Habitable zones versus photosynthetic habitable zones

The sunlight energy stored in coal is the product of oxygenic photosynthesis carried out by vast forests within the low elevation swamps of the supercontinent, Pangea. Before addressing the question of the prevalence of coal in the galaxy, we first need to determine the prevalence of oxygenic photosynthesis on exoplanets located within the habitable zones (HZ) of their stars, defined as the distance where liquid water can exist on planets with near-circular orbits (Hart, Reference Hart1979; Kasting et al., Reference Kasting, Whitmire and Reynolds1993; Kopparapu et al., Reference Kopparapu, Ramirez, Kasting, Eymet, Robinson, Mahadevan, Terrien, Domagal-Goldman, Meadows and Deshpande2013; Ward and Brownlee, Reference Ward and Brownlee2000). Although anoxygenic photosynthesis plays critical roles in diverse ecosystems by driving nutrient cycles and supporting microbial communities, it is restricted to various bacteria and archaea and did not contribute directly to the formation of major terrestrial ecosystems (Sleep and Bird, Reference Sleep and Bird2008).

Covone et al. (Reference Covone, Ienco, Cacciapuoti and Inno2021), following the exergetic arguments of Scharf (Reference Scharf2019), argued that oxygenic photosynthesis should be very common on other Earth-like planets. The question is, how many of the known exoplanets are sufficiently Earth-like for oxygenic photosynthesis to evolve? To answer this question Covone et al. analyzed the photon flux, exergy and exergetic efficiency of the radiation in the wavelength range useful for oxygenic photosynthesis as a function of host star effective temperature and planet–star separation. They concluded that, as of 2021, none of the observed terrestrial exoplanets were sufficiently comparable to Earth to support the evolution of oxygenic photosynthesis. However, this finding may have been due, at least in part, to a selection effect, since it is very difficult using existing methods (solar transits and radial velocity) to find Earth-like planets at Earth’s distance from the sun.

Since the formulation of the Drake Equation, most searches for life on exoplanets have tacitly assumed that the HZ is equally hospitable to life at every trophic level, both photosynthetic and non-photosynthetic. This approach has led to the neglect of parameters that are specific for photosynthesis. Recently, Hall et al. (Reference Hall, Stancil, Terry and Ellison2023) introduced a new term, the photosynthetic habitable zone (PHZ), defined as the distance from a star where both liquid water and oxygenic photosynthesis can exist. In principle, the PHZ can either be identical to the HZ or restricted to a narrower band within the HZ. Based on their analysis, Hall et al. concluded that under the most ideal conditions for life the PHZ is nearly as broad as the HZ. However, if the environmental parameters – including photosynthetically active radiation (PAR), atmospheric attenuation, greenhouse effects, stellar lifetime, orbital eccentricity and planetary daylengths or rotation speeds – are even slightly suboptimal, the PHZ becomes significantly narrower than the HZ and concentrated at greater distances from their more massive stars (Hall et al., Reference Hall, Stancil, Terry and Ellison2023). According to their calculations, of the 29 out of ∼5,000 exoplanets thought to be in the HZ, only five might be in the PHZ (from 0.6% down to <0.1%).

Previous studies by Lehmer et al. (Reference Lehmer, Catling, Parenteau and Hoehler2018) and Lingam and Loeb (Reference Lingam and Loeb2019) had established that M dwarf stars, which emit very low light levels due to their low temperature, have very limited emissions in the region of the spectrum (400 < λ < 700 nm) which brackets the PAR of green plants. On the other hand, laboratory studies by Battistuzzi et al. (Reference Battistuzzi, Cocola, Claudi, Pozzer, Segalla, Simionato, Morosinotto, Poletto and La Rocca2023a) indicated that cyanobacteria are able to photosynthesize normally under simulated M dwarf irradiation. However, Battistuzzi et al. (Reference Battistuzzi, Cocola, Liistro, Claudi, Poletto and La Rocca2023b) also found that both the nonvascular bryophyte moss Physcomitrium patens and the vascular plant Arabidopsis grew more slowly under simulated dwarf irradiation, and Arabidopsis exhibited a shade avoidance response due to the lower amounts of light and the lower red/far red ratio under M dwarf light conditions. As will be discussed later, coal on Earth was formed exclusively from vascular plants, and there is no evidence of coal formation from either cyanobacteria or bryophytes.

