Energy balance models are a fundamental tool used to understand the Earth's climate system and its energy dynamics. They represent the energy budget within the Earth's atmosphere, land, oceans and ice by quantifying the balance between incoming solar radiation and outgoing solar radiation. Although highly simplified compared to general circulation models, energy balance models (EBMs) are appreciated for their interpretability, mathematical tractability and ability to capture the essential dynamics of the Earth system . Two important feedback mechanisms are typically present in such models: the ice-albedo feedback and the Stefan–Boltzmann law. The positive ice-albedo feedback occurs when the melting of ice and snow reduces the surface reflectivity (albedo), causing the planet to absorb more solar radiation. According to the Stefan–Boltzmann law, a warmer body emits more radiation, thereby providing a negative feedback which stabilizes the planet's temperature. Depending on the precise configuration, these mechanisms may endow EBMs with bistability, suggesting the existence of two stable climates commonly referred to as the snowball climate and the warm climate. The snowball climate, supported by palaeoclimatic evidence from the Cryogenian period around 650 million years ago, is characterized by the absence of vegetation and the presence of ice caps extending over the entire planet's surface. In contrast, the warm climate exhibits relatively low albedo, ice caps limited to the polar regions, and the presence of oceans and vegetation. Additionally, EBMs typically allow for a third possible climate, albeit unstable. Transitions between stable climates in an EBM, as well as in general multi-stable models, can occur in various ways. But two important mechanisms are the following. The first consists of changes in factors influencing the climate system, such as variations in greenhouse gas concentrations like carbon dioxide (CO2), altering the balance of incoming and outgoing radiation and amplifying the greenhouse effect. Mathematically, this mechanism can be described by assuming that the model depends on one additional parameter, and changes in the parameter lead the model to undergo a bifurcation ; the second consists of noise-induced transitions resulting from unresolved processes in climate models or the representation of short-timescale weather as stochastic forcing acting on slow variables, as observed in stochastic reduced models . These two types of transitions correspond to mechanisms recognized to induce climate tipping, that is, rapid non-linear changes in the climate system with potentially irreversible and catastrophic consequences .

A zero-dimensional (0D) EBM is the simplest version of an EBM describing the evolution in time for the annual averaged global mean temperature T, without any space dependence . This model is given by an ordinary differential equation (ODE) of the form 1CTdTdt=Q‾0β(T)+q-σ0ε0T4,t>0,T∣t=0=T0. In this equation, CT>0 represents the heat capacity, Q‾0>0 is the globally averaged solar radiation and the co-albedo β is modelled by a continuous function (overbars typically denote globally averaged quantities). Further, q>0 is a positive parameter modelling the effect of the CO2 on the energy budget . The term R‾e(T)=σ0ε0T4 on the right-hand side of Eq. () accounts for the outgoing solar radiation, following the Stefan–Boltzmann law (where σ0 denotes the Stefan–Boltzmann constant, and ε0 is the globally averaged emissivity). The fixed points of the model are the solutions of the equation: dTdt=0, corresponding to points in Fig.  where the absorbed radiation R‾a(T)=Q‾0β(T)+q and the emitted radiation R‾e(T) intersect. Figure a furthermore illustrates that this model is generally characterized by bistability, with two stable fixed points TS and TW. These points correspond to the snowball and warm climate states mentioned earlier and are separated by an unstable intermediate fixed point TM. Furthermore, as highlighted by Fig. b, the stable points correspond to minimum points of a primitive function F‾ for the negative radiation budget R‾. In other words, F‾ is any regular function such that F‾′(T)=Re‾(T)-Ra‾(T)=-R‾(T). To better capture the variability of global mean surface temperature, it has been proposed to add a stochastic forcing, such as white noise, to the radiation balance. This is interpreted as the effect of the fast components of the climate system, i.e. the weather, over slow components . For this reason, we are interested in considering the stochastic differential equation (SDE) given by 2dT=R‾(T)dt+εdWt, where ε>0 is the noise intensity and (Wt)t≥0 is a Brownian motion . This SDE is of gradient type and possesses a unique Gibbs invariant measure ν‾ . An invariant measure is a probability distribution ν‾ in the state space of Eq. () (i.e. the real numbers in this case) with the property that if a solution T is distributed according to ν‾ at some time t, then it remains so for all later times. It is given by 3ν‾(dT)=1Zexp⁡-2ε2F‾(T)dT, where Z is a normalization constant, and dT denotes the standard volume element on R (we note the technical detail that to give meaning to Eqs. () and (), the radiation budget R‾ should be extended to negative values for the Kelvin temperature T in a way such that F‾→+∞ as T→-∞). The key observation from the explicit formula (Eq. ) is that ν‾ is concentrated around the minimum points of the function F‾. Indeed, if T0 is a strict minimum point and T1≠T0 is a point close to T0 such that F‾(T1)>F‾(T0), then the mass given by the measure ν in a small neighbourhood of T1 is exponentially lower than the mass around T0; more specifically, the ratio between the two masses is given by exp⁡-2ε2F‾(T1)-F‾(T0).

