1 Introduction: climate change mitigation investment

Global heating and climate change are driven by an imbalance between the quantity of solar energy received on Earth and that escaping back into space [1]. This can be affected by solar output, but the most important proxy measure and the only one that humanity can feasibly control is the increasing content of greenhouse gases (GHGs) in the air. Stabilising this is the focus of climate change mitigation investment, defined here as all action taken in the expectation of a reward for reducing net GHG emissions. The motivation depends upon the values of the investor and can take several forms, usually financial but often also political, reputational, emotional or diplomatic. Collective security can also be a motivation, since climate change is increasingly seen as an existential threat to humanity [2,3,4]. This is because cascades of transformation among connected Earth systems could soon render the world unable to support a viable human population. Mitigation efforts are increasingly driven by scientific and public awareness of these risks.

For each country, mitigation investment may be internal or external. Internally, it aims to decarbonise national economies through government action in the form of policies, laws, plans, fiscal incentives, direct investments, and partnerships with the private sector and other non-governmental actors. This has become increasingly associated with declarations of 'climate emergency' [5], and policies to achieve net zero national GHG emissions, usually by 2040–2060 [6]. Implementing them involves national budgets and structural change in whole economies, so investments can be very large but hard to apportion by effect, and their overall effects are proxied by national emissions reports [7].

To facilitate global cooperation, (1) internal mitigation investment by developed countries and (2) external investment by developed countries in mitigation and adaptation in developing countries are both needed. Overseas public mitigation investment has been shaped since 2009 by developed countries agreeing to provide or mobilise jointly 100 billion United States dollars (USD) each year by 2020 to meet the needs of developing countries [8]. This was reiterated and extended to 2025 by the Paris Agreement which came into force in 2016. Climate finance 'provided' (from public sources) and 'mobilised' (from private sources in partnership with public ones) increased thereafter from a mean annual total of USD 57.1 billion in 2013–2015 to USD 76.3 billion in 2017–2019 [9]. Most of the increase was from public sources, and mobilised private finance remained almost constant at about USD 13.8 billion/year in 2013–2019. Progress towards USD 100 billion annually is diplomatically vital, but the sums are small in a global economy with an aggregate GDP of USD 94.1 trillion in 2021 [10], and against global GHG emissions that in 2021 reached their highest-ever annual level [11] and have not yet peaked [12].

Total global climate-related primary investment, which includes all the above plus exclusively-private investment, was at an annual rate of USD 365 billion in 2013/14, USD 463 billion in 2015/16, USD 574 billion in 2017/18 and USD 632 billion in 2019/20 [13, 14]. In the two latter periods, almost all private financing went into renewable energy (RE) generation, a sector that attracted total investment (excluding large hydro) of about USD 2.7 trillion in 2010–2019 [15] but in which public investment was only USD 19 billion annually in 2018–2020 [16]. Solar and wind accounted for 73% of growth in RE generation capacity since 2016, and RE is now the cheapest form of power to develop in almost all countries [17].

This makes public financing for RE generation almost redundant, and effectively leaves the public to fund other parts of the climate response agenda. These include: (1) integrating RE into national grids; (2) promoting economy-wide emission reductions from energy efficiency (EE), green growth, low-carbon development and the protection of high carbon-density ecosystems; (3) building capacity and choice awareness among partner governments; (4) creating fiscal and regulatory conditions to incentivise private climate investments; (5) strengthening resilience of urban and rural communities against climate impacts in all their regional manifestations [18,19,20,21,22,23,24]; and (6) safeguarding environmental security and other services provided by ecosystems [25]. Hence it is the public sector that bears responsibility for leading some of the most complex tasks in the climate emergency response, including disaster aversion, strategic innovation and the creation of carbon markets and prices to which private investors can respond. Choices made by public investors are thus key to setting the pace, influencing the quantity, and maintaining the quality of climate-relevant investments.

For public overseas mitigation investment, the priority is to invest cost-effectively in ways that are practical within the bureaucratic machinery of partner governments. With this in mind, a 2020–2021 evaluation of Denmark's official climate aid portfolio [26] scored it well for design quality and performance, but identified weaknesses in how investments were chosen for strategic mitigation impact and cost-effectiveness. Evaluations in 2014 [27] and 2021–2022 [28] of the Swiss official climate aid portfolio found similar strengths and weaknesses. Both the Danish and Swiss portfolios showed: (1) limited appreciation that preventing harm to high carbon-density ecosystems can yield major mitigation, adaptation, biodiversity and other benefits; (2) limited use of expected and quantified GHG emission reductions to justify projects; (3) limited monitoring of actual emission reductions during projects; and (4) limited awareness of the strategic importance of delivering net emission reductions as quickly as possible. The second of these weaknesses is being addressed by other agencies, notably the Green Climate Fund which now requires mitigation investment projects at preparation stage to estimate the expected tonnes of carbon dioxide equivalent (tCO2e) to be reduced or avoided in emissions for every USD of GCF contribution [29, 30].