Recently, Chitnavis et al. (Reference Chitnavis, Gray, Rousouli, Gillen, Mullineaux, Haworth and Duffy2024) have suggested that the different responses of cyanobacteria and green plants are due to fundamental differences in their PSII and PSI antenna complexes. The large phycobilisome antenna complexes of cyanobacteria function on the membrane surface and act to funnel energy to the reaction centers. In contrast, the smaller antenna complexes of bryophytes and vascular plants are embedded in the chloroplast membranes and do not act as energy funnels. Chitnavis et al. conclude that hypothetical M dwarf ecosystems would therefore be dominated by cyanobacterial-type microbes and perhaps a few nonvascular plants, rather than the coal-forming forest ecosystems that evolved on Earth.

More recently, Vilović et al. (Reference Vilović, Schulze-Makuch and Heller2024) examined the effects of simulated light from K dwarf stars, which emit much more light in the visible range than M dwarfs. They found that both cyanobacteria and garden cress exhibit comparable photosynthetic efficiency under simulated K dwarf irradiation compared with artificial solar conditions, indicating that the K dwarf spectrum may not be limiting for green plant oxygenic photosynthesis. However, other parameters such as mineral nutrients or temperature might limit photosynthesis on actual K dwarf planets.

Unique properties of the PSII water-oxidizing enzyme

Another frequently discussed barrier to the evolution of oxygenic photosynthesis on exoplanets arises from thermodynamic and biochemical considerations. The PSII water-oxidizing enzyme is biologically unique in multiple ways (Fischer et al., Reference Fischer, Hemp and Johnson2016; Hohmann-Marriott and Blankenship, Reference Hohmann-Marriott and Blankenship2011; Oliver et al., Reference Oliver, Kim, Trinugroho, Cordón-Preciado, Wijayatilake, Bhatia, Rutherford and Cardona2023; Tan et al., Reference Tan, Liu, Jiao, Li, Hu, Lv, Qi, Li, Rao, Qu, Jiang, Soo, Evans, Hua and Li2024):

1. 1. It is the only currently known or modeled enzyme capable of splitting water into molecular oxygen, protons and electrons using sunlight.

2. 2. It contains a Mn₄CaO₅ cluster arranged in a cubane-like structure, which has no biological parallel in its composition or role in water oxidation.

3. 3. The reaction proceeds via the Kok cycle, a stepwise mechanism involving five intermediate states (S₀ to S₄). No other enzyme employs this multi-step, photon-driven mechanism for water oxidation.

4. 4. It undergoes frequent photoactivation to replace damaged components (e.g., the D1 protein) and reassemble the Mn₄CaO₅ cluster, a process absent in other metalloenzymes.

5. 5. No analogous natural enzyme has been discovered in non-photosynthetic organisms or abiotic systems, and artificial catalysts inspired by PSII remain inferior in efficiency and self-repair capability.

6. 6. Another unique feature of PSII is that it functions in a complex with a second photosystem, PSI, to lift electrons by a two-stage process from a low redox level (H₂O, +820 mV) to a high redox level (NADPH, −320 mV).

Although it is conceivable that a simpler type of oxygenic photosynthesis might evolve based on a single photosystem utilizing higher energy photons at the blue end of the spectrum, the fact that, as far as is known, no such reaction evolved or survived natural selection on Earth suggests that there may be thermodynamic obstacles or functional disadvantages to such a mechanism.

Was the evolution of PSII “hard” or “easy”?

