Timo Denk's Blog

Becoming Carbon-Neutral Is Not Enough to Stop Climate Change

· Timo Denk

My understanding of the dynamics of climate change has evolved over the past months. The cause is a bi-weekly paper discussion in which a group of friends and I read research papers dealing with climate change (in parallel to our normal, weekly machine learning paper discussion group). After reading a handful of papers I realized that assumptions I had previously held about climate change do not match the current scientific understanding. I further came to realize that these misconceptions are common among my friends and are also shaping the political debate (at least in Germany).

At that point I decided to write a blog post. My aim is to (1) make my main learnings accessible to a non-paper-reading audience and help rule out misconceptions that seem to be common. (2) Structure and sharpen my own understanding and thoughts. (3) Expose my recently-acquired scientific understanding to fact-based critique from experts who might come across this post.

In the following I want to clearly distinguish my personal opinions, interpretations, and suggestions from the scientific findings. My two cents will therefore be written with an italic font.


Many believe that climate change can be stopped by cutting CO2 emissions. According to Randers and Goluke (2020), however, climate change has already become self-sustained and humans currently contribute only to its acceleration. Temperature as well as the sea level of the Earth’s oceans would continue to rise, even if human-made CO2 emissions were cut to zero right now.

An effective way of stopping climate change could be climate engineering. It is the active human intervention into the Earth’s climate system, e.g., through solar radiation management (Randers et al. (2016)). As a society we currently have a strong focus on cutting CO2 emissions. The idea of climate engineering is by far less popular in everyday life and on political agendas. To stop climate change, it would be more effective to redirect parts of our efforts towards climate engineering.

Climate Models

The two references in the abstract are based on a simple climate model called ESCIMO by Randers et al. (2016). A climate model specifies physical relationships between important factors that influence the Earth’s climate – the large-scale change of weather. A concrete example is that a climate model would model the relationship between the amount of CO2 in our atmosphere, the effect it has on radiation being trapped (greenhouse effect), and the increase in temperature following therefrom. This is a very simple example and in practice, climate models consist of many more variables and physical relationships between them.

Climate models can typically forecast the development of climate under several circumstances. For example, “how would the Earth surface temperature develop, if the Sibirian perma-frost soil melted within the next 20 years?” or “by how much would the sea level rise by 2050, if several large volcanoes erupted by 2030?”. Evaluating the correctness of such forecasts is an active field of research, see for example Eyring et al. (2019) or Hausfather et al. (2020). While climate models cannot model climate with absolute certainty, they are the best predictions available to us, so it is sensible to base our decisions on them.

ESCIMO is a simple climate model which was developed with aim for interpretability and user-friendliness.

The Figure shows the main feedback loops in the ESCIMO climate model (source Randers et al. (2016)). Feedback loops are best explained with an example: The sun causes heat flow from space to Earth. Depending on the amount of ice and snow cover, more sunlight gets reflected back into space or is being absorbed on Earth. The absorbed light heats up the Earth causing more ice to melt. Less ice cover leads to more light being absorbed. This is an example for a positive feedback loop meaning it reinforces itself. In the Figure this feedback loop is indicated by the red “+4”.

Recent History

Up until humans interfered significantly with the Earth’s climate, it was circling on a somewhat stable trajectory called the Glacial-Interglacial cycle. In this cycle both temperature and sea level would periodically rise and decline over the course of around 100,000 years. See Steffen et al. (2018) for details.

Humans have interfered with the system in a number of ways, for example, deforestation, creation of impervious surfaces, agriculture, and burning of fossil fuel. The newly injected greenhouse gases have pushed Earth off of its stable cycle and into a new system of dynamics which we do not fully understand yet, but attempt to model with climate models like ESCIMO. Humans triggered the recent climate change that is not a part of the Glacial-Interglacial cycle anymore.

From the beginning of the industrial revolution until now, Earth has already heated up by around one degree Celsius (see for example this article by NASA). The easy-to-make, yet false, deduction is that climate change would stop, if humans disappeared or if the population shrank significantly. The problem is that Earth has already heated up and more CO2 than usual is already in the atmosphere at this point in time. The increase in temperature has had several other effects like thawing permafrost, reduction of the surface area covered by ice, more water vapor in the atmosphere (which is an important greenhouse gas). According to Randers and Goluke (2020) all of them lead to even more warming, even without further human intervention.