A one-dimensional (1D) EBM is given by a parabolic partial differential equation where the space variable is one-dimensional . Denoting the temperature averaged in the zonal direction by u=u(t,x), it extends the 0D EBM by introducing the sine of the latitude x=sin⁡(ϕ), where ϕ∈[-π2,π2] denotes the latitude and t≥0 represents time. We assume that the non-linear radiation balance of the planet, denoted by R(x,u;q), depends on the sine of the latitude and on an additive parameter q. This parameter models the effect of carbon dioxide concentration on the radiation budget . Atmospheric and ocean transport of heat between latitudes is modelled in a very simplified way by a diffusion term. Assuming spatially homogeneous diffusion in this introductory section and thus ignoring the dependence of κ on latitude and temperature, we obtain a non-degenerate reaction–diffusion equation: 4∂tu=κΔu+R(x,u;q),t>0,x∈(-1,1),ux(t,-1)=ux(t,1)=0,t≥0u(0,x)=ũ(x),x∈[-1,1], where Δ=∂xx denotes the Laplace operator in dimension one, the Neumann boundary conditions impose no-heat flux at the poles and ũ is an initial condition. The steady-state solutions of this model, representing the asymptotic solutions for the time-evolving dynamics, correspond to the non-negative solutions u=u(x) of the following elliptic problem: 50=κu′′+R(x,u;q),x∈(-1,1),u′(-1)=u′(1)=0, where u=u(x) depends only on the space variable. This elliptic problem forms a necessary condition for u=u(x) to be our extremal (in particular a local minimizer) for the potential functional 6Fq(u)=∫-11R(x,u;q)dx+κ2||u′||22, where ∂uR(x,u;q)=-R(x,u;q), and ||u′||22=∫-11u′2dx is the square of the norm of u′ in L2(-1,1). Calculus of variations is a widely employed technique for studying the existence of a solution to the previous problem . However, proving the existence of a local (but not global) minimum point is generally challenging, and this technique focuses on studying the existence of the global minimum point. The functional Fq in Eq. () has another interpretation though which renders it more important than being merely a characterization of solutions to the elliptic problem. Indeed, consider the stochastic partial differential equation (SPDE) on the Hilbert space H=L2(-1,1), given by 7du=κΔu+R(x,u;q)dt+εdWt, obtained by adding a space–time white noise (Wt)t≥0 modelled by a cylindrical Brownian motion on H=L2(-1,1) to Eq. (). R has a cutoff at negative temperature as in Sect. , and ε>0 is the noise intensity. We refer to for more details about SPDEs. It can be shown that this SPDE has a unique invariant Gibbs measure ν , given (broadly speaking) by an expression as in Eq. (), with Fq replacing F‾ (see Sects.  and ). Therefore, as in the zero-dimensional case, ν concentrates on minimum points of the functional Fq. These minimizers satisfy the elliptic problem (Eq. ), which therefore describes temperature profiles around which the solutions of the stochastic problem (Eq. ) tend to cluster.

