# How much warmer by 2100? Five futures from the Earth System Simulator

> The planet has warmed about 1.4 °C. Our simulator puts the end of the century anywhere from 1.3 °C to 4.4 °C warmer, depending almost entirely on the emissions path we take, and shows 1.5 °C being passed around 2030 in every scenario but one.

- Publication: Global Systems Lab, Transmission Report No. R-2026-02
- Published: 2026-11-01
- Authors: Global Systems Lab
- URL: https://globalsystemslab.com/transmission/how-much-warmer-by-2100
- Topics: Climate, Warming, Scenarios

**Notice.** This is experimental research by Global Systems Lab, provided as is and without warranty. It is not an official forecast, warning or statistic and must not be used for emergency, safety, planning, engineering, financial, legal or other professional decisions. Third-party data may be preliminary, delayed or revised, and their use does not imply endorsement by any agency. Full terms: https://globalsystemslab.com/disclaimer

## Abstract

We ran the lab's Earth System Simulator through the six standard emissions scenarios used by the
Intergovernmental Panel on Climate Change (IPCC), from the very low SSP1-1.9 to the very high
SSP5-8.5, using its default, observation-constrained ensemble of several thousand plausible climates.
By 2081–2100 the median warming above 1850–1900 ranges from 1.3 °C in the very low scenario to
4.4 °C in the very high one, closely matching the IPCC's Sixth Assessment Report. In every scenario
except the very low one, the 20-year average temperature passes 1.5 °C between 2028 and 2033; under
the intermediate scenario, which sits near the path implied by today's policies, it passes 2 °C in
the early 2050s and reaches 2.6 °C (very likely 2.0–3.5 °C) by the end of the century. Carbon dioxide
dominates the warming, while the cooling effect of air pollution is fading. The single largest
source of uncertainty after the emissions choice itself is how sensitive the climate is to carbon
dioxide, which spreads end-of-century outcomes in any one scenario across about 1.5 °C.

## Key findings

1. Warming in 2081–2100 ranges from 1.3 °C (very low emissions, SSP1-1.9) to 4.4 °C (very high, SSP5-8.5) above 1850–1900. The choice of emissions path matters more than any other factor.
2. The simulator's ranges agree with the IPCC Sixth Assessment to within 0.1 °C for every scenario's median.
3. In five of six scenarios the 20-year average passes 1.5 °C between 2028 and 2033. Only the very low scenario holds the peak near 1.5 °C, briefly, before temperatures fall back.
4. Under the intermediate scenario, close to current policies, 2 °C is passed around 2053 and warming reaches 2.6 °C (very likely 2.0–3.5 °C) by 2081–2100, still rising.
5. Carbon dioxide accounts for about 1.0 °C of the warming so far and about 2.2 °C of the 2100 total in the intermediate scenario; the masking effect of aerosols halves as air quality improves.

![Figure 1. Global warming since 1850–1900: observed (black, with WMO values for 2023–2025 as dots) and the simulator's median projections for five emissions scenarios to 2100. Shading shows the very likely range for the low, intermediate and very high scenarios.](https://globalsystemslab.com/transmission/how-much-warmer-by-2100/files/fig-century.png)

## Introduction

Global temperatures have now spent three years at or near 1.5 °C above their pre-industrial level.
The World Meteorological Organization (WMO) put 2024 at about 1.55 °C, the warmest year on record,
and 2025 at about 1.44 °C, among the three warmest. The eleven years from 2015 to 2025 are the eleven
warmest measured. With a strong El Niño under way since June 2026, this year is on course to be among
the two warmest, and next year could set a new record.

Single years are noisy, though. What matters for the long run is the trend underneath, and where it
goes next. This report asks three plain questions. How much warmer will the world be by the end of
the century? When will it pass the thresholds people talk about, 1.5 °C and 2 °C? And how much of
the answer is still in our hands?

To answer them we used the lab's own [Earth System Simulator](https://globalsystemslab.com/earth-system-simulator/), the same
tool anyone can open in a browser, and ran it through the standard set of future scenarios that the
climate research community shares.