The Figure from Randers and Goluke (2020) shows what would happen if humans were to cut emissions down to zero immediately (Scenario 2) or slowly bring them down to zero by 2100 (Scenario 1). In both cases the sea level keeps rising. With humans making the effort of cutting emissions the rise is delayed significantly, however, it is not prevented. This already feels like a surprising and counter-intuitive circumstance. Our main effort towards saving islands, cities, and countries from drowning is cutting CO2 emissions, but that will only delay it. Humans would have needed to cut emissions to zero, back as early as in the 1960s in order to not see a continued, self-sustained rise of temperature and sea level (see the conclusion section of Randers and Goluke (2020)).

Our Current Situation

Climate change is self-sustained. If humans keep on doing what they have been doing for the past decades, it will accelerate. If they cut emissions, the climate will still change, but more slowly. Cutting CO2 emissions globally down to zero is an enormous effort. And that won’t even solve the problem? Naturally, one would now ask for ways out. How can we stop climate change and not just slow it down? There are possibly solutions for that, but they are very different from the current “culture of renunciation” that we pursue and praise. These solutions are about active human intervention and steering of Earth’s climate. This might sound a bit scary. To me it does. That alone I consider problematic, because we’ll have to get acquainted with it – the sooner the better.

One example of such a human intervention is the injection of aerosols into the atmosphere (see the Figure below which is from Randers et al. (2016)). Aerosols reflect sunlight in the atmosphere before it reaches the Earth and can heat it up. Such an injection is not a one-time effort but would need to be carried out continuously and stopping it would make the Earth’s temperature likely bounce back up. It also comes with unpleasant side effects like harming water bodies of the planet.

The Figure shows the expected development of the sea level rise under different human policies. The pink policy is “injecting stratospheric aerosols” – one example of climate engineering.

Currently our society and politics are having an extreme focus on cutting CO2 emissions. It has become a popular topic, it is on political agendas, defines the debates, and is in people’s minds. Cutting CO2 emissions is arguably good – as we have seen above – it slows down climate change significantly. Further, we will have to become carbon-neutral eventually, because burning a limited resource (fossil fuel) is never sustainable, by definition.

The mental focus and the energy of humans is, however, a limited resource as well. From the cited climate research I draw the conclusion that our mental/political energy would be more effectively invested, if parts of it were redirected from reducing CO2 emissions, to developing climate engineering methods. An idea that I cannot spot in any political debates yet and that is not as omnipresent in our minds like “turning off the lights to save energy” or “taking the train rather than the car to reduce one’s carbon footprint”. In my current understanding:

  1. We need to get acquainted with the idea of controlling the planet’s climate, aka. climate engineering. We traditionally have this mindset of it being inherently good “to let Mother Nature do its own thing and not to interfere with it”. That is common for example in non-invasive medicine or in natural reserves where wildlife flourishes, but does not seem to hold true for the Earth’s climate anymore, given we have already interfered with it in the past and its warming is self-sustained.
  2. We need to invest into the development of climate engineering methods. This requires thinking big and out-of-the box. It also comes with uncertainty, because we do not know about its feasibility with certainty yet! Further, it is not an easy effort, but we have some time left and can potentially redirect resources from current efforts of cutting CO2 emissions.
  3. Actively controlling Earth’s climate comes with new challenges. Not everybody suffers from climate change alike. The climate, however, is a global matter and international agreements will be required to make an active control (e.g., through satellites shielding the planet from sunlight) feasible. We should start getting everyone on board here.

Reactive vs. Preventive Measures

Why should we act now? It’s worth considering the possibility of reacting to climate change once it really bothers us rather than trying to prevent it upfront. Reacting has several advantages. I mainly see that reactive measures would find high acceptance among the population as the issues are immediately apparent; measures can be more targeted; we would only act once climate change becomes real and wouldn’t need to worry about possible inaccuracies of current climate model predictions.

