Biodiversity as Biochemical Equilibrium
A monograph on thermal biology, metabolic scaling, and the thermodynamic foundations of biodiversity in a changing climate
Biodiversity as biochemical equilibrium
Biodiversity is not merely a catalogue of species. It is the macroscopic manifestation of an underlying biochemical equilibrium — a dynamic steady state in which thousands of metabolic networks, enzymatic cascades, and thermodynamic constraints balance one another across space and time. When we observe that the tropics harbour more species than the poles, we are witnessing the outcome of a thermodynamic gradient: the equator receives roughly 2.5 times more solar energy per unit area per year than 60°N, and this energy difference cascades through every layer of biochemistry, from the folding rates of proteins to the speciation rates of entire clades.
The latitudinal diversity gradient — the most ancient and robust pattern in ecology — is, at its deepest level, a consequence of the Boltzmann distribution. Organisms are chemical machines. Their rates of growth, reproduction, and mutation all depend on temperature through the Arrhenius relation. In warmer environments, metabolic rates are higher, generation times are shorter, and the molecular clock ticks faster. More mutations per unit time mean more raw material for selection; shorter generation times mean faster adaptation. The result is an exponential increase in species richness with temperature, observed across taxa from foraminifera to flowering plants.
This monograph develops the idea that biodiversity is best understood as a chemical equilibrium — one governed by the same thermodynamic principles that govern the equilibrium of any chemical system. We shall see that the free energy of protein folding sets the thermal limits of life, that the Michaelis-Menten equation determines how enzyme kinetics respond to warming, and that the metabolic theory of ecology provides a unified framework for predicting how species richness scales with temperature, body size, and energy availability. Along the way, we shall encounter the biochemistry of photosynthetic pathways (C3, C4, CAM), the thermoregulatory strategies of endotherms and ectotherms, and the population dynamics of climate-driven ecosystems.
“The diversity of life is not an accident of history; it is a consequence of thermodynamics.”
The kinetic explanation for the gradient was formalized by Allen, Brown & Gillooly (2002), who showed that species richness \(S\) scales with temperature as:
Boltzmann Diversity Scaling
\[ \ln S \;\propto\; -\frac{E_{\mathrm{a}}}{k_{\mathrm{B}} T} \]
where \(E_{\mathrm{a}} \approx 0.6{-}0.7\) eV is the average activation energy of metabolism, \(k_{\mathrm{B}}\) is Boltzmann’s constant, and \(T\) is absolute temperature (K). The slope of\(\ln S\) vs. \(1/k_{\mathrm{B}}T\) is remarkably consistent across taxa and continents.
This is, at its core, a statement about biochemistry: the rate at which new species arise is governed by mutation rates and generation times, both of which are set by the kinetics of the molecular machinery of life. The equilibrium number of species in a region is determined by the balance between speciation (driven by metabolic kinetics) and extinction (driven by environmental variability and competition). Temperature shifts this equilibrium because it shifts the rates of the underlying biochemical reactions.
Thermodynamics of living matter
Every organism exists as a thermodynamic system far from equilibrium, maintained by a continuous flux of free energy from the environment. The fundamental quantity governing the stability of biological macromolecules is the Gibbs free energy change \(\Delta G\):
Gibbs Free Energy of Folding
\[ \Delta G_{\text{fold}} = \Delta H_{\text{fold}} - T \,\Delta S_{\text{fold}} \]
For a protein to be functional, it must be folded into its native conformation, which requires \(\Delta G_{\text{fold}} < 0\). The stability of the folded state is surprisingly marginal: for most globular proteins, \(\Delta G_{\text{fold}} \approx -20\)to \(-60\) kJ/mol, equivalent to just a few hydrogen bonds. This marginal stability is not a design flaw — it is a feature. Proteins must be stable enough to function but flexible enough to catalyse reactions, bind substrates, and respond to allosteric signals.