This paper focuses on the study of the properties and the interpretation of the steady-state solutions of a 1D EBM depending on a bifurcation parameter. Motivated by 0D EBMs, there is a wide consensus in the literature, supported mainly by numerical simulations, regarding the existence of either one or three “interesting” steady-state solutions for 1D EBMs. Firstly, in Theorem 1 we prove the existence of a steady-state solution for the 1D EBM by solving the associated variational problem inf⁡Fq(u)∣u∈X, i.e. showing the existence of a global minimum point for the functional Fq over a suitable function space X. Secondly, in Theorem 2 we provide sufficient conditions to have at least three steady-state solutions. These consist of two local minima and one saddle point. The conditions can be summarized as follows: i.

the viscosity κ should be sufficiently large,

The fundamental mechanism of 1D EBMs is that the temperature u=u(t,x), averaged in the zonal direction, evolves in time due to (i) the diffusion of energy between adjacent regions, (ii) the energy absorbed by the planet and (iii) the energy emitted by the planet. The 1D EBM we consider in this paper is a Seller-type EBM where the absorbed radiation depends on an additive parameter . We only add a change in the diffusion term in order to get a non-degenerate parabolic PDE. Given an initial condition ũ, the non-linear, parabolic, reaction–diffusion PDE governing the model is given by 8CT∂u∂t=∂xκ(x)∂xu+Ra(x,u)-Re(u;q),t>0,x∈(-1,1),κ(-1)ux(t,-1)=κ(1)ux(t,1)=0,t≥0u(0,x)=ũ(x),x∈[-1,1], where Ra and Re represent the radiation absorbed and emitted by the planet per unit area, respectively. CT is the heat capacity, and the differential term parameterizes the meridional heat transport. The boundary conditions impose no flux at the poles. We now provide further details regarding the parameterization of these terms. The values of the constants of the model can be found in Table .

Firstly, the absorbed radiation is assumed to have the following form: Ra(x,u)=Q0(x)(1-α(u)), where Q0 is the solar radiation per unit area, and α is the albedo. The solar radiation is assumed to be Q0(x)=Q^0c1-c2x2,ci>0 where Q^0 is the mean solar radiation, and ci represents constants. The albedo, which is the proportion of the incident light or radiation that is reflected by a surface, is parameterized by a smooth monotonically decreasing function with a peak derivative in a reference temperature uref close to the melting point of ice. Specifically, α(u)=α1+(α2-α1)1+tanh⁡(K(u-uref))2, where K>0 is a rate parameter and α1>α2 are respectively the ice albedo and the water albedo.

Second, the emitted radiation is modelled using the Stefan–Boltzmann law, in other words assuming that the Earth radiates as a black body. Under this assumption, the energy radiated is proportional to the fourth power of its temperature and it is given by Re(u;q)=ε0σ0u4-q, where ε0 and σ0 are respectively the emissivity and Boltzmann's constant. The additive parameter q describes, in a simplified way, the radiative forcing by CO2, i.e. the effect of atmospheric CO2 on the energy budget . It is worth explaining (i) the additive structure of q and (ii) its independence on the spatial variable x. About the first point, denote by C the global CO2 concentration in part per million (ppm) and assume, just for explanation purposes, a dependence of the outgoing radiation both on u and C. To avoid confusion, we denote this by R^e=R^e(u,C), the outgoing radiation depending on temperature and CO2 concentration. If we linearize R^e with respect to temperature, we get R^e(u,C)≈A(C)+B(u-u^), where u^ is a reference temperature. In , three radiative transfer models are used in order to get that the dependence of the outgoing radiation with respect to changes in CO2 is given by A(C)=A1-A2⋅ln⁡CC0, where C0 is a reference CO2 concentration, and A1,A2>0 are explicit constants. In conclusion, R^e≈A1+B(u-u^)-q,q=A2⋅ln⁡CC0, and thus the radiative forcing of CO2 has an additive structure. About point (ii), we adopt the well-mixed hypothesis for CO2. In other words, we assume that atmospheric CO2 is globally homogeneous, thereby inducing a radiative forcing q independent of latitude . This assumption overlooks the spatial pattern of CO2 concentration, which affects many aspects of the climate system, such as the poleward heat transport . It was the state-of-the-art assumption 2 decades ago, although today it is common to keep in consideration the spatial distribution of radiative forcing .