This paper uses data from the 2020–2021 Danish study to explore the fourth of these issues: the importance of timing. It draws lessons from public grants to three developing countries: to promote high carbon-density ecosystem protection, to finance RE generation, and to build capacity in the RE/EE sector. It develops the idea that awareness of how complex systems behave under sustained stress, in particular their tendency to resist change but then suddenly to transform themselves, can help guide investments to the highest-value mitigation targets. A novel method is described and demonstrated to calculate the true biophysical value of returns on mitigation investments so they can be compared and individually justified. This method is widely applicable and could increase cost-effectiveness within all public mitigation portfolios, while offering new paths to profitability for private ones.

2 Problems arising from stressing complex systems

2.1 The heating of the biosphere

Earth's biosphere is an extremely complex global system that comprises complex subsystems such as the Arctic, Amazon, and various monsoons and ocean currents, all interconnected and influencing one another over deep, recent and current time [31,32,33]. Like all complex systems, these tend to resist change, owing to the elasticity and flexibility of relationships among the entities that comprise them [34,35,36,37,38]. They may therefore show few signs of reaction under sustained stress, while internal tensions escalate and they become increasingly likely to reach “a critical threshold beyond which a system reorganizes, often abruptly and/or irreversibly” [39], a threshold known as a tipping point [40, 41]. Continuing resistance may delay the onset of qualitative system change even after the tipping point, but it is by then inevitable and when it occurs will initiate a period of transitional chaos leading to a new kind of emergent stability. This whole process comprises a tipping pathway.

Resistance, delays, inadequate historical and current data, and evolving models can all make it hard to predict a tipping point with precision. But for purposes of precautionary planning it is possible to anticipate an approaching tipping point and try to prevent it, without knowing the details of exactly how any of the system actors will behave. This is vital with many little-known but connected subsystems under stress, each deforming in its own way, affecting one another, and contributing to the potential transformation of the biosphere. All their likely tipping points can then be aggregated into one foreseen event with a tentative date, and planned for. With uncertainty involving connected systems, events are moreover unlikely to unfold in a linear way. Potential cascade effects among connected Earth systems [4, 42, 43] imply that there is greater realism and safety in assuming an exponentially rather than a linearly increasing probability of linked changes occurring over time.

Atmospheric concentrations of 2 important GHGs, carbon dioxide (CO2) and methane (CH4), have increased dramatically since about 1950, relative to those over the past 800,000 years, and in June 2022 exceeded 421 parts per million and 1,900 parts per billion respectively [44,45,46]. The sudden escalation from about 1950 was due to past GHG emissions having saturated the ‘sinks’ (absorptive systems) that had previously removed them, or from annual emissions having exceeded the annual absorptive capacity of those sinks, or both, allowing a surplus to start accumulating in the air.

From about 1950, therefore, increased trapping of solar energy within the biosphere occurred. Total heat gain in 1971–2018 is estimated at 358 ± 37 ZJ, the heating rate increasing from about 0.50 W m–2 in 1971–2006 to 0.79 W m–2 in 2006–2018 and 0.87 W m−2 in 2010–2018; about 90 percent overall is accounted for by ocean warming and most of the rest by land warming and ice melting [1, 47]. Anomalously rising temperatures have now been confirmed for the Earth's air and surface since about 1970 [48], and for the upper 2000 m of the world's oceans since about 2000 [49]. This difference in timing makes sense because the Earth's air and surface heat up and cool down rapidly, but this is not so for the oceans. Water has a high heat capacity (see Sect. 2.2), while the mixing of water by upwellings and currents also ensured that the oceans warmed relatively slowly.

Energy flux is a fundamental attribute of all active systems, so adding energy to Earth systems will stress and alter them. Most of the effects are unknown, but a number of environmental changes are thought to be driven primarily by excess heat. These include: thermal expansion of water to help explain observed sea-level rise [50]; warm surface waters feeding heat and water vapour into higher-energy and more frequent storm systems [51, 52]; heat-waves becoming more common on land [53] and within the sea [54]; the advance of spring flowering northwards on land [55]; the ascent of ecological zones on mountains [56]; and the melting of multi-year ice formations in polar regions and mountain areas [57, 58]. The latter also contributes to sea-level rise and the potential distortion of marine currents and the on-shore weather systems that depend upon them.

2.2 The melting of the Arctic

The Arctic system is centred on the geographic north pole and includes the Arctic Ocean and its seabed and littoral, broadly within the Arctic Circle at about 66° 30' N [59] but with many biophysical connections to the south, including with the forests of the Arctic-boreal zone [60]. Sea ice is a key feature of the Arctic Ocean, and each year its extent varies between a maximum in April and a minimum in September. Some is deep, multi-year ice that is retained through the summers of consecutive years, while the rest is shallow, single-year ice that forms and melts seasonally [61].

Global heating is amplified in the Arctic and occurs 2–3 times faster than elsewhere [62, 63]. This is largely due to the lack of convection in cold air and the release of imported energy by condensation of water vapour [64]. As the Arctic warmed in the 1980s and 1990s, multi-year ice was lost from increasingly large areas [65]. As first-year sea ice came to dominate, there was a steady reduction in mean ice volume in each month from year to year [66,67,68,69,70,71,72]. The key metric is the minimum volume each September, which has declined from about 15,000 km3 in 1979 to around 5,000 km3 in 2022.