The temperature dependence of protein stability creates a characteristic “stability curve” — a plot of \(\Delta G_{\text{fold}}\) versus temperature that is approximately parabolic. The protein is most stable at some intermediate temperature \(T_{\mathrm{s}}\) (the temperature of maximum stability) and becomes unstable (unfolds) at both high and low temperatures. This is because:
Derivation: The Protein Stability Curve
Step 1. The enthalpy and entropy of folding are themselves temperature-dependent through the heat capacity change\(\Delta C_p\):
\[ \Delta H(T) = \Delta H(T_{\mathrm{m}}) + \Delta C_p (T - T_{\mathrm{m}}) \]
\[ \Delta S(T) = \Delta S(T_{\mathrm{m}}) + \Delta C_p \ln\!\left(\frac{T}{T_{\mathrm{m}}}\right) \]
Step 2. At the melting temperature\(T_{\mathrm{m}}\), \(\Delta G = 0\), so\(\Delta S(T_{\mathrm{m}}) = \Delta H(T_{\mathrm{m}})/T_{\mathrm{m}}\).
Step 3. Substituting into\(\Delta G = \Delta H - T\Delta S\):
\[ \Delta G(T) = \Delta H(T_{\mathrm{m}})\!\left(1 - \frac{T}{T_{\mathrm{m}}}\right) + \Delta C_p\!\left[(T - T_{\mathrm{m}}) - T \ln\!\left(\frac{T}{T_{\mathrm{m}}}\right)\right] \]
This is the Becktel-Schellman equation. The\(\Delta C_p\) term, which arises from the hydrophobic effect (exposure of non-polar surface area upon unfolding), creates the characteristic parabolic shape. A large negative \(\Delta C_p\) of unfolding means both hot and cold denaturation are possible.
The ecological implication is profound: every organism has a finite thermal window of viability, set ultimately by the thermodynamics of its constituent proteins. For most mesophilic organisms, this window spans roughly 0–45°C. Thermophiles extend the upper bound to ~80°C through increased hydrophobic core packing, more salt bridges, and a higher proportion of proline residues. Psychrophiles extend the lower bound to ~−15°C through increased glycine content (flexibility), reduced hydrophobic core size, and antifreeze proteins that inhibit ice crystal growth.
Key insight: The marginal stability of proteins (\(\Delta G \approx -20\) to \(-60\) kJ/mol) means that relatively small shifts in temperature — as little as 5–10°C — can push proteins from fully functional to partially or fully denatured. This is why climate warming of even 2–3°C can be catastrophic for organisms living near their thermal limits.
Enzyme kinetics and temperature
The rate of any enzyme-catalysed reaction depends on temperature through two opposing effects: (1) the Arrhenius acceleration of the catalytic rate constant, and (2) the thermal denaturation of the enzyme itself. The Arrhenius equation describes the first effect:
Arrhenius Equation
\[ k(T) = A \exp\!\left(-\frac{E_{\mathrm{a}}}{k_{\mathrm{B}} T}\right) \]
where \(k\) is the rate constant, \(A\) is the pre-exponential factor (related to the frequency of molecular collisions), \(E_{\mathrm{a}}\) is the activation energy, and\(k_{\mathrm{B}} T\) is the thermal energy.
Taking the natural logarithm gives the linearized Arrhenius form:
\[ \ln k = \ln A - \frac{E_{\mathrm{a}}}{k_{\mathrm{B}}} \cdot \frac{1}{T} \]
A plot of \(\ln k\) versus \(1/T\) (an Arrhenius plot) gives a straight line with slope \(-E_{\mathrm{a}}/k_{\mathrm{B}}\).
The \(Q_{10}\) coefficient — the factor by which a rate increases for a 10°C rise — is related to the activation energy:
Derivation: \(Q_{10}\) from Arrhenius Parameters
Step 1. The ratio of rates at temperatures\(T\) and \(T + 10\):
\[ Q_{10} = \frac{k(T+10)}{k(T)} = \exp\!\left[\frac{E_{\mathrm{a}}}{k_{\mathrm{B}}}\left(\frac{1}{T} - \frac{1}{T+10}\right)\right] \]
Step 2. For \(T \approx 298\) K (25°C) and \(E_{\mathrm{a}} \approx 0.65\) eV:
\[ Q_{10} \approx \exp\!\left[\frac{0.65 \times 11600}{298 \times 308} \times 10\right] \approx 2.5 \]
Most biological rates have \(Q_{10}\) values between 2 and 3, corresponding to activation energies of 0.4–0.8 eV (40–80 kJ/mol).
