The December 2020 update to En-ROADS includes a new baseline scenario, which leads to a lower global temperature in 2100: 3.6°C (6.5°F) instead of 4.1°C (7.3°F).
These changes arise from three improvements to the simulator:
1. Improved modeling of renewable energy. We now model wind and solar energy separately and account for historical soft costs and subsidies. As a result, renewable energy use now grows faster in En-ROADS.
2. Change in the definition of the pre-industrial temperature benchmark. The global temperature change in En-ROADS (e.g. 3.6°C in 2100 in the updated En-ROADS Baseline) measures how much the world has warmed relative to a benchmark pre-industrial time period. The temperature change in En-ROADS was previously relative to the mid-1700s (to match the approach of other models), and now it is relative to the mid-1800s, a more common time period used in climate policy.
3. Improved modeling of non-greenhouse gas forcings such as aerosols, soot, volcanoes, cloud albedo, and others.
Note that the baseline was created as a reasonable starting point of minimal climate action from which to test various changes in policies and assumptions to see the impacts on our global climate. It is not a prediction of the most likely future or an assessment of near-term policy action.
Climate Interactive has released an updated version of our En-ROADS Climate Change Solutions Simulator. Here we explain the update and its implications for the climate system.
The En-ROADS Climate Change Solutions Simulator enables users to explore the impact of policies that will help limit global warming to no more than 2°C (3.6°F) through 2100 with the goal of avoiding irreversible damage to the environment, global economy, and public health. The free, online software allows people with all levels of technical skills to simulate the interactions among energy, land, and climate through the end of the century in less than one second. If you are unfamiliar with En-ROADS, watch this video for a short introduction.
Background
The purpose of En-ROADS is to improve decisionmaker and citizen understanding of energy, land use, and climate dynamics as a means to effective action. Our approach is guided by these two commitments:
Continually updating En-ROADS in light of new data and scientific understanding related to climate, energy, land, forests, agriculture, equity, demographics, public policy, and economic growth. We release updates to the model monthly and document the changes in our release notes. This December 2020 release reflects several important changes to the model.
Stitching together scientific approaches for decision-makers. We aim to resolve differing approaches across disciplines, projects, and paradigms in order to create effective tools for decision-making under uncertainty.
En-ROADS does not make a prediction about the most likely future. Instead, the simulator enables the user to explore different possible future scenarios and pathways to climate mitigation.
What has the result been? New baseline with lower future temperature
The December 2020 update to En-ROADS includes a new baseline scenario. The En-ROADS Baseline scenario is a starting point from which to test various changes in policies and assumptions to see the impacts on our global climate.
The En-ROADS Baseline scenario 2020 version (hereafter “En-ROADS Baseline 2020”) now leads to a lower global temperature in 2100: 3.6°C instead of 4.1°C.
There are several caveats to note about this change in the baseline temperature projection:
The En-ROADS Baseline scenario is not our prediction of the most likely future, or even of the most likely scenario. The baseline scenario is a reasonable starting point for experimentation. It helps us understand what the world might look like in the future in the absence of efforts to mitigate climate change.
The Baseline scenario is not a scenario intended to capture current policies. It does not include the net-zero emissions pledges by China, South Korea, Japan, and the European Union. It also does not include the NDCs (pledges to Paris Agreement) or long term continuation of other national policies.
An increase in global temperature of 3.6°C would be devastating to the global population, ecosystems, and economies. This global average also obscures substantial regional differences in warming. Some regions will experience average temperature increases that are much higher than the global average, with accompanying negative consequences.
Significant uncertainties in the global biogeochemical system could lead the temperature to be higher or lower. Estimates of the planet’s climate sensitivity span a broad range. Climate sensitivity refers to how much global temperature increases in response to a doubling of atmospheric CO2equivalents. In other words, how much does a given amount of greenhouse gases in the atmosphere cause the planet to warm?
Additional uncertainties arise from the possibility of methane release from thawing permafrost, for example. These uncertainties can be explored in the Assumptions settings of En-ROADS.
What contributed to the changes in the En-ROADS Baseline?
The En-ROADS Baseline 2020 includes three changes to the structure and parameters of the simulator:
Improved modeling of renewable energy
Change in the definition of the “pre-industrial” temperature benchmark
Improved modeling of non-greenhouse gas forcings
Figure 1 below illustrates the relative contribution of these three factors to the change in the En-ROADS Baseline temperature. The growth in wind and solar energy contributed a -0.40°C shift in the baseline temperature in 2100. The change in the definition of the pre-industrial period contributed a further -0.16°C decline, and the updated effect of other forcings caused a +0.10°C increase. All together, these updates led the baseline temperature increase in 2100 in En-ROADS to be revised downward to 3.6°C.
Figure 1
Let’s explore each of these three factors:
1) Improved modeling of renewable energy
Previously, the growth in renewable energy in the En-ROADS Baseline 2018 was calibrated to match versions of the Shared Socioeconomic Pathway 2 (SSP2) Baseline from Integrated Assessment Models.