Given the complexity and many unique features of the PSII water-oxidizing enzyme, one might infer that chance played an outsized role in PSII evolution. It is widely accepted that the great oxidation event (GOE), which began around 2.4 Gya, was caused almost entirely by cyanobacterial oxygenic photosynthesis (Davin et al., Reference Davin, Woodcroft, Soo, Morel, Murali, Schrempf, Clark, Álvarez-Carretero, Boussau, Moody, Szánthó, Richy, Pisani, Hemp, Fischer, Donoghue, Spang, Hugenholtz, Williams and Szöllosi2025; Fournier et al., Reference Fournier, Moore, Rangel, Payette, Momper and Bosak2021). The precise date of origin of PSII remains contentious. Based on molecular clocks, the Last Universal Common Ancestor (LUCA) evolved toward the end of the Hadean, between 4.33 and 4.09 Gya (Westall, Reference Westall2025). Phylogenetic studies indicate that the last common bacterial ancestor evolved around 4.0–3.5 Gya (Wang and Luo, Reference Wang and Luo2025), while molecular clock and geological analyses together suggest that oxygenic photosynthesis arose around 3.5–2.8 Gya (Westall, Reference Westall2025). Unfortunately, the uncertainties surrounding these dates do not lend themselves to a determination of whether the evolution of oxygenic photosynthesis was “easy” or “hard.”

However, if the earlier estimates for origin of PS are correct, a delay of about a billion years occurred between the origin of oxygenic photosynthesis and the GOE. Proponents of this idea argue that the gap may reflect the time it took to saturate all of the oxygen sinks present on the early Earth, while skeptics believe the sinks would have been saturated in a much shorter time. Alternatively, competition between anoxygenic and oxygenic of photosynthesis may provide at least a partial explanation. Modeling studies suggest that anoxygenic photosynthesizers, which use electrons from reduced species like Fe(II) and H₂ instead of water, may have competed with early oxygenic photosynthesizers for essential nutrients and light, thus diminishing the amount of oxygen released prior to the GOE (Ozaki et al., Reference Ozaki, Thompson, Simister, Crowe and Reinhard2019). More recently, Horne et al. (Reference Horne, Goldblatt and Kump2025) used a new biogeochemical model to study interactions between oxygenic photosynthesis, nitrogen fixation and inorganic phosphorus cycle availability as a constraint on the timing of oxygenation. Their counter-intuitive finding was that the earlier oxygenic photosynthesis and nitrogen fixation arise, the longer the delay before the GOE, with phosphorous availability at the ocean surface serving as the limiting factor. Furthermore, strong ocean redox stratification and efficient phosphorus sequestration perpetuate the bottleneck and limit primary productivity and O₂ generation (Horne et al., Reference Horne, Goldblatt and Kump2025).

The plant origin of coal

The earliest stage of coal formation is peat, which is composed of partially decayed plant material in which fragments of vascular plants (e.g., stems, leaves and roots) can often still be recognized (Moore, Reference Moore, PC and B1989; Orem and Finkelman, Reference Orem, Finkelman and Mackenzie2003). The presence of plant fragments in peat and lower-grade coals, combined with chemical analyses, provides unequivocal evidence for the plant origin of coal. Even if the Precambrian Earth was covered with vast sheathes of microbial mats, as inferred from the preservation of Ediacaran fossils (McMahon et al., Reference McMahon, Matthews, Brasier and Still2021), there is no evidence that any coal was formed at this time. Since the rise of vascular plants in the late Silurian period (∼430 Mya), however, coalification has been going on more or less continuously, albeit at different rates and in different geographic regions (Butler et al., Reference Butler, Marsh and Goodarzi1988; Cleal and Thomas, Reference Cleal and Thomas2005; Dexin, Reference Dexin1989; Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016; Orem and Finkelman, Reference Orem, Finkelman and Mackenzie2003).

When, where and how much coal was formed

Two key factors – plate tectonics and climate – have been cited as the primary determinants of when, where and how much coal is deposited (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016). In this section, we will first outline the basic mechanism of coal formation. Next we will address the question of why so much coal was produced during the Carboniferous/Permian periods, as well as during the more geographically restricted second peak of coal deposition in the Late Mesozoic and Cenozoic eras. Finally, we will discuss the implications of the various contingent factors we have outlined for coal formation on other planets.