A tipping point in the Arctic system is anticipated when all the first-year ice melts during the summer, since thereafter: (1) no further multi-year ice can be formed and any remnants in sheltered and isolated places will in due course vanish; and (2) increasingly more of the Arctic Ocean will be ice-free and exposed to sunlight for increasingly longer periods in the summer [73]. Reasons to expect this transition to be irreversible include the increased heat content of the global ocean (see Sect. 2.1), and the effects of reduced ice albedo in the Arctic summer [33], but this view is contested by the Intergovernmental Panel on Climate Change (IPCC) and others [74, 75].

When displayed in a form known as the Arctic 'death spiral', the distance from the centre of a circle represents the mean ice volume in a particular month and year, so the declining ice volume spirals towards a centre point that represents zero sea ice [76, 77]. Extrapolated trends from submarine and satellite observations suggest that this centre point will be reached in a September before the early 2030s [78, 79]. Computer simulations used by the IPCC offer a more nuanced prospect, in which “the Arctic is likely to be practically sea ice-free in September at least once before 2050 under the five illustrative scenarios considered in this report, with more frequent occurrences for higher warming levels” [80]. In high CO2 emission scenarios, a practically sea ice-free Arctic is also expected to become the norm for late summer by the end of the twenty-first century [81].

Longitudinal observations of sea ice volume are credible and proposed mechanisms of change and irreversibility in sea-ice conditions are plausible, while climate simulations seem to be being outpaced by events such as the severe northern hemisphere droughts and heatwaves in 2022 [82,83,84,85]. With these points in mind, a date around September 2030 is accepted as a likely and precautionary one for a practically ice-free Arctic Ocean. But regardless of the precise timing, there are other factors at work in the Arctic system that increase the potential for cascade effects within and beyond the Arctic, which together make the Arctic a lynchpin of global climate security over the foreseeable future.

Most importantly, the Arctic system contains several trillion tonnes of carbon-rich organic peat that has existed for millennia as permafrost [86,87,88]. These permafrosts are now melting [89, 90] and the formerly-frozen peat has accordingly dried out and burned to release CO2 and soot [91, 92], and/or decayed to release CH4 [93,94,95]. Predictions based on continued gradual thawing range up to several hundred Gt of additional CH4 emissions in this century, with potential to upset efforts to meet the 1.5 °C temperature goal [96, 97]. A 2019 special report by the IPCC noted this potential but also a lack of certainty over whether plant growth and soil replenishment would offset the effect [98]. The 2021 IPCC assessment also noted that biogeochemical cycles can respond abruptly at regional scales [51], but had low confidence in the magnitude and timing of such increased CH4 emissions [99].

Some CH4 is also contained in sea-bed deposits of methane hydrate (or clathrate), water ice with a high density of CH4 in its crystal structure that is released on warming. The total amount of carbon stored in this form in the Arctic is unknown, but may amount to hundreds of Gt and some is being mobilised from shallow sea beds as a result of warming conditions [100]. Methane pluming from Arctic sea-beds is increasingly reported [101], but the 2021 IPCC assessment was that this is unlikely to cause much divergence from expected GHG emissions during this century [102].

An important additional factor, however, is the 80-fold difference between the ‘heat capacity’ and ‘latent heat of fusion’ of water [103, 104]. The first means that it takes 4.2 J to heat 1 g of water by 1 °C; the second means that it takes 334 J to melt 1 g of ice at − 1 °C. Much of the extra heat in the Arctic has thus far been melting ice rather than heating water, and when the last sea ice is gone this will no longer be the case [105, 106]. Rapid heating would then be expected to follow in and around the Arctic Ocean, accelerating the decay of formerly-frozen peat and the mobilisation of CH4 from sea-bed hydrates. Hundreds of Gt of potential CH4 emissions are available from both sources, and CH4 is much more potent as a GHG than CO2 (see Sect. 3.2), so this could severely impact climate change mitigation and adaptation plans. Uncertainties abound [107], but the scale of potential impact is such that urgent studies are needed to explore and model the response of the Arctic Ocean and the Arctic-boreal zone to the early loss of summer sea ice, and its implications for the biosphere and humanity's climate change response.

For reasons outlined above, a tipping point in the Arctic system can be anticipated when summer sea ice is lost almost entirely from the Arctic Ocean. Transitional chaos will likely follow, the details depending upon the rate and extent of heating and its broader effects. The destination of emergent stability could then be thought of as a swampy, forested Arctic inhabited by a biota newly arrived from warm temperate regions to the south. But southern ecosystems are also under stress, from ecosystem degradation, biodiversity loss, and climate change [108,109,110,111,112,113,114,115,116,117,118]. So the Arctic cannot be considered in isolation, which raises the question of whether other subsystems of the biosphere are also entering tipping pathways.

2.3 Other potential tipping pathways

Sustained heating within Earth systems interacts with other kinds of stress, making them more vulnerable to climate change, or likely to contribute to it. Several potential tipping elements in the climate system are known [119,120,121,122,123]. They include processes in the Arctic and Antarctic, such as melting ice and permafrosts, and instability of the east and west Antarctic ice sheets. Others involve the weakening or halting of the Atlantic Meridional Overturning Circulation (AMOC), the shifting of the West African and Indian monsoons, and the changing amplitude or frequency of the El Niño Southern Oscillation (ENSO). Linked processes in the ecological Arctic and its boreal zone, such as permafrost melt, decay and fire, have already been mentioned, and other important changes are underway in the ecosystems of the equatorial tropics, especially in the Amazon rainforest.