The temperature dependence of enzyme activity also determines the Michaelis-Menten parameters. The maximum velocity \(V_{\max}\)increases with temperature (more enzyme-substrate collisions, faster catalytic turnover) until denaturation sets in. The Michaelis constant\(K_{\mathrm{m}}\) is less temperature-sensitive, though it does increase at high temperatures as the enzyme’s binding affinity decreases with conformational loosening.
Michaelis-Menten with Temperature
\[ v(T) = \frac{V_{\max}(T) \cdot [S]}{K_{\mathrm{m}}(T) + [S]} \]
where \(V_{\max}(T) = k_{\text{cat}}(T) \cdot [E]_{\text{total}} \cdot f_{\text{native}}(T)\), with \(f_{\text{native}}(T)\) being the fraction of enzyme in the native (folded) state at temperature \(T\).
Thermal performance curves
The thermal performance curve (TPC) is perhaps the most fundamental concept in thermal biology. It describes how the performance of an organism — measured as growth rate, locomotion speed, photosynthetic rate, or any other fitness-relevant trait — varies with body temperature. The TPC is universally left-skewed: performance rises gradually with temperature (the Arrhenius acceleration), reaches a maximum at the optimal temperature\(T_{\text{opt}}\), and then drops precipitously as proteins denature.
The mathematical model that best captures this asymmetry is the Sharpe-Schoolfield equation, which incorporates both Arrhenius activation and high-temperature deactivation:
Sharpe-Schoolfield Model
\[ r(T) = \frac{r_{\text{ref}} \cdot \exp\!\left[\dfrac{E_{\mathrm{a}}}{\mathcal{R}}\left(\dfrac{1}{T_{\text{ref}}} - \dfrac{1}{T}\right)\right]}{1 + \exp\!\left[\dfrac{E_{\mathrm{d}}}{\mathcal{R}}\left(\dfrac{1}{T_{1/2}} - \dfrac{1}{T}\right)\right]} \]
where \(r_{\text{ref}}\) is the rate at a reference temperature\(T_{\text{ref}}\), \(E_{\mathrm{a}}\) is the activation energy,\(E_{\mathrm{d}}\) is the deactivation energy, \(T_{1/2}\) is the temperature at which 50% of the enzyme is denatured, and\(\mathcal{R}\) is the universal gas constant.
The optimal temperature can be derived by setting\(dr/dT = 0\):
Derivation: Optimal Temperature
Step 1. Let\(u = E_{\mathrm{a}}/(\mathcal{R} T^2)\) and\(v = E_{\mathrm{d}}/(\mathcal{R} T^2)\). Differentiating the Sharpe-Schoolfield equation and setting the derivative to zero yields:
\[ \frac{E_{\mathrm{a}}}{\mathcal{R} T_{\text{opt}}^2} = \frac{E_{\mathrm{d}}}{\mathcal{R} T_{\text{opt}}^2} \cdot \frac{\exp\!\left[\frac{E_{\mathrm{d}}}{\mathcal{R}}\left(\frac{1}{T_{1/2}} - \frac{1}{T_{\text{opt}}}\right)\right]}{1 + \exp\!\left[\frac{E_{\mathrm{d}}}{\mathcal{R}}\left(\frac{1}{T_{1/2}} - \frac{1}{T_{\text{opt}}}\right)\right]} \]
Step 2. Simplifying, the optimal temperature satisfies:
\[ T_{\text{opt}} = \frac{E_{\mathrm{d}} \cdot T_{1/2}}{E_{\mathrm{d}} + \mathcal{R} T_{1/2} \ln\!\left(\frac{E_{\mathrm{d}}}{E_{\mathrm{a}}} - 1\right)} \]
This shows that \(T_{\text{opt}}\) is always less than\(T_{1/2}\) and depends on the ratio\(E_{\mathrm{d}}/E_{\mathrm{a}}\). When deactivation energy is much larger than activation energy (sharp decline after the peak),\(T_{\text{opt}}\) is close to \(T_{1/2}\).
A critical ecological concept derived from the TPC is the thermal safety margin (TSM): the difference between an organism’s current habitat temperature and its\(T_{\text{opt}}\). Paradoxically, tropical ectotherms have the smallest TSMs despite living far from absolute thermal limits. This is because tropical species have evolved narrow thermal windows closely matched to the stable tropical climate. A warming of 2–3°C in the tropics can push many species past their \(T_{\text{opt}}\), onto the steep declining limb of their TPC, causing disproportionate performance loss.