## Data and approach

The Earth System Simulator is a fast climate model. It does not simulate weather, clouds or oceans
grid box by grid box, as the large models run on supercomputers do. Instead it tracks the planet's
energy budget as a whole: how much extra heat greenhouse gases trap, how much air pollution reflects
away, and how quickly the land and oceans warm in response. Because it runs in seconds, it can be run
thousands of times, each with a slightly different but plausible version of how the climate behaves.
Those versions are weighted by how well they reproduce the warming that has actually been observed,
so the spread of results is a fair picture of what we do and do not know.

The futures come from the Shared Socioeconomic Pathways (SSPs), the scenarios used in the IPCC's
Sixth Assessment Report. Each describes a different course for population, economic growth, energy
and land use, and the emissions that follow. We use five core scenarios and one overshoot path:

- **SSP1-1.9, very low.** Emissions fall steeply to net zero around mid-century.
- **SSP1-2.6, low.** Net zero carbon dioxide shortly after mid-century.
- **SSP2-4.5, intermediate.** Emissions stay near today's level until mid-century, then decline.
- **SSP3-7.0, high.** Emissions roughly double by 2100 in a fragmented world.
- **SSP5-8.5, very high.** Fossil-fuelled growth with emissions tripling.
- **SSP5-3.4 overshoot.** High emissions until 2040, then a rapid reversal.

All temperatures here are global surface averages relative to 1850–1900, the reference period the
IPCC uses for "pre-industrial". Where we talk about a threshold being passed, we mean the 20-year
average, not a single hot year. A "very likely" range covers 90% of the simulated outcomes.

## Results

### The century ahead

The opening figure shows the result in one picture. Up to the present the scenarios are
indistinguishable: the warming already locked in and the emissions of the next decade are similar in
all of them. They begin to separate in the 2030s and diverge sharply after 2050. By 2100 the
median paths range from about 1.3 °C to 4.8 °C.

The two lowest scenarios level off. Under the very low path, warming peaks at about 1.5 °C in the late
2030s and then falls slowly as carbon dioxide is drawn down; by 2300 it is back below 1 °C. Under the
low path the peak comes later, near 1.75 °C around 2070, and declines only gently. The other three
keep climbing through 2100. In the intermediate scenario warming is still rising at the century's end;
in the high and very high scenarios it is accelerating.

### End of the century, checked against the IPCC

Before trusting the projections we checked them against the IPCC's own assessment, which combined
complex models, observations and theory. For every scenario the simulator's median for 2081–2100
falls within 0.1 °C of the IPCC's best estimate, and the ranges overlap closely.

![Warming in 2081–2100 by scenario. Coloured bars and circles show the simulator's very likely range and median; grey bars show the IPCC Sixth Assessment's very likely range and best estimate.](https://globalsystemslab.com/transmission/how-much-warmer-by-2100/files/fig-end-century.png "Sources: Earth System Simulator; IPCC AR6 WGI Table SPM.1")

| Scenario | Simulator, 2081–2100 | IPCC AR6, 2081–2100 |
|---|---|---|
| SSP1-1.9 very low | 1.3 °C (1.0–1.8) | 1.4 °C (1.0–1.8) |
| SSP1-2.6 low | 1.7 °C (1.3–2.3) | 1.8 °C (1.3–2.4) |
| SSP2-4.5 intermediate | 2.6 °C (2.0–3.5) | 2.7 °C (2.1–3.5) |
| SSP3-7.0 high | 3.6 °C (2.8–4.7) | 3.6 °C (2.8–4.6) |
| SSP5-8.5 very high | 4.4 °C (3.3–5.8) | 4.4 °C (3.3–5.7) |

The gap between the lowest and highest medians, about 3 °C, is the clearest single message of this
report. It is larger than the uncertainty within any one scenario. How warm the world becomes is
mostly a question of what we emit.