Considering first the more strictly physical processes, the AMOC is an ocean current that transfers heat from the equatorial to the northern Atlantic. Its chief driver involves sinking of dense saline water around Greenland, so may be compromised by observed rapid melting of the Greenland ice sheet [124]. It is therefore linked to heating in the Arctic, and also, through its relationship with other currents in the Atlantic Ocean, to the West African monsoon. Monsoons depend on events at sea and on land, with the heating ocean and lands competing for influence, while the lands are also changing in vegetation cover (tending to increase albedo) and generating dust and aerosol pollution (tending to reduce solar heating). This makes future monsoons uncertain beyond a trend towards unpredictable rainfall [125]. Changes to the ENSO are similarly hard to predict, although a trend is recognised towards increased variability in rainfall and sea surface temperature, and an eastward shift and intensification of ENSO-related atmospheric connections between the Pacific and the Americas [126]. In the AMOC, monsoon and ENSO cases it is too soon to tell definitively if a tipping pathway has been entered, but they are linked to changing environmental conditions that are undermining the livelihood security of several billion people [127, 128].

Turning to more strictly ecological processes, deforestation in the moist tropics has been sustained, rapid and extensive [129,130,131]. Its drivers lie mainly in the global economy's demand for plantation and ranch produce, coupled with local settlement, exploitation and mining [132]. Large areas of intact moist tropical forest generate much of their own rainfall, thus maintaining themselves in a complex relationship with large-scale seasonal rhythms [133,134,135]. If they are reduced sufficiently, remaining forests may become drier and more prone to fire and disease. In the Amazon, this process is expected to become irreversible once 20–40 percent of its forests have been cleared [136,137,138]. The lower end of this threshold range has already been crossed, and deforestation is accelerating [139]. The increasing frequency and duration of dry weather since the early 2000s are thought to have caused an extensive loss of forest resilience, which is consistent with an approaching transition and a critical threshold of rainforest dieback [140]. Since both the South-east Asian and Amazonian rainforests are strongly affected by ENSO-related rainfall variability [137, 141], the relative contributions of ENSO changes and regional drying due to deforestation are unclear. A compounding effect is likely, with drought amplifying vulnerability to land-use change and vice versa.

The timing is contested [142], but it is possible that a regional breakdown in self-maintaining water and forest systems will soon lead to the Amazonian forests being swiftly replaced by fire-maintained grassland and/or cultivated land and pasture. With a mean carbon content of about 214 t/ha, the 576 million ha of Amazonian forest biomass and necromass that remained in 2000 were estimated to contain about 123 Gt C [143, 144]. By 2020, the area had been reduced to about 530 million ha [145], and the carbon content therefore to about 113 Gt C. This is equivalent to about 415 Gt CO2 [146] or some twelve times total worldwide CO2 emissions in 2020 [147]. If transformation of the Amazon occurs quickly in this century the global climate change effects would be severe, as well as being catastrophic for local biodiversity and peoples.

2.4 Is the whole biosphere entering a tipping pathway?

Evidence reviewed in Sect. 2.2 offers certainty that the Arctic system is rapidly melting, but also uncertainly over the timing and consequences of final sea-ice loss and the scale and timing of CH4 emissions from sea-beds and peatlands. The best fit to available information, taking into account the energy involved in melting water ice and the need to err on the side of caution, is that the Arctic Ocean will become practically sea-ice free in around September 2030, that this can be described as a tipping point, and that it will be swiftly followed by the start of a chaotic transition and the rapid emission of hundreds of Gt of CH4.

Evidence reviewed in Sect. 2.3 offers certainty that the Amazon system is being damaged and that this could disrupt the regional water cycles that sustain the rainforest. But there is uncertainty over the critical threshold of deforestation needed to induce a wide dieback of the forest, and therefore also the timing of transformative change and the release of GHGs from the dying forest. The best fit to available information is that a tipping point is approaching, perhaps in the 2040s or earlier at current deforestation rates, and that this will be followed by a chaotic transition and the rapid emission of hundreds of Gt of CO2.

Regardless of precise timings for these events in the Arctic and Amazon, it is a key point that these processes are underway in an entangled biosphere and climate system, along with ongoing changes to ENSO, AMOC and other ocean currents, major monsoon systems, and other trajectories such as ocean acidification and deforestation in South-east Asia and Africa. There is limited predictive understanding of how such entanglements operate and express themselves in the world as a whole, but their increasing influence can be proxied by annual insured losses (about 30% of economic losses) from weather-related disasters. These averaged about USD 54 billion in 2007–2016 and more than doubled to USD 118 billion in 2017–2021 [148]. The first half of 2022 yielded intense heatwaves, droughts and floods, particularly in Argentina, India, Pakistan, Spain, France and Japan, with insured losses up 18% over the twenty-first century mean [149], even before computing the full costs of the disastrous 2022 northern hemisphere summer (see Sect. 2.2).