“Tropical ectotherms are living on a knife-edge: their thermal performance curves are narrow, their safety margins slim, and the consequences of even modest warming are severe.”
Metabolic theory of ecology
The metabolic theory of ecology (MTE), developed by Brown, Gillooly, Allen, Savage & West (2004), provides a unified framework for understanding ecological patterns through the lens of individual metabolism. The central equation combines allometric scaling with Boltzmann kinetics:
MTE Fundamental Equation
\[ B = B_0 \, M^{3/4} \, \exp\!\left(-\frac{E_{\mathrm{a}}}{k_{\mathrm{B}} T}\right) \]
where \(B\) is the whole-organism metabolic rate,\(B_0\) is a normalization constant, \(M\) is body mass,\(E_{\mathrm{a}} \approx 0.65\) eV is the average activation energy of metabolism, \(k_{\mathrm{B}}\) is Boltzmann’s constant, and \(T\) is absolute temperature.
The \(M^{3/4}\) scaling (Kleiber’s law) arises from the fractal geometry of biological distribution networks (West, Brown & Enquist 1997). The Boltzmann term captures the temperature dependence of the underlying biochemistry. Together, they predict how metabolic rate — and therefore all rate-dependent ecological processes — varies across the tree of life.
Derivation: Why 3/4 Scaling?
Step 1. Biological distribution networks (circulatory, respiratory, vascular) are space-filling fractals that minimize transport costs. They branch hierarchically, with each level having \(n\) daughter branches.
Step 2. The network must service a 3D volume (the organism’s body), but the terminal units (capillaries, leaf veins) are invariant in size. The total flow rate through the network scales as:
\[ B \propto N_{\text{capillaries}} \propto M^{3/4} \]
Step 3. The 3/4 exponent emerges because the network is a 3D fractal that scales with an effective dimensionality of 4 (3 spatial + 1 hierarchical): the ratio of volumes between successive branching levels gives\(B \propto M^{d/(d+1)} = M^{3/4}\).
The MTE makes powerful predictions about ecological patterns. For species richness, the argument proceeds: (1) the total energy flux in a region is fixed by solar input; (2) each individual consumes energy at rate \(B \propto M^{3/4} e^{-E_{\mathrm{a}}/k_{\mathrm{B}}T}\); (3) the number of individuals is therefore \(N \propto E_{\text{total}} / B\); (4) species richness scales with the number of individuals (more-individuals hypothesis). The resulting prediction:
MTE Species Richness Prediction
\[ \ln S = \ln S_0 - \frac{E_{\mathrm{a}}}{k_{\mathrm{B}} T} + \frac{3}{4} \ln M + \ln A \]
where \(A\) is area. This predicts that species richness increases exponentially with temperature, decreases with body size (smaller organisms are more species-rich), and increases with area (species-area relationship).
An important consequence of the MTE is the energetic equivalence rule: the total energy flux through a population is approximately independent of body size. Because individual metabolic rate scales as \(M^{3/4}\) and population density scales as \(M^{-3/4}\) (Damuth’s rule), the total population energy use \(N \cdot B \propto M^{-3/4} \cdot M^{3/4} = M^0\) — it cancels out. This means that a hectare of forest allocates roughly the same total metabolic energy to its beetle population as to its bird population, regardless of the enormous difference in body sizes.
Plant biochemistry — C3, C4, CAM
The distribution of photosynthetic pathways across the globe is one of the clearest examples of how biochemistry shapes biodiversity. The three major pathways — C3, C4, and CAM — represent distinct evolutionary solutions to the fundamental biochemical challenge of fixing CO₂ while minimizing water loss and photorespiratory carbon waste.
C3 photosynthesis is the ancestral pathway, used by ~85% of plant species. The key enzyme is RuBisCO (ribulose-1,5-bisphosphate carboxylase/oxygenase), which catalyses the initial fixation of CO₂ into a 3-carbon compound (3-phosphoglycerate). RuBisCO is notoriously inefficient: it also fixes O₂ in a wasteful side reaction called photorespiration, which releases previously fixed carbon as CO₂.
