1.5.5 Final remarks

As anthropogenic global warming continues, tipping systems are at risk of crossing critical thresholds (Armstrong McKay et al., 2022). Several assessments have investigated the risk of crossing critical thresholds of individual tipping systems, whereas interactions between tipping systems are only more recently taken into account, mostly by conceptual models (e.g. Sinet et al., 2023; Wunderling et al., 2023b; Dekker et al., 2018). 

Based on the current state of the literature, we conclude that tipping systems interact across scales in space and time (see Figure. 1.5.1 and 1.5.2), spanning from subcontinental to nearly planetary spatial scales and timescales from sub-yearly up to thousands of years. We find that many of the discussed interactions between tipping systems are of a destabilising nature (Figure 1.5.3), implying the possibility of cascading transitions under global warming. Of the 19 discussed interactions, 12 are assessed as destabilising, two are stabilising, and five are unclear (see Figure 1.5.1). Assessing the overall stability of the Earth system, and the possibility of a chain of nonlinear transitions, will however require more detailed assessments of their interactions, strengths, timescales and climate state-dependence.

While there is increasing research on individual thresholds of climate tipping systems, substantial uncertainties prevail in the existence and strength of many links between tipping systems. In order to decrease such uncertainties, we propose three possible ways forward:

1. Observation-based approaches: Satellite observations, reanalysis and palaeoclimate datasets may be evaluated using correlation measures (Liu et al., 2023), or advanced methods of inferring causality (e.g. Runge et al., 2019; Kretschmer et al., 2016; Runge et al., 2015). In-situ monitoring is also very important for most of the tipping systems as well, and in particular for the biosphere (see Chapters 1.3 and 1.6).
2. Earth system model-based approaches: With recent progress, Earth system models of full or intermediate complexity could be used to evaluate interactions between climate tipping systems in detail at the process level, and quantify their interactions using specifically designed experiments (see Chapters 1.2, 1.3, and 1.4).
3. Risk analysis approaches: Since relevant parameter and structural uncertainties are large within Earth system models, analysing model ensembles with a considerable number of ensemble members is very helpful in order to comprehensively propagate uncertainties for risk assessments (Daron and Stainforth, 2013; Stainforth et al., 2007; Murphy et al., 2004). 
4. Finally, all three approaches above have their limitations, and could probably benefit from direct expert input. Therefore, expert elicitation exercises on tipping system interactions remains of high value to update and move beyond early investigations of this kind (Kriegler et al., 2009).

To summarise, the approaches above (and likely more) are required to obtain more reliable estimates of the existential risks potentially posed by tipping events or even cascades (Kemp et al., 2022; Jehn et al., 2021). They could be used to inform an emulator model for tipping risks, taking into account properties of individual tipping systems as well as their interactions. In addition, there also exist large uncertainties, not only among the known interactions as discussed above, but also because not all interactions are known or quantified (i.e. known unknowns versus unknown unknowns). 

Further, in certain systems there are forcings of non-climatic origin that could interact with climate change and lead to tipping, and thus to interactions and possibly cascades with other systems. For instance, land use change and specifically deforestation are threatening the Amazon and decreasing its resilience to climate change (e.g. Staal et al., 2020; Boulton et al., 2022) (1.3.2.1). Lastly, systems do not necessarily tip fully in one go, but can also have stable intermediate states (such as through the formation of spatial patterns). This has mostly been reported in ecological systems, but is not limited to them (Rietkerk et al., 2021; Bastiaansen et al., 2020).

Taken together, assessing and quantifying tipping system interactions better has great potential to advance suitable risk analysis methodologies for climate tipping points and cascades, especially because it is clear that tipping systems are not isolated systems. The relevance for developing such risk analysis tools to assess tipping events and cascades is clear given the potential for existential risks and long-term irreversible changes (Kemp et al., 2022).