### When the thresholds are passed

The 1.5 °C threshold arrives soon in almost every future. In five of the six scenarios the 20-year
average passes it between 2028 and 2033; the order is set mainly by how quickly air pollution is
cleaned up, since cleaner air removes a cooling haze, and by near-term methane. Only the very low
scenario holds off until about 2039, and there the excess is small and temporary.

![The year in which the 20-year average temperature first passes 1.5 °C (circles), 2 °C (squares) and 3 °C (diamonds), for the median projection of each scenario.](https://globalsystemslab.com/transmission/how-much-warmer-by-2100/files/fig-thresholds.png "Source: Earth System Simulator")

The 2 °C threshold is where the scenarios truly part ways. Under the very high path it is passed around
2042, under the high path around 2047, and under the intermediate path around 2053. The low and very
low paths stay below it in the median, although the low scenario still carries roughly a one-in-six
chance of exceeding 2 °C by 2100. The overshoot scenario passes 2 °C in the early 2040s, peaks near
2.4 °C around 2060 and then declines, ending the century close to 2 °C. Three degrees, a level with
far-reaching consequences for ice sheets, ecosystems and food production, is reached in the 2060s under
very high emissions and in the 2070s under high emissions.

These timings are consistent with independent estimates. The latest annual update of the Indicators of
Global Climate Change, published in June 2026, put human-induced warming at about 1.37 °C in 2025,
rising at roughly a quarter of a degree per decade, and found the remaining carbon budget for 1.5 °C
small enough to be used up in about three years at current emissions.

### What drives the warming

The simulator can separate the warming into its causes. The figure below does that for the
intermediate scenario, today and in 2100.

![Contributions to warming since 1850–1900 under the intermediate scenario SSP2-4.5, in 2025 and 2100. Bars to the right warm the planet; bars to the left cool it.](https://globalsystemslab.com/transmission/how-much-warmer-by-2100/files/fig-drivers.png "Source: Earth System Simulator")

Carbon dioxide is already responsible for about 1.0 °C of warming and grows to about 2.2 °C by 2100,
because it accumulates and lingers for centuries. Methane adds roughly 0.3 °C, and nitrous oxide,
industrial gases and ozone smaller amounts. The main cooling influence is aerosols: tiny particles
from burning fuels, which reflect sunlight and brighten clouds. They currently offset about half a
degree. As countries clean their air, that offset roughly halves by 2100, which is good for health but
adds to the warming. Changes in land use cause a little cooling, mostly by making the land surface
brighter. Volcanoes and the Sun play only minor, short-lived roles.

The same split shows why the scenarios diverge. In 2100, carbon dioxide alone contributes about
1.4 °C under the low scenario and about 4.0 °C under the very high one. Net-zero carbon dioxide is
what stops the climb.

### How much depends on the climate's sensitivity

Even with emissions fixed, the outcome is uncertain, because we do not know exactly how strongly the
climate responds to carbon dioxide. The usual measure is the warming from a doubling of carbon dioxide
once the climate settles. The IPCC's best estimate is 3 °C, very likely between 2 °C and 5 °C. The
simulator's ensemble, constrained by the observed record, spans a similar range.

![The spread of simulated warming for 2081–2100 under the intermediate scenario. Each bar is the share of plausible climates landing in that 0.1 °C range. The red line marks the warming the UN expects from current policies.](https://globalsystemslab.com/transmission/how-much-warmer-by-2100/files/fig-spread.png "Sources: Earth System Simulator; UNEP Emissions Gap Report 2025")

For the intermediate scenario, that uncertainty spreads end-of-century warming from about 2.0 °C to
3.5 °C, with a median of 2.6 °C. A low-sensitivity climate would give about 2.1 °C by 2100; a
high-sensitivity one about 3.8 °C. The distribution has a longer tail on the warm side: outcomes well
above the median are more likely than outcomes well below it. That tail is a reason to treat the
higher end seriously rather than planning only for the middle.