Although the basic mechanism of coal formation has been understood for decades, literature estimates of the amounts and rates of coal deposited at different times and geographic regions have varied widely (Bao et al., Reference Bao, Hu, Scotese, Li, Guo, Lan, Lin, Yuan, Wei, Li, Man, Yin, Han, Zhang, Wei, Liu, Yang and Nie2023; Cleal and Thomas, Reference Cleal and Thomas2005; Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016). Presently, there appear to be four major points of general agreement:

1. 1. Relatively little coal was formed during the Devonian period (419–359 Mya), prior to the advent of the tall arborescent lycophytes (spore-bearing trees related to clubmosses) that dominated the coal swamp forests of the Carboniferous.

2. 2. The vast majority of the coal that fueled the Industrial Revolution in England, Europe and North America (perhaps as much as 70–90%) was deposited during a ∼70 million-year window spanning the Carboniferous and Permian periods (330–260 Mya), as the supercontinent Pangea was being assembled (Lane Reference Lane2002; McGhee Reference McGhee2018). The arborescent lycophytes, tree ferns, seed ferns and giant horsetails that dominated the Pangean coal swamp forests of the Carboniferous became extinct as the climate grew drier and hotter, replaced by broad-leafed gymnosperms, such as cycads, Ginkgo and Gnetales. However, the tropical peat-forming biomes responsible for a substantial drawdown of CO₂ were lost by the end of the Permian.

3. 3. The Triassic (252–201 Mya) “coal gap” was a period of little or no coal formation associated with the catastrophic Permian/Triassic extinction event, or Great Dying, caused by the massive volcanic eruptions of the Siberian Traps (Retallack et al., Reference Retallack, Veevers and Morante1996). Along with the majority of marine species, terrestrial vertebrate species and insects, the P/T extinction devastated Earth’s forests, greatly reducing plant productivity (Cascales-Miñana et al., Reference Cascales-Miñana, Diez, Gerrienne and Cleal2016). In the absence of peatification and carbon burial, atmospheric CO₂ levels rose, causing a super greenhouse effect that may have prolonged the P/T Great Dying event by five million years  (Xu et al., Reference Xu, Yu, Yin, Merdith, Hilton, Allen, Gurung, Wignall, Dunhill, Shen, Schwartzman, Goddéris, Donnadieu, Wang, Zhang, Poulton and Mills2025). So profound was this extinction among plants that river systems changed from dominantly meandering, as they are today, to braided river systems due to the extinction of larger trees necessary for bank stability (Ward et al., Reference Ward, Montgomery and Smith2000).

4. 4. During the Upper Cretaceous period of the Mesozoic era and the Paleogene, and Neogene periods of the Cenozoic era (∼100–20 Mya) a second broad peak of elevated coal formation occurred (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016; Orem and Finkelman, Reference Orem, Finkelman and Mackenzie2003). These younger deposits consist largely of lignite and sub-bituminous coals, which are less energy dense than the bituminous and anthracite coals of the Carboniferous and hence less suitable for industrial applications (Han et al., Reference Han, Zhou, Zhang and Li2022; Mills, Reference Mills2016).

Coalification is a multi-step process

The conversion of plant biomass to coal is a multi-step process, leading to several different grades of coal, ranging from the lowest carbon content (lignite, 25–35%) to the highest carbon content (anthracite, 86–97%) (Flores, Reference Flores2014). It is useful to understand coalification in terms of the subdiscipline of taphonomy. The taphonomy of coal involves two main phases: biostratinomy, the events between the death of the tree and its burial; and diagenesis, the process whereby the buried plant debris undergoes chemical change through heating and pressure (Lyman, Reference Lyman2010). Each step in the overall process has its own conditions and time frames. Only a few of the many possible pathways bring about the removal of water, oxygen and hydrogen that is required to produce carbon-rich (and thus energy-rich) coal.

The first step, peatification, is the conversion of plant biomass to peat. But for peat to accumulate, the rate of plant biomass production must exceed the rate of microbial decay (Krausel, Reference Krausel and Nairn1964; Schopf, Reference Schopf, Tarling and Runcorn1973). The thickest coal seams are produced by tropical wetlands (coal swamps or mires) which provide the anoxic, water-logged environment in which microbial decay by aerobic bacteria is inhibited. However, long-term accumulation requires rapid burial through abundant sedimentation, enhanced by crustal subsidence, to avoid erosion (Schopf, Reference Schopf, Tarling and Runcorn1973; Taylor et al., Reference Taylor, Teichmüller, Davis, Diessel, Littke and Robert1998).