RuBisCO Carboxylation vs. Oxygenation
\[ v_c = \frac{V_{\max} \cdot [\text{CO}_2]}{[\text{CO}_2] + K_{\mathrm{c}}\!\left(1 + \frac{[\text{O}_2]}{K_{\mathrm{o}}}\right)} \]
\[ v_o = \frac{V_{\max} \cdot [\text{O}_2]}{[\text{O}_2] + K_{\mathrm{o}}\!\left(1 + \frac{[\text{CO}_2]}{K_{\mathrm{c}}}\right)} \]
The ratio \(v_c/v_o\) depends on the CO₂/O₂ratio and the specificity factor\(\tau = (V_c \cdot K_{\mathrm{o}})/(V_o \cdot K_{\mathrm{c}})\). At 25°C, \(\tau \approx 80{-}100\), but it decreases with temperature, making photorespiration worse in hot environments.
The temperature dependence of the specificity factor is the key to understanding the geographic distribution of C3 vs. C4 plants. As temperature rises: (1) \(\tau\) decreases (RuBisCO becomes less able to discriminate CO₂ from O₂); (2) CO₂ solubility decreases relative to O₂; (3) the CO₂ compensation point (\(\Gamma^*\)) rises. The crossover temperature at which C4 photosynthesis becomes advantageous is approximately 22–30°C (daytime temperature), corresponding to the biogeographic boundary between C3- and C4-dominated grasslands.
C4 photosynthesis evolved independently at least 66 times across 19 families of angiosperms — a remarkable case of convergent biochemical evolution. The C4 pathway uses PEP carboxylase to initially fix CO₂ into a 4-carbon compound (oxaloacetate) in mesophyll cells, then shuttles this to bundle sheath cells where CO₂ is released and refixed by RuBisCO at high concentration. This CO₂-concentrating mechanism effectively eliminates photorespiration:
C4 CO₂ Concentrating
\[ [\text{CO}_2]_{\text{bundle sheath}} \approx 10 \times [\text{CO}_2]_{\text{ambient}} \]
At these elevated concentrations, RuBisCO operates near\(V_{\max}\) for carboxylation, and the oxygenation rate becomes negligible. The cost is 2 additional ATP per CO₂fixed (total: 5 ATP vs. 3 ATP for C3).
CAM (Crassulacean Acid Metabolism)temporally separates CO₂ fixation from the Calvin cycle. Stomata open at night (cooler, less evaporative demand) to fix CO₂ via PEP carboxylase into malate, which is stored in vacuoles. During the day, stomata close; malate is decarboxylated to release CO₂ for RuBisCO behind closed stomata. This achieves extraordinary water-use efficiency (WUE):
Water-Use Efficiency Comparison
\[ \text{WUE}_{\text{C3}} \approx 1{-}3 \;\text{mmol CO}_2/\text{mol H}_2\text{O} \]
\[ \text{WUE}_{\text{C4}} \approx 3{-}6 \;\text{mmol CO}_2/\text{mol H}_2\text{O} \]
\[ \text{WUE}_{\text{CAM}} \approx 6{-}35 \;\text{mmol CO}_2/\text{mol H}_2\text{O} \]
The implications for biodiversity are direct. Under climate warming, the competitive balance between C3 and C4 plants shifts: C4 species expand poleward, and C3 species retreat to higher elevations or are replaced. Elevated CO₂ partially counteracts this by favouring C3 plants (CO₂ fertilization reduces photorespiration), but the net effect depends on the interaction between temperature and CO₂ concentration — a quintessentially biochemical problem.
Animal metabolism and thermoregulation
Animals face the same thermodynamic constraints as all living organisms, but have evolved two fundamentally different strategies for managing body temperature: ectothermy(body temperature tracks the environment) and endothermy (body temperature is maintained by internal heat production). These strategies have profound consequences for metabolic rates, geographic ranges, and vulnerability to climate change.
The Scholander-Irving model (1950) describes the relationship between metabolic rate and ambient temperature for an endotherm. Below the thermoneutral zone (TNZ), the animal must increase heat production to maintain body temperature; above the TNZ, it must increase evaporative cooling:
Scholander-Irving Heat Balance
\[ B_{\text{met}} = C \cdot (T_{\mathrm{b}} - T_{\mathrm{a}}) \quad \text{for } T_{\mathrm{a}} < T_{\text{LCT}} \]
where \(B_{\text{met}}\) is metabolic rate,\(C\) is thermal conductance (W/°C),\(T_{\mathrm{b}}\) is body temperature,\(T_{\mathrm{a}}\) is ambient temperature, and\(T_{\text{LCT}}\) is the lower critical temperature of the thermoneutral zone.
