Climate extremes likely to drive land mammal extinction during next supercontinent assembly

Anthropogenic greenhouse gas emissions are driving Earth toward a warmer climate state with implications for ecosystem resilience, but it is unclear whether or when terrestrial mammals will encounter a climatic tipping point that threatens their dominance. Prior work suggests that warming could exceed mammalian physiological limits in some regions, rendering them uninhabitable. The study frames mammalian thermal limits via dry-bulb temperature (Td), wet-bulb temperature (Tw ~35 °C as a critical hyperthermia threshold), and Humidex (≥45 dangerous; ≥54 heat stroke), alongside cold-stress thresholds (frost and ≤0 °C limiting freshwater and plant growth). The planetary habitability index (PHI) provides a complementary, astrophysical perspective on habitability. The next supercontinent, Pangea Ultima, is predicted to form in ~250 Myr, coincident with a ~2.5% increase in solar luminosity and potential extremes in atmospheric pCO2 due to tectonics, rifting, and outgassing. The central question is whether the assembly and evolution of Pangea Ultima will force climates that surpass mammalian physiological tolerances long before a classic runaway greenhouse, and how geography, pCO2 and increased insolation jointly control future planetary habitability.

The geologic record shows large swings in atmospheric CO2 over billions of years and apparent coupling between supercontinent cycles (assembly, tenure, breakup) and climate. Past supercontinents (Vaalbara, Kenorland, Nuna, Rodinia, Gondwana/Pangea) experienced climates from panglacial to greenhouse states, with CO2 ranging from ~200 ppm to >2,000 ppm, and even orders-of-magnitude higher during Snowball Earth deglaciations. Large Igneous Provinces and rifting have been linked to CO2 outgassing and climate change. Prior physiological research indicates conserved upper thermal limits in mammals through geologic time, with core constraints at Tw ~35 °C and Td >40 °C leading to mortality, and Humidex providing an indicator for hot-dry stress. IPCC assessments project future exceedance of some heat-stress thresholds regionally this century. Studies also emphasize the role of silicate weathering, tectonic degassing, and moisture pathways in modulating long-term CO2, with feedbacks operating on 10^6-year scales. Astrophysical habitability indices (ESI/PHI) provide a complementary framework to assess Earth-like habitability as a function of planetary parameters and temperature.

Climate modeling: The study employs the UK Met Office HadCM3LB-M2.1aD fully coupled atmosphere–land–ocean GCM (3.75° × 2.5° resolution; 19 atmospheric and 20 ocean vertical levels) with dynamic vegetation (TRIFFID, MOSES 2.1 land-surface scheme) and an interactive ozone scheme responsive to tropopause height. Sea-ice is simulated with a zero-layer model; ocean salinity is adjusted to an ice-free reference. Vegetation–albedo feedbacks include interactive desert soil albedo based on soil carbon. Simulations are initialized from an equilibrated pre-industrial climate and integrated 5,000 model years to equilibrium (surface trend <0.1 °C per century; TOA imbalance <0.3 W m−2), except low-CO2 cases that glaciate. Boundary conditions: A +250 Myr reconstruction of Pangea Ultima (PU) provides land–sea masks, topography, and bathymetry (no terrestrial ice sheets). Orbital parameters are modern. Solar constant is set to 1,364.95 W m−2 (modern) and 1,399.07 W m−2 (+2.5% by +250 Myr). Atmospheric CO2 sensitivity experiments span 0, 70, 140, 280, 560, 1,120 ppm under both solar forcings; an additional experiment doubles orography at 560 ppm with +2.5% solar. Heat/cold stress metrics and habitability: Wet-bulb temperature is computed via accurate thermodynamic methods; Humidex is calculated from temperature and dew point with thresholds at monthly mean Humidex ≥45 (dangerous) and ≥54 (heat stroke). Mammalian habitability is defined by simultaneous satisfaction of: (1) cold-month mean temperature >0 °C for at least three consecutive months; (2) Tw <34.5 °C; (3) monthly mean Humidex <45; and exclusion of desert grid cells (bare soil fraction >0.5) due to limited water/food. Hibernation is implicitly accounted for via the cold-stress criterion; aestivation and expanded dormancy scenarios are explored in sensitivity analysis. Model skill is checked by comparing pre-industrial simulations against observed mammalian species distributions with human influence removed. Energy balance analysis quantifies contributions from albedo, emissivity, and heat transport to temperature changes between configurations and forcings. Astrophysical indices: The Earth Similarity Index (ESI) and Planetary Habitability Index (PHI) are used, with GMAT the only changing parameter for future Earth; habitability is defined at ≥0.8. Carbon cycle modeling: The SCION climate–biogeochemical model estimates long-term background pCO2 for PU by combining 2D spatial weathering informed by the GCM outputs with tectonic CO2 degassing prescribed from a hypothesized PU plate boundary network. Degassing is inferred as 1.3–1.9× present, and ensemble simulations (n=1,000) propagate uncertainty in degassing and silicate reactivity to produce a probabilistic pCO2 range.