Taken together, all lines of evidence and reasoning considered here suggest an approaching biosphere tipping point (BTP), which is a similar concern to that expressed by other authors [150]. The BTP would be the moment when enough Earth systems become committed to transformation that the whole biosphere can no longer resist profound system change, regardless of human efforts thereafter. This might tentatively be dated to the year 2050 ± 10. If there is even a small chance that this is a correct prognosis, it would justify exceptional efforts to prevent the BTP, including the refocusing of mitigation investment strategies to maximise their impact on the drivers of the BTP itself. The rest of this paper explores ways in which these urgent new priorities could be translated into specific kinds of action within the portfolios of public mitigation investors.

3 Preventing the biosphere tipping point

3.1 Mitigation investment goals

Like any other public expenditure, public mitigation investment must be justified politically, in competition with alternatives, then budgeted, scheduled and delivered, and its performance assessed against dated timelines. For mitigation, these things cannot be done effectively by individual governments using a goal like reducing global mean surface temperature, heat content of the oceans, or GHG content of the air, since these are not measurably responsive to any one country's actions. To turn the 1.5 °C temperature goal into a practical objective, therefore, some governments and other institutions have adopted policies or laws to achieve net zero GHG emissions (see Sect. 1). This goal implies a zero sum after taking into account all emission consequences of all relevant actions at all scales. Such a binding schedule requires a target-consistent approach, with planners setting carbon prices and other incentives and rules that are consistent with the time-bound emissions limit [151, 152]. But the opportunity to change outcomes by relieving heat stress on Earth systems may only be available prior to the BTP. This is why it would be wise to take a precautionary approach and consider how mitigation projects can be targeted most cost-effectively on postponing the BTP and ultimately preventing it. This idea will now be explored, first by describing a way to compare results relative to the BTP, and then by using that method to compare the effects of some representative mitigation projects.

3.2 Calculating mitigation value according to an approaching BTP

As well as CO2 and CH4, anthropogenic GHGs include nitrous oxide (N2O) from the breakdown of chemical fertiliser, and sulphur hexafluoride (SF6), the hydrofluorocarbons (HFCs) and perfluorocarbons (PFCs) from industrial processes. The various GHGs do not have the same capacity as CO2 to trap heat: CH4 is about 85 times more potent over two or three decades (but declines to about 20 times over a century); N2O is nearly 300 times more potent; and SF6 and the HFCs and PFCs are more potent still, as well as being very persistent [146, 153, 154]. Taking into account the different potencies of each, all can be converted into a single figure for their collective CO2 equivalence (CO2e). Thus, for example, 35–40 billion tonnes (Gt) of CO2 (GtCO2) were emitted globally in 2020, but these and other GHG emissions made the total 50–60 GtCO2e [113, 147].

Public mitigation investments must be justified at least in part by specifying the quantity of reduced emissions expected of them. But to compare their true effectiveness the timing of any potential BTP must also be accounted for. To reflect rising probabilities in a very complex system with many moving parts and much uncertainty (see Sect. 2.1), this can be represented by an exponential decline in the true biophysical value of GHG savings each year as the BTP is approached. The equation y = a (1-b)x can be used to estimate this value in each year of a project ('y'), with each year's starting value ('a') discounted by a percentage expressed as a decimal ('b') to the power of each year ('x') [155]. The relevant unit of account is created by exponentially discounting annual GHG savings each year at a rate by which the value of each tCO2e saved approaches zero at the expected BTP. The discount rate will determine the pace of divergence between current and discounted emission reductions, and will depend on when a potential BTP is anticipated.

Whatever the rate and date chosen, a unit of biophysical mitigation value will have been created that factors in a dated BTP. This can therefore be called a dated mitigation value (dmv), and GHG emissions can then be expressed in tCO2edmv rather than tCO2e. In this way the expected mitigation effects of an investment might be given, in 2025 for example, as tCO2edmv2025, and each of these would be worth more to mitigation than one tCO2edmv2030, and so on. A BTP in 2050 is assumed in the proof of concept cases that follow, and a discount rate of 10% (0.1) is used because it reduces mitigation value to approach zero then. The choice of 2050 is consistent with an understanding of recent events in major and connected Earth systems, as explained in Sect. 2, along with the sense of widespread and increasing biophysical stress within those systems, and trajectories of change that seem to be converging on a common crisis point in this mid-century. This timing cannot be tested in advance, but climate change may be an existential threat to humanity (see Sect. 1) so a precautionary approach is needed. Should convincing new reasons be found to justify slowing the likely timelines of tipping points in Arctic, Amazon and other major Earth systems, then later dates and different discount rates could be used. Forthcoming elections in Brazil, for example, might lead to favourable political conditions for halting deforestation [156,157,158]. Similarly, the European Union's 2022 decision to strengthen engagement with the Arctic, in recognition that the loss of sea ice and permafrost thawing could trigger tipping points in the climate system [159], might induce a step-change in efforts to resist global heating.

Meanwhile, the choices made here on timing and discounting capture the deadline nature of the BTP while recognising that mitigation efforts will retain some value far into the future, for example to societies rebuilding after a series of global catastrophes. Table 1 shows how to transform the value of one tCO2e in this way. The purpose of doing so is to focus mitigation investments on projects that will have maximum impact in terms of postponing the approaching BTP and then, through further focused investment, preventing it entirely.