Thermal conductance scales with body mass as\(C \propto M^{0.5}\), while basal metabolic rate scales as\(B_{\text{BMR}} \propto M^{0.75}\). The ratio\(B_{\text{BMR}}/C \propto M^{0.25}\) gives the width of the thermal gradient an animal can sustain without increasing metabolic output. This means larger endotherms have proportionally wider thermoneutral zones and can tolerate colder environments — Bergmann’s rule emerges naturally from metabolic scaling.
Derivation: Bergmann’s Rule from Metabolic Scaling
Step 1. The minimum survivable temperature is where maximum metabolic rate equals heat loss:
\[ B_{\max} = C \cdot (T_{\mathrm{b}} - T_{\min}) \]
Step 2. Maximum sustainable metabolic rate scales as \(B_{\max} \propto M^{0.75}\) (typically 5–10 \(\times\) BMR), and conductance as\(C \propto M^{0.5}\). Therefore:
\[ T_{\mathrm{b}} - T_{\min} = \frac{B_{\max}}{C} \propto \frac{M^{0.75}}{M^{0.5}} = M^{0.25} \]
Step 3. Larger animals can sustain a larger thermal gradient, meaning they can survive at lower ambient temperatures. This predicts that, within a lineage, populations at higher latitudes (colder climates) should be larger-bodied — Bergmann’s rule.
The ecological consequences of thermoregulatory strategy are profound. Ectotherms are “cheap to run” (low metabolic overhead) but are slaves to their thermal environment; their activity windows shrink as climate warms beyond their TPC optima. Endotherms are “expensive to run” (high basal metabolism) but can remain active across a wide range of ambient temperatures. However, endotherms face an upper thermal limit set by their capacity for evaporative cooling — above the upper critical temperature, metabolic rate rises sharply as the animal activates costly heat-dissipation mechanisms (panting, sweating, saliva-spreading), and dehydration becomes lethal.
Key insight: The metabolic cost of thermoregulation means that climate warming affects endotherms and ectotherms through different mechanisms. Ectotherms lose performance (thermal performance curve effects). Endotherms lose energy budget (increased cooling costs reduce energy available for reproduction and growth). Both pathways lead to population decline, but through distinct biochemical mechanisms.
Population dynamics with climate
The interaction between climate and population dynamics can be modelled by incorporating temperature-dependent parameters into classical ecological equations. The Lotka-Volterra competition model, extended with metabolic temperature dependence, reveals how warming can shift competitive outcomes and restructure communities.
Consider two competing species with temperature-dependent intrinsic growth rates \(r_i(T)\) derived from their thermal performance curves:
Temperature-Dependent Lotka-Volterra Competition
\[ \frac{dN_1}{dt} = r_1(T) \, N_1 \left(\frac{K_1(T) - N_1 - \alpha_{12} N_2}{K_1(T)}\right) \]
\[ \frac{dN_2}{dt} = r_2(T) \, N_2 \left(\frac{K_2(T) - N_2 - \alpha_{21} N_1}{K_2(T)}\right) \]
where \(r_i(T)\) and \(K_i(T)\) follow Sharpe-Schoolfield kinetics, and \(\alpha_{ij}\) is the competition coefficient (effect of species \(j\)on species \(i\)).
The coexistence condition from classical theory requires\(\alpha_{12} < K_1/K_2\) and\(\alpha_{21} < K_2/K_1\). When carrying capacities are temperature-dependent, a temperature shift can violate these inequalities, causing competitive exclusion of one species. Specifically, if species 1 has a lower\(T_{\text{opt}}\) than species 2, warming decreases\(K_1(T)/K_2(T)\), tightening the coexistence condition until species 1 is excluded.