The third component of the model is the term ∂xκ(x)ux. It parameterizes the meridional heat transport, that is, the phenomenon resulting from the poleward transportation of heat by the Earth–atmosphere system due to the surplus of net radiation heating in the tropics and the deficit in the poleward regions. Usually, the diffusion function κ(x) is assumed null at the poles, i.e. with a form such as κ(x)=D(1-x2), where D is a diffusion constant. This choice is based on the paradigm of mimicking the conduction of heat on a sphere; see for a derivation. On the other hand, it leads to mathematical difficulties in the treatment of the singular PDE arising, in particular in the study of the corresponding variational problem, from which all our results follow. To avoid these difficulties, which at the moment remain an open problem to solve, we add as a simplifying assumption that κ is non-degenerate and given by κ(x)=D(1-x2)+δ,D,δ>0. We choose δ=0.003, but its value is not important for the results of this work, and different choices can be made.

For the parabolic problem (Eq. ), the global existence and uniqueness of the solution can be demonstrated, given a regular initial condition . Furthermore, if the initial condition is non-negative, the solution remains non-negative for any time t>0. This can be shown proving that [0,+∞) is an invariant region for Eq. (), exploiting the fact that there exist C1,C2>0 such that R(x,u;q)>C1>0 for all x∈[-1,1] and u∈[0,C2] .

We recall the formulation of stochastic EBMs using the theory of SPDEs . Denote by Δ the Laplace operator with Neumann boundary conditions. Given an initial condition ũ∈H, we consider the following SPDE: 9∂tu=κΔu+Q0(x)β(u)-Re(u;q)+εdWtu∣t=0=ũ, where ε>0, and (Wt)t≥0 is a cylindrical Brownian motion on H. Under the minor cutoff modifications introduced in Sect. , it can be proved that the H-valued stochastic process (ut)t which solves in a mild sense () is unique and has continuous trajectories . In addition to this there exists a unique Gibbs invariant measure 10ν(du)=1Zexp⁡-2ε2∫-11R(x,u;q)dxμ(du), where R is as in Eq. (), and μ∼N(0,-ε22κΔ-1) is a symmetric Gaussian measure on H with covariance Q=-ε22κΔ-1 . The covariance operator Q:H→H is the unique linear operator such that ∫H〈h1,ϕ〉〈h2,ϕ〉μ(dϕ) for each h1,h2∈H, where 〈⋅,⋅〉 denotes the scalar product in H. Further, it can be shown that Q is symmetric and positive-definite, and its eigenvalues (λk)k∈Z satisfy ∑k∈Zλk<∞. In the following lines, given a symmetric, positive-definite operator Q such that ∑k∈Zλk<+∞, we are going to explain how to construct an H-valued random variable X with law N(0,Q). Indeed, consider a sequence (Rk)k∈Z of independent and identically distributed R-N(0,1) random variables defined on a probability space (Ω,F,P). We can assume without loss of generality that the eigenvectors (ek)k∈Z associated with the eigenvalues (λk)k∈Z form an orthonormal basis of H. Then, the H-valued random variable X=∑k∈ZλkRkek is well defined, i.e. the series defining X converges in L2(Ω,F,P;H) and has law N(0,Q) Proposition 2.18. Further, the convergence also holds P almost surely in H Proposition 2.13.

As mentioned in the introduction, this measure is concentrated on minimum points of the functional Fq. A heuristic explanation of this fact can be found in Sect. .

The stationary problem associated with the 1D EBM is given by the elliptic equation for u=u(x): 11κ(x)u′′+Q0(x)β(u)+q-ε0σ0u4=0,x∈(-1,1),u′(-1)=u′(1)=0,u(x)≥0. These solutions can be either stable or unstable, depending on the long-term behaviour of their infinitesimal perturbations. As pointed out in , if the reaction–diffusion equation for u=u(t,x) was space-homogeneous, i.e. of the form 12∂tu=κΔu+R(u), then the stable steady-state solutions would correspond to functions that are constant in space and time, with values given as the roots of R(y)=0. A rigorous result in this direction has been shown in . Indeed, for a fixed double-well symmetric potential, it has been proved that (i) if κ is large enough, the only steady-state solutions of Eq. () are the constants where the potential is critical, and (ii) the number of unstable steady-state solutions to Eq. () can be made arbitrarily large as κ→0. Introducing a spatial dependence in R=R(x,u) leads to a space-heterogeneous model. Depending on the space heterogeneity, it can exhibit any number of both stable and unstable steady-state solutions . The variational approach to the study of steady-state solutions provides a tool for characterizing the stable ones, which are the local minimum points of a functional.