Table 1 Exponential decay in the mitigation value of one tCO2e expressed in tCO2edmv
Full size table

3.3 Examples of dated mitigation value analyses

3.3.1 Information sources and units

The Danish climate aid portfolio was presented for evaluation with the four focal countries of Ethiopia, Indonesia, South Africa and Vietnam having already been selected by the Ministry of Foreign Affairs of Denmark. These were chosen "with the aim of sampling developing countries with high and growing emissions where Danish interventions had already contributed much to mitigation and could contribute more, and also where the experience of mitigation work had much to teach." [160]. The three projects that are described below were encountered within these national portfolios. One (Harapan in Indonesia) was of special interest because it was the longest-running site-specific Danish intervention in Indonesia, and represented an important tradition in community-based forest conservation with mitigation benefits that had not been fully documented at the time. Another (Assela in Ethiopia) was of special interest because it was by far the largest investment in the Ethiopian mitigation portfolio, on its own representing 37% of all Danish mitigation grants to all four focal countries over two decades. The third (capacity building in South Africa) was of special interest because it represented an expanding element in the Danish mitigation approach, which was also being applied in Vietnam.

In each of these three projects, net physical emission savings expected in each year were estimated in tCO2e, converted to tCO2edmv, and are simply displayed in Figs. 1, 2 and 3 alongside the approximate costs [161]. Brief project descriptions and references to more detailed information are given in the respective sections. The data used are given in the Additional file 1 Table S1. Three units of account are employed: the mitigation unit is tCO2e; the mitigation effectiveness unit is tCO2edmv; and the mitigation cost-effectiveness unit is tCO2edmv/euro (EUR, €). Figures 1, 2 and 3 show patterns of mitigation investment and returns over 21 years in mitigation effectiveness per unit cost (Σ tCO2edmv/Σ EUR) over the same period.

3.3.2 Harapan rainforest, Sumatra (Indonesia)

Figure 1 is based on the Harapan project in a lowland (< 100 m asl) area of about 100,000 ha that straddles the border between the provinces of South Sumatra and Jambi in the island of Sumatra, Indonesia. This started in 2002 and is a complex story involving many stakeholders and donors, with continuous support from the BirdLife family of conservation charities and with Danish official aid funding in 2011–2018 [162,163,164,165]. The area was set aside in 2008 and 2010 in two Ecosystem Restoration Concessions, a category of land use recognised by the Government of Indonesia, for management by local communities in partnership with the concession holders (PT Restorasi Ekosistem Indonesia, mostly owned by BirdLife). Attributed to their success in addressing encroachment through trust building and government recognition for social forestry managed by indigenous communities and transmigrant community groups, the Harapan forest was still standing and regenerating in 2021, with an Indonesian and international research and conservation presence [166].

Fig. 1
figure 1

Danish public investment (€) and mitigation effectiveness returns (tCO2edmv) from the Harapan rainforest project in Sumatra, Indonesia, showing the large initial saving of GHG emissions that arose in the first year from all parties agreeing to protect and restore the forest, and also the low carbon sequestration rate in regrowing forest

Full size image

Forest protection, regeneration and restoration at Harapan can be used to explore the mitigation value both of avoided deforestation (for intact forest and residual stands of damaged forest) and of natural and assisted regrowth of disturbed forest after protection. The latter can also shed light on the mitigation value of reforestation investments, although here involving the costs of community engagement and alternative livelihoods rather than those of land acquisition, preparation and tree planting. The total amount of carbon contained in the Harapan forest was estimated in 2010 at 10–15 Mt [167]. A new estimate was made in 2021, based on a conservative average of 200 tC/ha for intact tropical lowland rainforests on mineral soils [168,169,170,171,172,173,174,175,176]. This was assumed to be reduced to 125 t/ha by light selective logging and to 75 t/ha by heavy or repeated logging. The approximate distribution of forests in the Harapan area at the start of the project was: 25% pristine (200 tC/ha), 25% lightly logged (125 tC/ha), and 50% heavily logged (75 tC/ha). Carbon sequestration rates in these forests were taken to be about 0.5 tC/ha/yr in pristine forests (25% of the area) and about 2.15 tC/ha/yr in disturbed forests (75% of the area) [177]. Based on these assumptions, the Harapan forests were estimated to contain at least 10 MtC at the start of the project, thus confirming the 2010 estimate, and to have absorbed a total of about 3.5 MtC since the start of the project.

Using these numbers, Fig. 1 indicates the large initial saving of GHG emissions that arose in the first year from all parties agreeing to protect and restore the forest, and also the low carbon sequestration rate in regrowing forest. The former is a retrospective view, and while there are other ways to represent the effect of a conservation decision, this is the most realistic in this case because it describes what actually happened. Carbon sequestered thereafter, at about 175,000 tCO2e/year, is barely visible in Fig. 1 because of the high initial gain in conserved carbon. But it does contribute to the total GHG emissions prevented over 21 years, which amounts to 11.439 MtCO2edmv at a unit gain of 0.40 tCO2edmv/EUR from a total public investment of EUR 28.8 million.