The plate tectonic process known as subduction (in which one tectonic plate slides beneath another plate) is not directly involved in coal formation, although it can influence coal formation indirectly by creating basins adjacent to subduction zones (foreland basins) and by augmenting subsidence and metamorphism (Butler et al., Reference Butler, Marsh and Goodarzi1988). In the case of Earth’s major coal formation period, the Carboniferous and Early Permian periods (∼350–270 Mya), the process of plate tectonics and continental drift were crucial in producing the down-dropped basins (such as the modern-day basin and range topography of western North America) where plant growth and accumulation and then burial occurred. The roles of temperature and pressure during burial will be discussed later in the section. However, it appears likely that without the plate tectonic processes that occurred from the middle to the end of the Paleozoic Era, and near the end of the Mesozoic, the amount of coal on Earth would have been minuscule compared to today’s coal reserves.

The requirements for the three major coal types are as follows (Flores, Reference Flores2014):

1. 1. Lignite formation requires the burial of peat under sediment at shallow depths (up to 1 km), which initiates compaction through water loss and additional microbial decay by anaerobic bacteria.

2. 2. Bituminous coal forms at depths of 1.7 to 6 km, which raises the temperature to 50–200^oC.

3. 3. Anthracite coal requires burial at 6+ km, at temperatures in excess of 200^oC.

The amount of time required for the formation of each grade of coal depends on the temperatures achieved at each stage of the process. Conversion of peat to lignite takes ∼1–10 Myr; conversion of lignite to bituminous coal takes ∼10–100 Myr; and conversion from bituminous to anthracite takes ∼100–300 Myr (Flores, Reference Flores2014). Because of its prolonged incubation period, anthracite coal is extremely rare, comprising about 1% of the total world coal reserves. Given enough time at the required temperature, however, the younger coal deposits of the Mesozoic and Cenozoic eras could eventually be transformed into higher grades of coal.

The role of plate tectonics in coalification

Plate tectonics – the movement and recycling of Earth’s crust by convection currents within the mantle – has played a profound role in shaping the evolution and diversity of life on our planet, particularly during the Carboniferous (Montañez, Reference Montañez2016; Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016). The transition from a “stagnant lid” tectonic regime to modern plate tectonics in the Neoarchaean era, ∼2.5 Gya (Holder et al., Reference Holder, Viete, Brown and Johnson2019; but see Marshall, Reference Marshall2024), is ultimately linked to the rapid diversification of complex life, particularly during the Cambrian explosion. Subduction and volcanism support life by recycling carbon, nitrogen, phosphorus and other key nutrients from the Earth’s interior to the surface (Stern and Gerya, Reference Stern and Gerya2024).

Plate tectonics has also played a central role in coal formation (Butler et al., Reference Butler, Marsh and Goodarzi1988; Cleal and Thomas, Reference Cleal and Thomas2005; Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016; Orem and Finkelman, Reference Orem, Finkelman and Mackenzie2003). The Variscan orogeny, or mountain-building event, that created the vast coal swamps of the Carboniferous involved the complex assembly and collisions of microplates governed by a combination of deterministic geological forces and some elements of chance, which ultimately produced the supercontinent Pangea. These tectonic collisions brought about the formation of the 10,000 km Variscan-Alleghanian-Ouachita mountain belts between 480 and 250 Mya.

Specifically, the widespread coal deposits of both current day Appalachia and western Europe were aided by a combination of rapidly down-dropping basins next to rapidly rising mountains, creating conditions of sedimentation so rapid that the shallowly rooted, and then fallen trees were rapidly buried, and thus creating vast reservoirs of reduced organic carbon that subsequently became the high grade coals of North America and Europe. Plate movement and plate collision were thus fundamental to the greatest high grade coal accumulation events on Earth. That the level of plant evolution and a stochastic result of global tectonics intersected in this way was itself improbable.