In the following paragraph, we describe the properties of the solutions of Eq. (). As the parameter q changes, numerical simulations for Eq. () suggest the existence of either one or three steady-state solutions. That is, there exists q1<q2 s.t. Eq. () has one steady-state solution if q<q1 or q>q2, and the steady-state solutions are three if q1≤q≤q2. In the latter case, we denote the solutions by uS≤uM≤uW, corresponding respectively to the snowball climate, a middle (or intermediate) climate and the warm climate. As an analogy, we denote by uS the unique steady-state solution for q<q1 and by uW the unique one for q>q2. Figure a shows the bifurcation diagram of the model in the (q,u‾) plane, where u‾=∫-11u(x)dx denotes the average temperature. Figure b depicts the three steady-state solutions for q=25∈(q1,q2).

A stability analysis can be conducted to determine the stability of the steady-state solutions. The results show that uS and uW are stable, while the middle climate, uM, is unstable. Furthermore, it is worth noting that special values q=q1,q2 correspond to bifurcation points of saddle-node type, where the unstable solution uM collides with either uW (for q=q1) or uS (for q=q2) and then disappears. These numerical findings regarding the number and stability of the steady-state solutions will be supported and validated using rigorous arguments, as in the next section.

In this section, we (i) provide an intuitive motivation for why the invariant measure for the stochastic EBM concentrates on minimum points of the functional Fq, (ii) prove the existence of global minimum points for Fq using the direct method, and (iii) present sufficient conditions on the viscosity κ and the space-averaged potential R‾(u)=12∫-11R(x,u)dx, with ∂uR=-R=Re-Ra, to ensure that the 1D EBM has at least three steady-state solutions.

Firstly, consider the stochastic EBM (). Assume that for a negative value of u, where the model has no physical meaning, the Stefan–Boltzmann law is extended as Re(u)=ε0σ0u4 if u≥0,0 if u<0. And β is smoothly extended to β̃ by setting it to zero outside the physically relevant range, as described in the Supplement. Then, Eq. () possesses a unique Gibbs invariant probability measure given by 13ν(du)∝exp⁡-2ε2∫-11ε0σ0u5+5-Q0(x)B(u)-qudxμ(du),μ∼N0,-ε22κΔ-1, where (u)+=max⁡{u,0} is the positive part, N(0,-ε22Δ-1) denotes a symmetric Gaussian measure with covariance operator Q=-ε22κΔ-1 over the Hilbert space H=L2(-1,1) and Z is the normalization constant. See for a rigorous derivation of the invariant measure for a reaction–diffusion model with a polynomial homogeneous reaction term. We move to explain in what sense ν is concentrated around minimum points of Fq. In fact, for u∈H the Gaussian measure μ is formally given by μ(du)=1Z1exp⁡-12〈Q-1u,u〉du, where Q-1=-2κε2Δ. Here, Z1 is a normalization constant, 〈⋅,⋅〉 denotes the scalar product in H and du is a formal notation for the Lebesgue measure on H. If we perform an integration by parts, we get μ(du)=1Z1exp⁡κε2〈u′′,u〉du=1Z1exp⁡-κε2||u′||22du. Plugging the previous identity into Eq. (), we obtain ν(du)∝exp⁡-2ε2∫-11ε0σ0u5+5-Q0(x)B(u)-qudx+κ2||u′||22du∝exp⁡-2ε2Fq(u)du. From this heuristic formula, we see that points u such that Fq(u) is not a global minimum have exponentially smaller density than the minimum points. Indeed, if u1 is a global minimum point and u≠u1, then the mass given by ν in a small neighbourhood around u is exponentially smaller than the mass given to a neighbourhood of the same size around u1; in particular, the ratio between the two masses is given by exp⁡-2ε2Fq(u)-Fq(u1). The previous derivation is formal, because the Lebesgue measure cannot be defined on an infinite-dimensional Hilbert space. For a more rigorous explanation, see Sect. S2 in the Supplement.