3.3.3 Assela wind farm (Ethiopia)

Figure 2 is based on the the Assela Wind Farm Project in Ethiopia, starting in 2020 [178,179,180]. It is being built at Iteya, between the towns of Adama and Assela in the Oromia Region of Ethiopia, about 150 km south-east of the capital Addis Ababa. A total of about EUR 208 million divided almost equally between grant and loan financing was committed to a project in which GHG emission savings were expected from fossil fuel substitution. The assumed mitigation effects start at zero and are expected to rise as turbines come on line during the five-year construction phase. The cost of maintaining the wind turbines thereafter, at about EUR 400,000 annually, is barely visible in Fig. 2 because of the high initial cost, but it does contribute to total cost of a project that was expected to save a total of 1.2 MtCO2edmv over 21 years, at a unit gain of 0.01 tCO2edmv/EUR.

Fig. 2
figure 2

Danish public investment (€) and mitigation effectiveness returns (tCO2edmv) from the Assela wind farm project and assumed RE substitution for fossil fuels in Ethiopia, showing mitigation effects starting at zero and rising as turbines come on line during the five-year construction phase, and also the cost of maintaining the wind turbines thereafter

Full size image

3.3.4 Energy sector capacity building (South Africa)

Figure 3 is based neither on real experience (as in Fig. 1) nor detailed design and feasibility work using familiar technology (as in Fig. 2). Rather it is inspired by capacity-building partnerships between Danish and developing country institutions, for example with Vietnam since 2009, South Africa since 2013, Indonesia since 2016 and Ethiopia since 2017 [181, 182]. These and other such partnerships involve dialogue and advice on policy and regulation in the energy sector, as the partner country is encouraged and enabled to incentivise RE and EE investment and integrate increased RE generation into its national grid. This scenario assumes no emission savings at first, but these are expected to result later as the government responds to increased awareness of options, and takes advantage of technical and policy guidance. These effects are represented by a small increase in the RE contribution to the national energy supply, incorporating pulsed reforms to represent election cycles, international agreements and policy changes, each with a -0.1% effect on annual national emissions starting at 500 MtCO2e, which is based on South African GHG emissions in 2018. With these assumptions, an investment of EUR 5.0 million was expected to yield 4.4 MtCO2edmv in emission savings over 21 years, at a unit gain of 0.9 tCO2edmv/EUR.

Fig. 3
figure 3

Danish public investment (€) and mitigation effectiveness returns (tCO2edmv) from building capacity in the RE/EE sector in South Africa, showing that emission savings are delayed until choice awareness and technical and policy guidance result in successive reforms with pulsed effects on economy-wide GHG emissions

Full size image

4 Discussion

4.1 The proof of concept cases

The line of enquiry in Sect. 3.3 was prompted by noting that project justifications in the Danish mitigation aid portfolio seldom revealed quantitative estimates of expected net GHG emission changes, an explicit preference for early rather than late mitigation effects, or an appreciation of ecosystem carbon storage and the co-benefits linked to it. These issues were also highlighted in evaluations of the Swiss climate aid portfolio. The 3 Danish projects examined here were representative of different kinds of mitigation project, and close inspection of project documents and progress reports allowed their approximate costs and mitigation benefits to be calculated, and their potential co-benefits to be listed. Hence they offered an opportunity to compare costs and benefits across the projects.

While studying them it became clear that the projects yielded different distributions of costs and benefits over time. This was considered important in view of time running out to address mass extinction, ecosystem collapse and runaway climate change at a global level. It therefore seemed necessary to understand the performance of these projects in relation to a potential deadline driven by deteriorating global environmental conditions. But no practical way was then available to quantify differences in mitigation value between early and late emission reductions. A new unit of account, tCO2edmv, was developed to correct this, and its use resulted in different total GHG savings being calculated with and without discounting for these three projects: 13.5 MtCO2e against 11.4 MtCO2edmv for 'avoided deforestation'; 3.2 MtCO2e against 1.2 MtCO2edmv for 'renewable energy'; and 16.0 MtCO2e against 4.4 MtCO2edmv for 'capacity building' (Additional file 1: Table S1).

This method also results in the projects delivering different amounts of early and therefore more valuable GHG savings per unit cost. Here a 40-fold difference between avoided deforestation and wind farming is particularly noteworthy. This would disappear if only the gain from forest regrowth was included, shedding light on the scale of likely returns from tree-planting. With the assumptions used here, both growing trees and farming wind are more expensive than avoiding deforestation, but the latter is more expensive than capacity building in the energy sector which, however, offers the least predictable gains since it depends on events over which the project has no control. Had this analysis been available beforehand, different investment choices might have been made: (1) community-based avoided deforestation projects might have been appreciated as high-performing mitigation actions; (2) large public wind farm investments might have been reconsidered; and (3) capacity-building partnerships might have been more securely justified and focused for the greatest possible mitigation impact. A more complete picture of the relative value and cost-effectiveness of mitigation strategies can be developed through similar analysis of other mitigation investments in various economic sectors.