Globally, the immense crustal thickening associated with the growing mountain belts that were a consequence of the Late Paleozoic supercontinent formation caused downward flexural bending of the lithosphere ahead of the mountain belts, forming vast shallow basins in the equatorial region. Such low-lying sedimentary basins provided an ever-wet, ever-warm climate ideal for dense plant growth and peat formation, as exemplified by the Appalachian and European basins. However, long-term peat accumulation requires ongoing crustal subsidence to ensure continued deposition, while minimizing erosion (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016; Schopf, Reference Schopf, Tarling and Runcorn1973). For these reasons, continental flexures are commonly associated with coal-bearing deposits, as their rates of subsidence and coal accumulation can be roughly comparable (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016).

The key role of climate oscillations in coalification

Although the equatorial region of Pangea is often thought of as a vast, unchanging tropical swamp forest, it experienced regular dry intervals in which xerophytic vegetation displaced wetland flora and became dominant over most of the central and western Pangean supercontinent (DiMichele, Reference DiMichele2014; Horton et al., Reference Horton, Poulsen, Montañez and DiMichele2012; Montañez et al., Reference Montañez, McElwain, Poulsen, White, DiMichele, Wilson, Griggs and Hren2016). The equatorial wetland and dryland biomes oscillated during single glacial-interglacial cycles, which were primarily restricted to the southern hemisphere near the South Pole.

During the glacial periods, lower sea levels exposed continental shelves, creating vast equatorial wetlands (peat mires) in foreland basins of the Central Pangean Mountains. Despite the cooler global temperatures, equatorial regions remained warm and humid, sustaining the growth of dense stands of arborescent lycophytes and tree ferns. Rainfall also increased, and the resulting waterlogged, anoxic swamps and mires allowed organic matter to accumulate without fully decomposing.

During interglacial periods, global temperatures increased, and rising sea levels inundated the swamps and peat mires, burying organic material under sediment (DiMichele, Reference DiMichele2014). Repeated cycles of exposure and submersion over the course of many glacial-interglacial cycles led to layered coal deposits. Such climate oscillations were a key factor leading to the extraordinary abundance of Carboniferous coal, and becomes yet another variable for assessing the prevalence of coalification on exoplanets.

Coal accumulation: Plant productivity versus temporal stability

As previously noted, organic matter can only accumulate when plant biomass production exceeds decay. According to one theory, the tall lycophytic trees that dominated the Carboniferous coal swamps grew very rapidly and were short-lived (10–15 years), leading to a rapid turnover of the population. In this scenario, biomass accumulation was driven by the productivity side of the equation (Phillips and DiMichele, Reference Phillips and DiMichele1990). Alternatively, Boyce and DiMichele (Reference Boyce and DiMichele2016) have provided compelling evidence that the lycophytic trees of the Carboniferous coal swamps grew very slowly, with lifespans of 100–200 years. Under low productivity conditions, it was essential for microbial decay to proceed slowly enough to allow accumulation to occur. The coal swamps and mires of Pangea appear to have minimized oxidative decay by creating a stagnant, waterlogged, anoxic substrate for the preservation of organic plant debris. What was remarkable about the Carboniferous was not the high rates of productivity of the coal forests, but the temporal stability of the subsidence-driven deposition of peat throughout the vast, equatorial region of Pangea, allowing coal seams to thicken over hundreds of millions of years (Boyce and DiMichele, Reference Boyce and DiMichele2016; Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016).

A second peak of coal production

A second, smaller peak of coal production occurred during the Cretaceous, Paleogene and Neogene periods that was comparable in terms of deposition rates to the Carboniferous, although much more restricted geographically. During this time, “regional coal accumulation rates…approximated those of the Carboniferous, albeit those coal accumulation rates were not integrated over so extensive a geographic area globally as in the Carboniferous” (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016).