I still believe a proactive approach, i.e., acting right now to stop climate change, is the better choice. The reasons are:

  • Climate injustice. Developing countries might suffer more and earlier from climate change than rich countries, so we would harm others by not preventing climate change. This is morally unfair, not only because they suffer first, but also, because first-world countries are the ones who mainly caused climate change in the first place.
  • Fighting climate change once it is already painful could be more expensive than preventing it. Some things like the melting of arctic ice are likely to be much harder to undo than to prevent (see Guarino et al. (2020)).


In the context of climate change, reducing CO2 emissions is currently omnipresent in both our minds and politics. It is commonly seen as the go-to strategy against climate change. This is seemingly ignoring the fact that climate change is already self-sustained and that cutting CO2 emissions alone is not going to stop it. I advocate for redirecting some of our effort towards researching and implementing more effective solutions to the climate change problem. Such solutions likely involve active control of Earth’s climate through humans, potentially by partly shielding Earth from the Sun’s radiation.

Further Remarks

I am by no means an expert in climate change or climate engineering. The article is based on a handful of papers that I have read together with friends in a climate paper discussion group.

I want to emphasize that I am not against cutting CO2 emissions. It is undoubtedly an effective way of slowing down climate change. I do think we have to be aware of the limits of this approach to combating rising temperatures and sea levels. Further, I think the explanations above could be misused by justifying decreased effort in fighting climate change. I do not endorse that.

Why do I speak of “redirecting” resources away from cutting CO2 emissions rather than making an “additional” effort with climate engineering? When saying that I treat our mental resources and financial means as a fixed-sized quantity which we can distribute. Within that constraint, I think climate engineering is underfunded (relative to CO2 emission reduction), thus resources should be redirected. Having more budget overall would certainly be great.

When commenting on this article, please distinguish between opinions and research-supported assumptions like I did in the article as well. This helps understanding whether we are having a discussion about the current state of research or about moral values and what political agendas should contain.


Randers, J., Golüke, U., Wenstøp, F., and Wenstøp, S.: A user-friendly earth system model of low complexity: the ESCIMO system dynamics model of global warming towards 2100, Earth Syst. Dynam., 7, 831–850, https://doi.org/10.5194/esd-7-831-2016, 2016. The ESCIMO climate model that this article is mainly based on. The paper contains – among other things – several projected climate developments under different interventions (cutting CO2, injecting stratospheric aerosols, etc.) or natural disasters (more volcanic eruptions, more high clouds, etc.).

Randers, J., Goluke, U.: An earth system model shows self-sustained thawing of permafrost even if all man-made GHG emissions stop in 2020. Sci Rep 10, 18456 (2020). https://doi.org/10.1038/s41598-020-75481-z Application of the ESCIMO climate model to show that climate change is self-sustained.

Guarino, MV., Sime, L.C., Schröeder, D. et al.: Sea-ice-free Arctic during the Last Interglacial supports fast future loss. Nat. Clim. Chang. 10, 928–932 (2020). https://doi.org/10.1038/s41558-020-0865-2 This paper applies climate models to the last interglacial during which the arctic ice also melted. It is referenced here in the context of how long it takes to reverse natural events like the melting of arctic ice cover.

Steffen, W., Rockström, J., Richardson, K., Lenton, T. M., Folke, C., Liverman, D., Summerhayes, C. D., Barnosky, A. D., Cornell, S. E., Crucifix, M., Donges, J. F., Fetzer, I., Lade, S. J., Scheffer, M., Winkelmann, R., Schellnhuber, H. J.: Trajectories of the Earth System in the Anthropocene. Proceedings of the National Academy of Sciences Aug 2018, 115 (33) 8252-8259; DOI: 10.1073/pnas.1810141115

Eyring, V., Cox, P.M., Flato, G.M. et al.: Taking climate model evaluation to the next level. Nature Clim Change 9, 102–110 (2019). https://doi.org/10.1038/s41558-018-0355-y

Hausfather, Z., Drake, H. F., Abbott, T., & Schmidt, G. A. (2020): Evaluating the performance of past climate model projections. Geophysical Research Letters, 47, e2019GL085378. https://doi.org/10.1029/2019GL085378