The United Nations Environment Programme's Emissions Gap Report 2025 estimated that policies in place
today lead to about 2.8 °C this century, and full delivery of countries' pledges to about 2.3–2.5 °C.
Both figures sit inside this distribution, close to the intermediate scenario. That is the closest of
the five to the world as it is currently heading.

### The longer view

The year 2100 is not an end point. Under the intermediate scenario, warming continues to about 3.1 °C
by 2300. Under the high and very high scenarios it reaches roughly 7–8 °C, a planet unlike any human
civilisation has known. Only in the low scenarios does warming stabilise or fall. Much of the heat
goes into the ocean, which the simulator estimates has gained about 450 zettajoules since 1971, and
the planet is still taking in about one watt of extra energy per square metre. That heat commits the
world to further sea-level rise long after temperatures stop climbing.

## Uncertainty and limitations

The simulator is a global model. It projects the world's average temperature and energy budget, not
regional climate, rainfall, extremes or sea level, which depend on details it does not represent. It
also does not include abrupt changes such as the collapse of ice sheets or large releases of carbon
from thawing permafrost, beyond what is implicit in its calibration. Those would push results toward
the warm end.

The scenarios are not forecasts. None is labelled most likely, and real emissions will not follow any
one exactly. The year in which a threshold is passed is given for the median; in the warmer members of
the ensemble it comes several years earlier, in the cooler members later. Natural variability, such as
the current El Niño, can push individual years well above or below the trend, which is why we use
20-year averages.

## Conclusions

The world is already close to 1.5 °C, and the simulator shows that threshold being crossed as a
20-year average around 2030 in almost every plausible future. What happens after that is far less
settled. If emissions fall to net zero by mid-century, warming peaks below 2 °C and may decline. If
they stay near today's path, the world is heading for about 2.6 °C by the end of the century, with a
real chance of 3 °C or more, and still rising. Higher-emission futures lead to 3.5–4.5 °C.

Of all the uncertainties in these projections, the largest is the one we control. The difference
between the lowest and highest scenarios, about 3 °C by 2100, exceeds every uncertainty in the physics.
Each tenth of a degree avoided reduces the heat, drought, flooding and sea-level rise that follow. You
can run these scenarios yourself, change the climate's sensitivity, and see the drivers for any year
in the [Earth System Simulator](https://globalsystemslab.com/earth-system-simulator/).

## References

1. IPCC (2021). Summary for Policymakers. In: *Climate Change 2021: The Physical Science Basis.* Contribution of Working Group I to the Sixth Assessment Report. Cambridge University Press. https://www.ipcc.ch/report/ar6/wg1/chapter/summary-for-policymakers/
2. World Meteorological Organization (2026). State of the Global Climate 2025. WMO, Geneva. https://wmo.int/
3. Forster, P. M. et al. (2026). Indicators of Global Climate Change 2025: annual update of key indicators of the state of the climate system and human influence. *Earth System Science Data*, 18, 3889. https://essd.copernicus.org/
4. United Nations Environment Programme (2025). Emissions Gap Report 2025. UNEP, Nairobi. https://www.unep.org/resources/emissions-gap-report-2025
5. Copernicus Climate Change Service (2026). Monthly climate bulletins, September 2026. https://climate.copernicus.eu/
6. Riahi, K. et al. (2017). The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications. *Global Environmental Change*, 42, 153–168. https://doi.org/10.1016/j.gloenvcha.2016.05.009

## Data availability

The projections behind every figure are available below as a CSV file: for each scenario and each
year from 1850 to 2300, the median and the 5th and 95th percentiles of global warming above
1850–1900. They come from a single run of the Earth System Simulator on 1 November 2026 with its
default settings. You can reproduce and extend them in the [Earth System Simulator](/earth-system-simulator/).

## How to cite

Global Systems Lab (2026). How much warmer by 2100? Five futures from the Earth System Simulator. Global Systems Lab Transmission R-2026-02. https://globalsystemslab.com/transmission/how-much-warmer-by-2100