However, the regional tectonic processes that produced this later peak of coal are similar to those that occurred during the Carboniferous. During the Cretaceous, rapid plate movement led to the breakup of Pangea, the formation of new ocean basins, and the flooding of continental interiors by the high seas (Nelsen et al., Reference Nelsen, DiMichele, Peter and Boyce2016; Scotese et al., Reference Scotese, Hay, Wicander, Boucot, van der Voo, Hart, Batenburg, Huber, Price, Thibault, Wagreich and Walaszczyk2025). In western North America, the Western Interior Seaway inundated vast regions, creating extensive coastal marshes, back-barrier swamps and inter-deltaic wetlands along its shifting shorelines. The Cretaceous was characterized by a warm, humid climate, with CO₂ concentrations reaching as high as 650–2,000 ppm (Scotese, 2025). The dominance of lignin-rich woody gymnosperms allowed coal-forming plants to thrive not only in tropical but also in temperate latitudes (40°–80°N and S), expanding the potential for coal deposition as continents moved northward (Nelsen, 2016).

The Cenozoic era was characterized by major mountain-building events, such as the Laramide orogeny in North America, which uplifted the Rockies and created deep basins (e.g., the Powder River Basin) (Flores, Reference Flores1986, Reference Flores and Raynolds2003; McCabe and Parrish, Reference McCabe and Parrish1992; Morley, Reference Morley1999; Nichols, Reference Nichols1995). These basins experienced significant subsidence, allowing thick accumulations of peat and coal in wetlands, lake margins and river floodplains. The continued movement of continents into higher northern latitudes increased the area available for coal-forming environments, especially in the rainy zones of the Northern Hemisphere. Offshore tectonic activity also created coal-bearing basins on continental shelves, as seen in the Tertiary of the East China Sea and offshore regions of Asia and Russia (Scotese, 2025). The climate of the early Cenozoic remained warm, supporting extensive peatlands. As the era progressed, global cooling and the development of more seasonal climates shifted coal-forming environments to higher latitudes and influenced the types of vegetation contributing to peat. Angiosperms became dominant during the Cenozoic, diversifying the flora from which younger coals developed.

Contingency, determinism and the two-phase “evolution” of coal

In his magnum opus, The Structure of Evolutionary Theory (2002), published shortly before his untimely death, Stephen Jay Gould defined contingency as “the tendency of complex systems with substantial stochastic components, and intricate nonlinear interactions among components, to be unpredictable in principle from full knowledge of antecedent conditions, but fully explainable after time’s actual unfoldings” (Gould, Reference Gould2002). In his earlier book, Wonderful Life: The Burgess Shale and the Nature of History (1990), Gould had argued that the history of life has been strongly shaped by contingency, ruling out any predetermined or inevitable path towards increasing complexity or adaptation. He pointed out that many of the animal phyla that arose during the Cambrian Explosion became extinct, not because they were inferior or less well-adapted to their environments, but because of contingent events. According to Gould, evolutionary history consists of:

Is Gould’s conception of biological evolutionary history applicable to coal formation? Zachary Blount (Reference Blount2016) restated Gould’s provocative argument as follows:

According to Blount, determinism in evolution derives from “[t]he power of natural selection to find the limited set of high-fitness solutions to the challenges imposed by environments” (Blount et al., Reference Blount, Lenski and Losos2018).

Convergent evolution has often been cited as evidence for evolutionary determinism, as in the case of cacti and euphorbias, whose phenotypic resemblances have arisen from the fact that they are both adapted to arid conditions (Orgogozo, Reference Orgogozo2015). Convergent evolution is path-independent, meaning that it is relatively independent of its starting conditions. However, convergence is neither inevitable nor deterministic. Some populations may simply lack specific genes that are required to evolve certain adaptive traits, creating evolutionary barriers to convergence. For example, Antarctic fish have lost genes required for heat tolerance, preventing them from ever adapting to warmer waters (Sackton and Clark, Reference Sackton and Clark2019). Nevertheless, the phenomenon of convergent evolution raises the possibility that, if the tape of the Carboniferous were replayed, at least some of the descendants of the vascular plants that evolved during the Middle Silurian period would be able to adapt to the tropical wetlands of Pangea and give rise to coal. This assumes, of course, that the tectonic and climate conditions that prevailed throughout the Carboniferous remained essentially the same during the replay.

