnitude of local climate warming and precipitation change per teraton (Tt) of CO2 emission (Tt C);

• Derive relationships to define the patterns of climate change resulting from emissions of aerosols and non-CO2 greenhouse gases such as methane -CH4 -, nitrous oxide -NO2 - and CFCs;

• Test the resulting emissions-scaled patterns of climate change across a range of climate scenarios to evaluate their robustness and reliability for decision-making.

This project extensively analyzed the data from the Coupled Model Intercomparison Project Phase 5 (CMIP5) ensemble of global climate models to:

• Estimate the patterns of climate response to CO2 alone through the subset of all available simulations in which CO2 concentration increases by 1% per year to define the “Regional Transient Response to cumulative CO2 Emissions” (RTCRE);

• Consider annual and seasonal scales, for temperature and precipitation.

• Evaluate uncertainty associated with the different CMIPS climate models.

• Compare CO2 -only climate change patterns to those associated with the response to other greenhouse gases and aerosol emissions in the climate model projections.

This project has produced a new set of data that quantifies the climate response to cumulative human gas CO2 emissions, considering both annual-mean and seasonal patterns of temperature and precipitation. Patterns remain approximately stable with time, and appear to be robust across a range of emissions scenarios with increasing forcing during the 21st century. However, some small discrepancies appear in the patterns derived from the emissions scenarios assuming a stabilization or decline of greenhouse gases emissions. These discrepancies are the subject of ongoing research.

Figure 1 shows the average response of an ensemble of climate models for the annual-mean surface air temperature, as described in Leduc et al. (2016). These values represent the “Regional Transient Response to cumulative CO2 Emissions” (RTCRE) and range from less than 0.5°C per Tt C emitted in the North Atlantic and parts of the Southern Ocean, to more than 5°C per Tt C in parts of the Arctic. The global-mean response (the TCRE) from this model ensemble is 1.7°C per Tt C, with an uncertainty range of ±0.4°C per Tt C (1-sigma).

We found that over most land regions, the local temperature change is strongly linear with cumulative CO2 emissions and increases at a rate that is considerably higher than the global average. A very good example of this typical behaviour is shown on the right panel with the graph of the local temperature change per unit of cumulative emissions for the “Eastern North America” (ENA) region, which includes Southern Quebec (2.4°C per TtC for the ENA region; 3°C for Quebec as a whole, not shown).

Figure 1. Left: Local temperature change per 1 TtC of cumulative CO2 emissions (Regional Transient Climate Response to Cumulative Carbon Emissions: RTCRE). Right: Regional average temperature change for “Eastern North America” (ENA) region, as marked with the red box (land only) on the map

These patterns also hold well for seasonal temperature change, as well as for annual-mean and seasonal precipitation (Partanen et al. 2017). Both temperature and precipitation appear to have stable patterns with time, implying a linear regional climate response to cumulative emissions. Although temperature patterns are very robust across models, it is worth noting that the level of uncertainty is higher for precipitation patterns.

This project also explore how non-CO2 greenhouse gas and aerosol emissions contribute to alter the patterns derived from the relationship between cumulative CO2 emissions and the local climate response (Partanen et al. 2018, in preparation). Our analyses suggest that regardless of what emissions scenario we follow over the next century, the local annual-mean temperatures in Quebec will increase by 2.5° C (Southern Quebec) to 3.5°C (Northern Quebec) for every trillion tons of carbon emitted as CO2 . Winter warming will be larger (4-6°C per TtC) than summer (2-3°C per TtC), and will be accompanied by an increase of 5-15% per TtC in annual mean precipitation.

We are currently working to submit a paper that quantifies the local climate change resulting from non-CO2 emissions.