Next, we discuss the properties of the functional Fq:H1,2(-1,1)∩u≥0→R given by Fq(u)=∫-11u55ε0σ0-Q0(x)B(u)-qudx+12∫-11κ(x)(u′(x))2dx, where B is a primitive of the co-albedo β(u)=1-α(u), and H1=H1,2(-1,1) denotes the Sobolev space of order 1 and exponent 2, i.e. the function space where a function u and its derivative u′ (in a weak sense) are both square integrable over [-1,1]. See for more details about Sobolev spaces. The functional Fq, depending on the parameter q, is known in the literature as a potential functional or Lyapunov function . The study of the functional Fq gives useful information thanks to its links with the invariant measure for the stochastic 1D EBM, as we have seen, and the stable steady-state solutions for the deterministic 1D EBM which emerge as necessary conditions for the stationarity of Fq. Going deeper with the former point, the first variation of Fq in the point u in direction h is given by δFq(u,h)=ddsFq(u+sh)∣s=0=∫-11u4ε0σ0-Q0(x)β(u)-qhdx+∫-11κ(x)u′(x)h′(x)dx=∫-11u4ε0σ0-Q0(x)β(u)-q-(κ(x)u′(x))′hdx where in the last identity we have used integration by parts. Since h is arbitrary, u is a stationary point for the functional Fq if and only if it is a steady-state solution for the EBM. In particular, local extremum points for Fq correspond to steady-state solutions of the EBM. Any local minimizer of Fq represents a locally attractive solution of the deterministic 1D EBM. In view of our interpretation of Fq in terms of the invariant measure, however, global minimizers play a special role since if present and unique they are exponentially more likely than any other state (including minimizers that are just local). The following result establishes the existence of a global minimum point for Fq.

A rigorous proof of the previous result can be found in Sect. S3. The proof relies on standard arguments from the direct method of calculus of variation, exploiting the fact that the outgoing radiation in the EBM model prevents the temperature from being too high.

Concerning the existence of two local minimum points, let us describe a sufficient condition. Consider the potential function R‾:R→R coming from the space-averaged model R‾(u)=12∫-11R(x,u)dx. If the viscosity κ>0 is sufficiently large and the function R‾ has a double-well shape with sufficiently deep minimum values attained at the minimum points, then we are able to prove the existence of two minimum points for Fq. Further, it is possible to prove that the functional Fq satisfies a compactness condition known as the Palais–Smale condition. This property and the mountain pass theorem give the possibility to deduce the existence of a third steady-state solution. Next, we characterize a situation in which there are three steady-state solutions, two of which are local minimizers . This is summarized in the following result.

i.

R‾′′(ui)>f(ε‾), for i=1,2;

Note how the previous result can be also interpreted as giving sufficient conditions for the convergence of the stable solutions of a space-inhomogeneous EBM to the stable solution of the corresponding space-averaged model, as the diffusion becomes large.

The key element of this section is the value function, which is given by V(q)=inf⁡Fq(u)∣u∈H1,u≥0. From Sect. , we know that the previous infimum is indeed a minimum, and so V(q) can be interpreted as the minimum possible value attained by the potential functional over the possible temperature profiles u. Since a minimum point for Fq is also a stationary point for the functional, the value function can be evaluated numerically by computing the minimum of the three steady-state solutions uS,uM and uW. Following this strategy, Fig.  shows q↦Fq(u*), with u*∈uS,uM,uW. Particularly, there exists a point q3 such that uS is the global minimum point of Fq for q<q3, while uW is the global minimum point for q>q3. Further, for q=q3 the function Fq has two different global minimum points uS and uW, and q=q3 corresponds a non-differentiability point for V. In addition to this, the value function appears to be concave, thus with a decreasing derivative, where it exists.

Summarizing, the numerical evaluations of V(q) suggest the following result, a rigorous proof of which is included in the Supplement.

i.

V is Lipschitz continuous.