What makes the history of coal so interesting is that its evolution is biphasic. It begins as a living organism, subject to both Darwinian natural selection and Gouldian contingency, but attains “maturity” as a nonliving rock, subject to a multitude of contingent factors. Unlike biological evolution, the physico-chemical process of coalification is highly dependent on its initial starting conditions and is thereafter subject to (in Gould’s words), “a staggeringly improbable series of events,” which are “utterly unpredictable and quite unrepeatable.” Indeed, the further back in time we rewind the tape of Earth’s history, the greater the number of contingent events must be repeated to obtain the same or similar result when time’s tape is replayed, and the less predictable the outcome.

On the positive side, the number of stars in the Cosmos (and thus planets) is so large that even the seemingly lowest probability events, or sequence of events, must have been repeated a large number of times. The question is, did it (and does it) occur frequently enough to allow back-and-forth communication between neighboring ATCs? Are there any exoplanets within even a hundred light years of our own planet where an Industrial Revolution and the construction of interstellar communication was made possible?

The nexus of coalification and intelligence

Nearly all of the bituminous coal used in the Industrial Revolution was derived from deposits formed during the Carboniferous period between around 330 and 260 Mya (Lane, Reference Lane2002). Given that the conversion of peat to bituminous coal takes roughly 10 to 110 Myr, the majority of Carboniferous coal would have reached the bituminous stage between around 320 and 150 Mya. Although complex life has existed on Earth since at least 540 Mya, the first species sufficiently intelligent to have the potential to harness the energy of coal was Homo sapiens, which did not evolve into its modern form until the Pleistocene Epoch, ∼10,000 years ago. However, it was not until the rise of agriculture during the Holocene around 12,000 years ago that complex civilizations capable of exploiting natural resources on a large scale emerged. On a geologic time scale, the Industrial Revolution of the 18^th and 19^th centuries is barely resolvable from the present.

Fortunately, or unfortunately depending on one’s point of view, Homo sapiens evolved long after Carboniferous coal had reached the bituminous and anthracite stages and could be profitably mined and combusted in steam engines to jump start the Industrial Revolution. The smaller amounts of coal that formed during the second peak of coalification in the Cenozoic, being younger deposits, are mostly composed of lower rank lignite and sub-bituminous coals (Dai et al., Reference Dai, Arbuzov, Chekryzhov, French, Feole, Folkedahl, Graham, Hower, Nechaev, Wagner and Finkelman2022). Lignite is characterized by high water content, high ash content, high volatile content and low calorific value. It is also prone to spontaneous combustion and weathering, making storage and long-distance transportation problematical. These characteristics make it a dirtier fuel and far less productive and profitable (Han et al., Reference Han, Zhou, Zhang and Li2022). Because sub-bituminous coal has a lower carbon content and higher water content, it, too, produces less heat and is less efficient as a fuel. Sub-bituminous coal is also considered a non-coking coal, meaning that it does not soften, swell and resolidify into a hard, porous mass when heated, the property needed for metallurgical coke. For this reason, it is unsuitable for blast furnaces for steel manufacture (Mills, Reference Mills2016). Thus, although sub-bituminous coal could have provided energy for home heating purposes and some industrial activity in the 18^th and 19^th centuries, it could not have led to the extraction of oil and gas in the 20^th century because of its non-coking properties.

As luck would have it, the maturation of large amounts of energy-dense bituminous coal preceded the evolution of Homo sapiens by more than a 100 Myr, in time to spark the Industrial Revolution. This might not have happened if humans (or some other highly intelligent species) had evolved much earlier, say, in the Carboniferous, Permian or Triassic periods, before Carboniferous coal had progressed from peat to bituminous. Mammalian species have finite evolutionary lifespans (1–10 Myr), so it is unlikely that, say, a Permian version of Homo sapiens could persist for another ∼200 Myr until Carboniferous coal matured. We therefore propose that the need for synchronicity of bituminous or anthracite coal and intelligence represents a temporal contingent factor that must align with the other biotic and physical contingent factors we have already discussed for an ATC to arise on an Earth-like planet.