Climate states over PU: Under +2.5% solar luminosity, PU grid-weighted global mean annual temperature (GMAT) spans 19.9–27.3 °C across 0–1,120 ppm CO2; land-only GMAT spans 24.5–35.1 °C. Relative to a pre-industrial control (280 ppm; modern Fo), PU GMAT anomalies are +8.2 to +16.1 °C (land-only +12.2 to +29.8 °C). Changing geography alone to PU (280 ppm; modern Fo) warms GMAT by +3.5 °C and land-only temperatures by +13.9 °C due to continentality. Habitability: Pre-industrial Earth has 66% habitable land. PU at 280 ppm (modern Fo) yields 54% habitable land. With +2.5% solar luminosity, habitable land declines to 25% (280 ppm), 16% (560 ppm), and 8% (1,120 ppm). Doubling orography at 560 ppm and +2.5% solar raises habitability marginally from 16% to 19%. Desert area increases from 30% (pre-industrial) to 42–49% in PU scenarios. Heat-stress exceedance expands from the tropics (at 280 ppm with +2.5% Fo) into mid-to-high latitudes by 1,120 ppm; exceedances persist for >30 days in large regions. Aestivation sensitivity increases habitability only slightly: at +2.5% Fo, from 25% to 32% (280 ppm), 16% to 21% (560 ppm), and 8% to 13% (1,120 ppm). Hydrology and aridity: Reduced moisture advection and runoff into continental interiors produce extensive arid zones; at 1,120 ppm, vapor pressure deficit between 50°N–50°S is 3–4× that of the modern Gobi Desert. Energy balance: With +2.5% Fo, GMAT increases are driven directly by insolation (~+1.8 °C) and indirectly via albedo (~+2.7 °C), emissivity (~+3.7 °C), and minor heat-transport changes; desertification enhances terrestrial albedo feedbacks. Equilibrium climate sensitivity (ΔTeq) across 280→1,120 ppm at +2.5% Fo ranges from ~2.3–3.9 °C; with modern Fo it varies more (2.4–4.8 °C), indicating non-linear feedbacks with insolation. Astrophysical indices: No future PU scenario with ≥280 ppm CO2 meets the ESI ≥0.8 habitability threshold; combined ESI–PHI indicates Earth leaves its astrophysically defined habitable zone (temporarily) before a runaway greenhouse. Long-term CO2: SCION predicts a mean background pCO2 of ~621 ppm (range 410–816 ppm) for PU given 1.3–1.9× present degassing and altered continental weatherability. Implications: Predicted background pCO2 combined with +2.5% solar insolation likely pushes Earth across a mammalian habitability tipping point, risking mass extinction. Short-lived CO2 spikes ≥560 ppm (e.g., LIPs, rifting) over 10^2–10^3 years could render Earth inhospitable to mammals. Alternative supercontinent configurations: Most arrangements with tropical centering resemble PU climates; an Amasia configuration centered over the North Pole may reduce heat stress but heighten cold-stress risks and glaciation potential.

The assembly and evolution of Pangea Ultima, together with increased solar luminosity, substantially increases Earth’s sensitivity to atmospheric CO2, driving widespread exceedance of mammalian thermal limits. Geography alone moves the climate toward critical thresholds via enhanced continentality and albedo/emissivity feedbacks; adding +2.5% solar luminosity pushes large areas beyond habitable heat-stress thresholds for sustained periods. Under plausible long-term background CO2 (410–816 ppm) derived from tectonic degassing and weathering feedbacks, global mean temperatures rise sufficiently (∼21–25 °C GMAT) to trigger extensive uninhabitability across the tropics and beyond. Hydrological changes produce continent-scale aridity that acts as a biogeographic barrier, limiting potential migratory refugia and compounding heat stress via limited water availability and elevated VPD. Behavioral and physiological strategies (hibernation/aestivation, nocturnal activity, burrowing) provide only marginal gains under these climates. Weathering feedbacks operate too slowly relative to the time scales relevant for mammalian survival, allowing short to medium-term hyperthermal events to devastate habitability. Astrophysical habitability metrics corroborate that Earth would fail standard exoplanet-based habitability thresholds prior to a true runaway greenhouse, underscoring that continental configuration and atmospheric composition strongly modulate habitability within traditional insolation-defined ‘habitable zones’.

This study integrates fully coupled climate simulations, energy balance diagnostics, physiological heat/cold-stress thresholds, astrophysical habitability metrics, and long-term carbon-cycle modeling to demonstrate that the formation and tenure of Pangea Ultima—under modestly increased solar luminosity—are likely to render large portions of Earth uninhabitable to mammals. With plausible background CO2 (410–816 ppm), critical heat-stress thresholds are exceeded widely, and even brief CO2 pulses ≥560 ppm could trigger mass extinction-level uninhabitability. Continental configuration, atmospheric CO2, and insolation jointly control planetary habitability, implying that Earth-like exoplanets’ surface habitability may depend as much on land distribution and atmospheric composition as on stellar flux. Future work should explore alternative supercontinent scenarios (e.g., Amasia), more comprehensive carbon-cycle–tectonic couplings (including weathering pathway sensitivities and lithology), dynamic biosphere feedbacks, and species-specific physiological and behavioral adaptation limits across time scales relevant to evolutionary change.

Predicting atmospheric CO2 on ~250 Myr time horizons is inherently uncertain due to unknown tectonic configurations, degassing rates, and lithologic/weatherability distributions; SCION estimates depend on assumed plate boundary networks and extrapolated spreading rates. Supercontinent geography is uncertain, and different configurations (e.g., Amasia) could shift the balance between heat and cold stress. Weathering feedbacks are represented in long-term models and may lag climate changes critical for mammals; transient hyperthermals and volcanic events are difficult to constrain. The habitability criteria adopt generalized mammalian thresholds and coarse metrics (monthly means, desert exclusion via bare soil fraction), which may not capture microrefugia or species-level variability (e.g., juveniles, small burrowers). The GCM resolution and parameterizations, including vegetation and ozone schemes, impose structural uncertainties; energy balance decompositions simplify complex feedbacks. Short-term perturbations and human influences are excluded by design, and validation relies on pre-industrial benchmarks with human impacts removed.