Derivation: Climate-Driven Competitive Exclusion
Step 1. At equilibrium,\(dN_i/dt = 0\). The isoclines are:
\[ N_1^* = K_1(T) - \alpha_{12} N_2^* \]
\[ N_2^* = K_2(T) - \alpha_{21} N_1^* \]
Step 2. Coexistence requires the isoclines to intersect in the positive quadrant. The intersection is stable when:
\[ \alpha_{12} \alpha_{21} < 1 \quad \text{(mutual limitation)} \]
Step 3. With temperature-dependent carrying capacities, the equilibrium densities become:
\[ N_1^* = \frac{K_1(T) - \alpha_{12} K_2(T)}{1 - \alpha_{12}\alpha_{21}}, \quad N_2^* = \frac{K_2(T) - \alpha_{21} K_1(T)}{1 - \alpha_{12}\alpha_{21}} \]
Species 1 is excluded when \(N_1^* \leq 0\), i.e., when \(K_1(T) \leq \alpha_{12} K_2(T)\). This defines a critical temperature above which species 1 cannot persist.
The predator-prey dynamics are similarly affected. The Lotka-Volterra predator-prey model with temperature-dependent attack rates becomes:
Temperature-Dependent Predator-Prey
\[ \frac{dN}{dt} = r(T) N - a(T) N P \]
\[ \frac{dP}{dt} = e \, a(T) N P - d(T) P \]
where \(a(T) \propto e^{-E_{\mathrm{a}}^{(a)}/k_{\mathrm{B}}T}\) is the temperature-dependent attack rate and\(d(T) \propto e^{-E_{\mathrm{a}}^{(d)}/k_{\mathrm{B}}T}\) is the predator death rate. If the activation energies differ (\(E_{\mathrm{a}}^{(a)} \neq E_{\mathrm{a}}^{(d)}\)), warming can decouple predator-prey oscillations and destabilize the interaction.
Species coexistence and niche overlap
The thermal niche of a species is defined by its thermal performance curve. When two species have overlapping thermal niches, they compete for resources at temperatures where both can be active. The degree of niche overlap determines whether coexistence is possible or whether competitive exclusion will eliminate one species.
MacArthur & Levins (1967) quantified niche overlap using the overlap integral. For two species with resource utilization functions\(p_1(x)\) and \(p_2(x)\) along a resource axis\(x\) (here, temperature):
Niche Overlap (Pianka’s Index)
\[ O_{12} = \frac{\int p_1(x) \, p_2(x) \, dx}{\sqrt{\int p_1^2(x) \, dx \cdot \int p_2^2(x) \, dx}} \]
\(O_{12} = 0\) indicates no overlap (complete niche separation);\(O_{12} = 1\) indicates identical niches (maximum competition).
The limiting similarity principle states that stable coexistence requires a minimum separation between niche optima. If thermal performance curves are approximately Gaussian with width \(\sigma\), the critical separation is:
Limiting Similarity Criterion
\[ |T_{\text{opt},1} - T_{\text{opt},2}| > d_{\min} \approx \sigma \]
The exact threshold depends on the competitive environment, resource productivity, and environmental variability, but the niche width\(\sigma\) sets the fundamental scale.
Climate change threatens coexistence by compressing thermal niches. If two species coexist because their thermal optima are separated by \(d > d_{\min}\), warming shifts both species’ experienced temperatures. But if the warm-adapted species is already near its upper thermal limit, it cannot shift further, while the cold-adapted species is pushed toward (and eventually past) the warm-adapted species’ niche. The result is increased niche overlap and potential competitive exclusion — a mechanism for climate-driven biodiversity loss that operates entirely through biochemical thermal constraints.
“Climate change does not merely shift ranges; it reshuffles the biochemical deck, altering which species can coexist and which must yield.”
Climate change and the thermal window
We now bring together all the threads of this monograph to address the central question: how will climate change reshape the biochemical equilibrium that sustains biodiversity? The answer emerges from the intersection of protein thermodynamics (§ 02), enzyme kinetics (§ 03), thermal performance curves (§ 04), metabolic theory (§ 05), and population dynamics (§ 08–09).