Since the time of that calibration, the cost of wind and solar energy production continued to fall rapidly. Lazard reports the levelized cost of energy from utility-scale solar power dropping 89% from 2009 to 2019, and wind dropping 70%.
Figures 2 and 3 below show the substantial declines in the marginal costs of wind and solar energy, respectively, from 1990 to 2019 across several sources of measured historical data. Data from Lazard is shown in purple, the International Renewable Energy Agency (IRENA) in green, and the International Energy Agency (IEA) in yellow. The marginal cost of solar declined from $1456 per MWh in 1990 to around $60 per MWh in 2019, and the marginal cost of wind declined from around $200 per MWh in 1990 to around $50 per MWh in 2019.
Figure 2 and Figure 3
The falling prices and resulting growth in production led us to investigate our modeling of renewables and consider improvements. When we took a closer look at this data, we noticed that the growth rate of renewable energy as modeled in the En-ROADS Baseline 2018 was slightly lower from ~2005-2019 than in reality. Figure 4 below compares the growth of electricity generated by renewables from 1990 to 2019 in the En-ROADS Baseline 2018 to historical data from the International Energy Agency (IEA) and British Petroleum (BP). You can see that renewable energy as modeled in En-ROADS was growing more slowly from 2005 through 2019 than it did in the real world.
Figure 4
What was going on here? To figure this out, we made two improvements to how En-ROADS models wind and solar energy:
We disaggregated wind, solar, and geothermal energy – they used to be modeled together as “renewables.”
We recalibrated the costs of renewable energy to recent data by accounting for ancillary costs and subsidies.
We disaggregated wind, solar, and geothermal energy in En-ROADS in order to model them separately, including separate learning curves and prices. (This change to the modeling structure will not be readily apparent to the user of the online simulator, as wind and solar energy are still reported together with geothermal and hydro as “renewables” in most graphs.)
We also improved the modeling of historical ancillary costs and subsidies for wind and solar energy. Ancillary costs are soft costs, including difficulties with installation, siting, and permitting of renewable energy infrastructure. These ancillary costs, initially higher than we had accounted for in the model, increased the cost of renewable energy in the 1990s and early 2000s. Subsidies for renewable energy had an opposing, and larger, effect. The net result is that subsidies lowered the cost of wind and solar energy in the first decade of the 2000s enough to jump-start a reinforcing feedback loop, leading to an accelerated decline in the cost of renewable energy and growth in renewable energy production. We updated the model structure to account for these factors.
As a result, the En-ROADS Baseline 2020 is a better fit to history. The figures below show the fit to history of the new En-ROADS Baseline 2020 (in blue) from 1990 to 2019 for wind and solar marginal cost (Figures 5 and 6) and electricity generated (Figures 7 and 8).
Figure 5 and Figure 6Figure 7 and Figure 8
Figure 9 below aggregates wind, solar, and other renewables to show the electricity generated by renewables. The En-ROADS Baseline 2020 is now a better fit to the measured historical data.
Figure 9
The greater growth of renewables in En-ROADS Baseline 2020 as a result of these modeling changes has important implications. Cheaper wind and solar energy compete with other sources of electricity. As a result, coal and natural gas grow less in the new En-ROADS Baseline 2020 than before the update, because renewables grow more. Figures 10 and 11 compare the electricity generated by coal and natural gas in En-ROADS to the historical measured data from the IEA (in yellow) and BP (in orange).
Figure 10 and Figure 11
Comparison to other models
When we explore the long-term implications of improved model structure, we see that our new 2020 baseline now exceeds the growth of renewables in the SSP2 Baseline scenarios from the Integrated Assessment Models (IAMs).
As such, we no longer calibrate the En-ROADS Baseline to the SSP2 baseline from the IAMs. Instead, we compare to those scenarios, as well as others such as those of the IEA, Shell, and DNV-GL.
Note that we continue to use the IAM versions of the SSPs to calibrate other parts of En-ROADS, particularly the impact of various carbon prices (e.g., those in the scenarios that create radiative forcing levels such as 2.6, 4.5, etc.) on energy production from coal, oil, gas, renewables, nuclear, and biomass, as well as CO2 and greenhouse gas emissions.
Figure 12 compares the growth in renewable energy in the En-ROADS Baseline 2020 (solid green line) to the En-ROADS Baseline 2018 (dotted green line) and baseline scenarios in other models. Historical data is shown in orange. You can see that the growth in renewables in the En-ROADS 2020 Baseline is similar to the IEA World Energy Outlook’s (WEO) Stated Policies (STEPS) scenario, and higher than the SSP2 Baseline scenarios for the IAMs.
Figure 12
The higher growth in renewable energy in the En-ROADS Baseline 2020 as compared to the En-ROADS Baseline 2018 causes demand for the other main sources of electricity—coal, natural gas, and nuclear—to be lower. Figure 13 and 14 below compare the growth in coal and natural gas, respectively, in the En-ROADS 2020 and 2018 Baseline scenarios to scenarios from other models.