We also see numerically that uM is actually never a global minimizer for the specific functional Fq considered here, but we do no have rigorous proof of this fact. Let us briefly discuss the proofs of the previous points. The proof of (i) follows from the facts that the sup norm of the minimizer u0 can be bounded uniformly in q and that, given a family {gi}i∈I of Li Lipschitz functions gi, the inf⁡i∈Igi is Lipschitz continuous if the constants Li can be bounded uniformly. In our case, given u∈H1 is non-negative, we have |Fμ1(u)-Fμ2(u)|≤|μ1-μ2|∫-11|u(x)|dx≤2M|μ1-μ2| where M>0 is the constant appearing in Eq. (). On the other hand, the proof of point (ii) is less straightforward, although being very similar to the one for the existence of a solution for the variational problem. The proof of point (iii) makes use of the concept of semiconcavity, a generalization of that of concavity, which is fundamental in optimal control . The main reason for the concavity of V though is that V is an infimum over functions that are affine in q. Hence, the fact that q is additive is essential for this result. More details can be found in Sects. S4 and S5 in the Supplement.

An additional property of the value function can be observed when comparing the bifurcation diagram (Fig. a) and the graph of the value function (Fig. ). Corollary 4. If V is differentiable, then V′(q)=-∫-11u0(x)dx, where u0 is the only minimizer for Fq. In other words, the part of the bifurcation diagram that corresponds to the global minimizer, represented by the subgraph (q,∫-11u0(x),dx), can be determined based on the knowledge of V′, and vice versa. Figure  compares Figs. a and , highlighting in magenta the corresponding parts of the two graphs. From the mathematical point of view, the previous result is a consequence of the proof of Theorem 3.

It is worth pointing out that by combining Theorem 3 and Corollary 4, a valuable property emerges, i.e. the global mean temperature of the functional minimizer is non-decreasing with respect to q. In other words, as the concentration of CO2 rises, the global mean temperature increases. Additionally, through this monotonicity and Froda's theorem Theorem 4.30, we also establish that the global mean temperature is continuous, except for, at most, a countable number of upward jumps.

In the second part of this section, we demonstrate the applicability of Corollary 4 to other reaction–diffusion equations. We use as an example a spatially heterogeneous Allen–Cahn equation (ACE), already considered in . For an initial condition ũ, this model is given by 16∂tu=1100Δu+u(1-u2)+q+12cos⁡(πx),x∈(-1,1),t>0,ux(t,-1)=ux(t,1)=0,t≥0,u∣t=0=ũ. The associated elliptic problem for u=u(x) is 170=1100u′′+u(1-u2)+q+12cos⁡(πx),x∈(-1,1),u′(-1)=u′(1)=0. In this case, the potential functional takes the form Jq(u)=∫-11u4(x)4-u2(x)2-u(x)(q+12cos⁡(πx))dx, and all the properties discussed in Sect.  and  can be extended to this equation. Specifically, Theorems 1, 2 and 3 hold. But in this case, the structure of the bifurcation diagram is more complex, even if symmetric with respect to q=0. Indeed, through numerical experiments, it is possible to deduce the existence of 0<q4<q5 such that (a) for |q|>q5 or |q|<q4, there exists a single steady-state solution, which is stable, (b) for q4<|q|<q5 there are three steady-state solutions, two of which are stable while the third is unstable. Further, q=q4,q5 are bifurcation points of the saddle-node type. We denote by u1 the steady-state solution for q<-q5, by u2 and u3 the steady-state solutions appearing at the bifurcation point q=-q5 and existing for -q5<q<-q4 in addition to u1, and by u4 and u5 the steady-state solutions appearing at q=q4 and existing for q4<q<q5 in addition to u3. Regarding the potential functional Jq, in this case there exists q6∈(q4,q5) such that u1 is the global minimum point for the functional for q<-q6, and u3 is the global minimum point for -q6<q<q6, while u5 becomes the global minimum point for q>q6. A picture for the bifurcation diagram just described and the value function is shown in Fig. . Note that q=±q6 represents the only values of the parameter q for which the value function is not differentiable and also the only points in which the global minimizer of the variational problem is not unique.