The thermal window of a species is the range of temperatures over which its thermal performance exceeds some critical threshold (e.g., \(r(T) > 0\), positive population growth). This window is bounded by the critical thermal minimum (\(\text{CT}_{\min}\)) and maximum (\(\text{CT}_{\max}\)). The thermal safety margin (TSM) is the distance from the current habitat temperature to \(\text{CT}_{\max}\):
Thermal Safety Margin
\[ \text{TSM} = \text{CT}_{\max} - T_{\text{habitat}} \]
Global analyses reveal that tropical ectotherms have TSMs of only 1–3°C, while temperate and polar species have TSMs of 10–20°C. This means tropical species are disproportionately vulnerable to even modest warming — a pattern sometimes called the tropical vulnerability paradox.
The fraction of species in a community that will experience thermal stress under a given warming scenario can be estimated by integrating the distribution of TSMs across species:
Derivation: Fraction of Species at Risk
Step 1. Let \(f(\text{TSM})\)be the probability density function of thermal safety margins across species in a community. A species is at risk when warming\(\Delta T\) exceeds its TSM.
Step 2. The fraction of species at risk is:
\[ F_{\text{risk}}(\Delta T) = \int_0^{\Delta T} f(\text{TSM}) \, d(\text{TSM}) = \text{CDF}(\Delta T) \]
Step 3. If TSMs are approximately normally distributed with mean \(\mu_{\text{TSM}}\) and standard deviation \(\sigma_{\text{TSM}}\):
\[ F_{\text{risk}}(\Delta T) = \Phi\!\left(\frac{\Delta T - \mu_{\text{TSM}}}{\sigma_{\text{TSM}}}\right) \]
where \(\Phi\) is the standard normal CDF. For a tropical community with \(\mu_{\text{TSM}} = 2\)°C and\(\sigma_{\text{TSM}} = 1\)°C, a warming of +3°C puts ~84% of species at risk. For a temperate community with \(\mu_{\text{TSM}} = 12\)°C and\(\sigma_{\text{TSM}} = 5\)°C, the same warming puts only ~4% at risk.
The total biodiversity change under warming integrates across all the mechanisms discussed in this monograph. We can write a master equation for the rate of change of species richness:
Integrated Biodiversity Change
\[ \frac{dS}{dt} = \underbrace{\lambda(T) \cdot S}_{\text{speciation}} - \underbrace{\mu(T) \cdot S}_{\text{extinction}} - \underbrace{\gamma(T) \cdot S}_{\text{climate-driven loss}} \]
where \(\lambda(T) \propto e^{-E_{\mathrm{a}}/k_{\mathrm{B}}T}\) (speciation rate increases with temperature), \(\mu(T)\) is the background extinction rate, and \(\gamma(T)\) is the additional extinction rate due to climate change (thermal stress, niche compression, phenological mismatch).
The tragedy is that while warming increases the speciation rate (Boltzmann kinetics), this effect operates on evolutionary timescales (10\(^5\)–10\(^7\) years), while climate-driven extinction operates on ecological timescales (10\(^1\)–10\(^3\) years). The rate of warming far exceeds the rate at which biodiversity can regenerate through speciation. We are drawing down a thermodynamic capital that took millions of years to accumulate.
The biochemical perspective clarifies why certain thresholds matter. The Paris Agreement target of 1.5°C is not arbitrary — it corresponds approximately to the mean thermal safety margin of tropical ectotherms. Below 1.5°C warming, most tropical species remain within their thermal performance envelope. Above it, an increasing fraction of tropical species are pushed past their biochemical limits: enzymes denature, membrane fluidity becomes dysfunctional, and the molecular machinery of life begins to fail.
Similarly, the often-cited “tipping point” at 2°C of warming corresponds to the threshold where niche compression begins to cause widespread competitive exclusion (Section 09). And the catastrophic scenarios above 4°C correspond to temperatures at which the Boltzmann diversity scaling predicts a fundamental restructuring of the global biodiversity gradient — not merely a loss of species, but a reorganization of life’s thermodynamic equilibrium on Earth.
Final synthesis: Biodiversity is a thermodynamic phenomenon. It emerges from the Boltzmann distribution of molecular energies, is maintained by the delicate balance of protein folding free energies, is organized by the metabolic scaling laws that govern all living systems, and is threatened by any perturbation that disrupts the biochemical equilibrium on which life depends. The equations of thermodynamics and enzyme kinetics are not abstract theoretical tools — they are the governing equations of biodiversity itself.
“In the end, the fate of biodiversity is written in the language of thermodynamics. The question is whether we will read it in time.”