Figure 13Figure 14
Primary energy demand for oil is also lower in En-ROADS Baseline 2020 scenario, although it shows less of a change than natural gas and coal. Figure 15 below compares primary energy demand of oil in En-ROADS to other models. Oil (in the form of gasoline and diesel) is a major source of energy for transportation. In order for cheap renewable energy to displace oil, the transport system must be electrified. Hence, in a baseline scenario without policy changes to increase electrification, renewable energy does not displace much oil.
Figure 15
2) Change in the definition of “pre-industrial” temperature benchmark used in En-ROADS
The global temperature change in En-ROADS measures how much the world has warmed on average relative to pre-industrial levels.
Climate and energy models use different dates as benchmarks for defining “pre-industrial” climate, and therefore the temperature increase they report varies due to these different starting points. The En-ROADS Baseline 2018 was calibrated to the IAM versions of the SSPs, which as far as we can tell, were calibrated to observed temperature increase relative to the mid-1700s. Figure 16 below shows global temperature change from 1990 to 2019 with respect to the mid-1700s. Historical observational data from two different sources are shown: NASA’s GISS Surface Temperature Analysis (GISTEMP) in green (with 95% confidence intervals) and the Met Office Hadley Centre HadCRUT4 dataset in purple. These temperatures are adjusted to be relative to the mid-1700s. Model output data from a suite of six Integrated Assessment Models (IAMs) for the SSP2 Baseline is shown in yellow. The light blue line is the old En-ROADS Baseline 2018. (The dip in temperature in the early 1990s is due to the 1991 eruption of the Mount Pinatubo volcano).
Figure 16
The temperature increase in the 2020 En-ROADS version is now defined relative to the temperature of the mid-1800s, which is the definition used by the IPCC in their Fifth Assessment Report and a more common definition in the policy world. In keeping with our mission to translate scientific knowledge into policy-relevant knowledge, we decided to adopt this convention in En-ROADS. This change in definition lowered the temperature rise in the En-ROADS Baseline 2020 in 2100 by 0.16°C.
Figure 17 compares the En-ROADS Baseline 2020 (dark blue line) to the historical observational data from GISTEMP (green) and HadCRUT4 (purple) relative to the mid-1800s. Note, the observational data in Figure 16 above was adjusted to be relative to the mid-1700s, while the observational data in Figure 17 is relative to the mid-1800s.
Figure 17
3) Improved modeling of non-greenhouse gas forcings, including aerosols and solar irradiance
We also improved the modeling of non-greenhouse gas forcings, which include those from aerosols (black carbon, organic carbon, sulfates, nitrates, and biomass burning-related), cloud albedo, stratospheric ozone, tropospheric ozone, stratospheric water vapor from methane oxidation, land use, black carbon on snow, volcanic stratospheric aerosol, solar irradiance, and direct forcings from mineral dust aerosol. More details can be found in the En-ROADS Technical Reference. These changes in modeling of non-greenhouse gas forcings resulted in a 0.10°C increase in the En-ROADS Baseline 2020 temperature in 2100.
Notes on new interface changes in En-ROADS
This update includes two main changes to the En-ROADS interface:
New home graphs. We changed the default graphs to improve the usability of the simulator. The left graph is now “Global Sources of Primary Energy” and the right graph is “Greenhouse Gas Net Emissions.” (See Figure 18 below).
This helps users understand the causal flow: energy sources (on the left) create greenhouse gas emissions (on the right) which leads to an increase in temperature (number on the far right). The “Global Sources of Primary Energy” graph is the same as the previous En-ROADS version, except this one uses stacked area wedges rather than lines. The stacked version has the benefit of showing the total amount of primary energy used and alleviates the complexity of overlapping lines.
If you prefer the old default graph, you can find it by clicking on the title bar of the graph and selecting “Global Sources of Primary Energy” under “Primary Energy Demand Totals.”
The graph of “Greenhouse Gas Net Emissions” better shows the changes as a result of slider adjustments while building scenarios and highlights the connection between energy demand and global temperature change. The “Temperature Change” graph is still available under “Impacts” in the Graphs menu.
Figure 18
“New Zero Carbon” replaces “New Technology.” In En-ROADS, users can test the implications of discovering a brand-new, cheap source of electricity that does not emit greenhouse gases. This slider was previously labeled “New Technology,” but we have renamed the label to “New Zero Carbon.” The new label should help prevent confusion, as some users had been incorrectly assuming that “New Technology” included technologies outside the energy supply, such as carbon removal and electrification.
Want more detail? Follow along with Climate Interactive’s Co-Director Andrew Jones in a video tour to explore all of the latest features and updates to the En-ROADS simulator.
Interested in learning how to use En-ROADS?The En-ROADS Training Plan will prepare you to successfully lead events using the En-ROADS model – covering everything from event planning, to advanced facilitation tips, to deep dives on the system dynamics driving model behavior. You can find the training plan here.
The En-ROADS Climate Ambassador Program is a unique leadership opportunity intended for highly-motivated facilitators. Joining the En-ROADS Climate Ambassador program is free, and applications are open to any member of the public who has completed the En-ROADS Training Plan.
Upcoming events: Sign up for upcoming Climate Interactive webinars and events here.