4.2 Priorities for preventing the BTP

The UN Secretary-General [183, 184], many scientists [185] and much of the global public [186] are unconvinced that current efforts to mitigate climate change will prevent runaway global heating. Signs of accelerating climate change are causing anxiety, frustration and political turbulence even in relatively wealthy societies, while desperation and violence are increasing elsewhere. There is widespread concern that worse is yet to come, and no sense that the worst has yet been defined. These anxieties are appropriate, since current trajectories of change in various physical and ecological Earth systems suggest that a biosphere tipping point may be reached in the middle years of this century, after which events could move swiftly beyond human control or influence. The BTP thus offers a hard deadline against which to plan, implement, monitor, adjust and accelerate our mitigation efforts. These investments should therefore be selected and their design and performance judged against the standard of how well they perform against this deadline.

In practice, this means valuing early net GHG emission savings much more highly than late ones, while also considering co-benefits. Among the latter, avoided deforestation offers gains for biodiversity conservation and the well-being of indigenous peoples that cannot be obtained in any other way. This and ecosystem restoration offer environmental security from the protection and restoration of water catchments and wetlands. Meanwhile, RE offers reliable power for housing, industry and employment, and capacity building opens doors for a country in the direction of economy-wide green growth. But environmental co-benefits especially are seldom fully recognised in justifying mitigation projects, offering a field of urgently-needed research. Moreover, every country will have a variety of policy priorities, so every donor portfolio and government programme will tend to be diverse.

To address specifically the time-critical nature of mitigation investments in a precautionary way, mitigation investments can be compared using tCO2edmv as the key unit of mitigation effectiveness. Different strategies can have very different returns on investment in tCO2edmv per unit cost, thus offering guidance on the following best ways to postpone and prevent the BTP. (1) The highest single priority is to protect all remaining high carbon-density natural ecosystems globally. This will prevent the most GHG emissions most quickly, and even temporary success is useful since it will take some of the pressure off Earth systems while more permanent solutions are found. (2) Efforts to restore natural ecosystems are also useful, and rich in environmental co-benefits, but slow in mitigation terms relative to protecting intact ecosystems. And (3), net decarbonisation of land use, energy, transport, and other sectors requires whole-economy thinking, improved choice awareness, technical support and capacity building, all of which are needed everywhere if national net zero commitments are to be put into effect.

Public mitigation investment portfolios that emphasise these three strategic elements are most likely to be effective in postponing the BTP, but the Danish and Swiss portfolios offer additional insights. These include: (1) that the most cost-effective way to protect natural ecosystems is often through local communities having secure resource tenure and support in managing their own resources sustainably; (2) that protecting and restoring catchments and wetlands can often best be achieved by arranging for users of their ecosystem services to pay for these actions; and (3) that improved community-based management of lower carbon-density ecosystems such as soils in farmlands and pastures can also contribute to mitigation if large areas are involved. Many proven variants of these and other approaches are known, but the key point is that this experience must be available to mitigation planners, who must also become used to justifying investments using realistic quantitative estimates of their net GHG emission effects. Only then can the necessary strong bias towards the early delivery of large net GHG emission reductions with valuable co-benefits be expressed in practice, creating the possibility of preventing the BTP. This is certainly desirable, since it would give humanity time in which to achieve a 'just transition' to sustainability [187, 188] and to build the foundations of the 'peace with nature' for which the UN Secretary-General has called [189].

5 Conclusions

By the reckoning in Sect. 2, the BTP may now be only 20–40 years away. Preventing it will require vast social, financial and economic investment, so an ability to identify the most cost-effective investments will be useful. This paper describes a method to compare and select mitigation investments according to the unit cost of the dated mitigation value (tCO2edmv) of their outcomes. The tCO2edmv metric is created by exponentially discounting annual GHG savings in tCO2e against a dated BTP, the rationale being that it will be harder and harder to prevent the BTP as it approaches. Three proof of concept cases are described using a BTP in 2050 and a 10% discount rate, highlighting key ways to prevent the BTP. These include (1) protecting and restoring high carbon-density natural ecosystems, and (2) improving choice awareness and building capacity to promote decarbonisation in all economic sectors. Public mitigation portfolios should emphasise these strategic elements, while private ones continue to focus on renewable energy and linked opportunities.

The approach can be explored further through sensitivity analysis with various experimental assumptions, inter alia: (1) about the timing of GHG savings delivery in different projects and in a variety of economic sectors; (2) on likely dates for the BTP; (3) on the discount rate for future GHG emission gains relative to the BTP; and (4) on price inflation and currency exchange rate variations that may affect cost calculations.

Once practitioners have found the most robust and useful assumptions, the tCO2edmv metric can be standardised and will then be useful: (1) in public investments that are intended or likely to have net GHG emission reduction consequences; and (2) in private investments where net GHG emission reductions are rewarded by markets that favour tCO2edmv savings over tCO2e savings, as they should do after some market education. Since all mitigation investments should foresee GHG emission gains, and all should prefer these to be achieved early rather than late, the tCO2edmv metric will be of use across the whole mitigation arena. This will help governments ensure that adequate resources are assigned to all the highest-value mitigation targets in every country, whatever else is done for other purposes.

Meanwhile, three urgent research priorities have also been identified by this study: (1) on the consequences of an Arctic Ocean imminently free of summer sea ice, and options for remedial action including the recapture of methane at large scale and high speed; (2) on testing the tCO2edmv metric with various assumptions to justify precautionary mitigation investments in multiple sectors and contexts; and (3) on integrating diverse co-benefit values into mitigation investment decision making.