Transient Climate Response (TCR)

Updated 22 January 2026

Transient Climate Response (TCR) is a measure of global warming at CO₂ doubling achieved over 70 years, capturing both fast climate feedbacks and early ocean heat uptake.
Various methodologies, including GCMs, energy-balance models, and regression analyses, are employed to estimate TCR with distinct uncertainty bounds.
TCR informs near-term climate projections and policy decisions by linking process-level feedbacks with integrated assessments of mitigation and economic impacts.

The Transient Climate Response (TCR) is a central metric in climate science quantifying the transient (non-equilibrium) surface warming at the time atmospheric CO₂ doubles under a standardized increasing-forcing scenario. Defined as the simulated or observed increase in global-mean surface temperature at the moment of CO₂ doubling, typically achieved by a 1 % yr⁻¹ CO₂ ramp over 70 years, TCR captures the climate’s response incorporating both fast feedbacks and oceanic heat uptake, but before slow components reach equilibrium. This parameter is foundational in both physical climate modeling and integrated assessment, linking process-scale feedbacks with policy-relevant projections on multidecadal horizons.

1. Formal Definition and Theoretical Framework

TCR is formally defined as the change in global-mean surface temperature, denoted as $T_{\mathrm{AT}}(t)$ , after 70 years of a 1 % yr⁻¹ increase in CO₂, resulting in a doubling of atmospheric concentration:

$\mathrm{TCR} = T_{\mathrm{AT}}(t_{\mathrm{dbl}}) - T_{\mathrm{AT}}(0), \quad t_\mathrm{dbl}=70\,\mathrm{yr}$

where $T_{\mathrm{AT}}(0)$ is the pre-industrial global-mean temperature and $T_{\mathrm{AT}}(t_{\mathrm{dbl}})$ the temperature at the time of doubling (Folini et al., 2021). This protocol has become the community standard, aligning closely with CMIP and IPCC practices. TCR should be distinguished from Equilibrium Climate Sensitivity (ECS), which characterizes the long-time, equilibrated system response under the same increment in radiative forcing.

The TCR quantifies a system’s "inertia-dominated" warming, with oceanic heat uptake and time-dependent feedbacks leading the response well before the much slower components (deep ocean circulation, multi-centennial carbon cycle feedbacks) fully equilibrate (Lembo et al., 2019, Royce et al., 2013).

2. Methodologies for Computing TCR

Multiple frameworks are employed to compute or constrain TCR:

A. General Circulation Model (GCM) Protocols

State-of-the-art GCM studies (e.g., MPI-ESM v1.2) utilize ensembles of step or ramp CO₂ increase experiments:

Step-responses: Instantaneously double CO₂ and track global-mean surface temperature over centuries.
1 % yr⁻¹ ramp: CO₂ increases by 1 % per year; TCR is evaluated as the ensemble mean warming at year 70 (Lembo et al., 2019).

The system's response is expressed as a convolution of the radiative forcing history with the Green’s function $G_{\Phi}(t)$ for each climatic observable $\Phi$ :

$\langle \Phi \rangle (t) - \langle \Phi \rangle_0 \simeq \int_0^t G_{\Phi}(\tau) f(t-\tau) d\tau$

The convolution approach enables flexible computation under arbitrary forcing protocols, not limited to the canonical ramp (Lembo et al., 2019).

B. Energy-Balance Models and Climate Emulators

Two-layer energy-balance models (EBMs), widely used in climate emulation and economics, encode the TCR via parameters representing mixed-layer (fast) and deep-ocean (slow) heat uptake:

$C \frac{dT_1}{dt} = F(t) - \lambda T_1(t) - \gamma [T_1(t) - T_2(t)],$

$C_0 \frac{dT_2}{dt} = \gamma [T_1(t) - T_2(t)]$

where $T_1$ is mixed-layer temperature, $T_2$ the deep-ocean, $\lambda$ the feedback parameter, and $\gamma$ the deep-ocean exchange rate. TCR is nearly

$\mathrm{TCR} \approx \frac{F_{2\times}}{\lambda + \gamma}$

with corrections for deep-ocean lag. Analytical and numerical solutions are available via Laplace transform and eigenmode decompositions (Bauer et al., 21 Jul 2025, Folini et al., 2021).

C. Empirical and Regression-Based Estimates

Emergent-constraint and regression analyses apply to observed or reconstructed time series. For instance, Stefani applies a detrended regression protocol:

$\Delta T'_i = S \log_2 \left( \frac{C_i}{280\,\mathrm{ppm}} \right ) + \varepsilon'_i$

where $S$ is the transient climate response per doubling, extracted after accounting for other covariates (e.g., aa-index) (Stefani, 16 Jan 2026).

Others apply time-dependent regression metrics, regressing temperature anomalies onto time-varying forcings such as TSI (solar cycle) or CO₂ radiative forcing (Li et al., 2023).

3. Numerical Results and Uncertainty Bounds

TCR estimates, and their uncertainties, arise from both physical modeling and empirical constraints.

Estimation Method	TCR K	Uncertainty (95%/1 $\sigma$ )
CMIP5 Multi-model Mean (MMM)	$\approx$ 1.9	1.3, 2.3
Response Operator (MPI-ESM v1.2)	1.8–1.9	$\pm 0.2$
Solar-cycle emergent constraint	2.2	1.9, 2.7
Regression (SST, aa-index)	1.1–1.4	$\pm 0.15$ –0.3
Fractal response (van Hateren)	1.5 (strong solar); 1.9 (weak solar)	$\pm 0.2$ –0.3
Long-memory EBM	2.1	$\sim$ 15–20% (model param. error)
Simple EBM (Kramm & Dlugi)	0.5–3	"O( $\pm$ 1)" (dominant param. error)

References: (Lembo et al., 2019, Folini et al., 2021, Li et al., 2023, Stefani, 16 Jan 2026, Hateren, 2011, Rypdal et al., 2013, Kramm et al., 2010)

The spread in TCR estimates reflects both intrinsic parameter uncertainty (e.g., deep-ocean heat uptake rate, radiative feedback strength) and structural model differences, especially over whether slow feedback processes are active within the 70-year horizon (Bauer et al., 21 Jul 2025, Royce et al., 2013).

The solar-cycle-based emergent constraint yields the narrowest uncertainty envelope to date, with a central TCR of 2.2 °C and a range $\sim23\%$ about the mean (Li et al., 2023). By contrast, historical-warming and model-based methods often exhibit uncertainty exceeding 70–140% of the central value.

4. Fast versus Slow Feedbacks and Ocean Heat Uptake

TCR is determined largely by fast feedbacks (water vapor, clouds, sea-ice) and the efficiency of ocean mixed-layer heat uptake, but is significantly muted relative to ECS due to incomplete deep-ocean adjustment.

The impulse-response formalism demonstrates that the Green’s function for temperature, $G_{T_{2m}}(\tau)$ , exhibits a rapid initial rise (fast feedback) and a long tail (slow, oceanic processes) (Lembo et al., 2019, Hateren, 2011):

Typically, ~50% of equilibrium warming occurs in the first few years, ~30% in the next 10–30 years, and the remainder over centuries.
In models with a dynamic ocean, the ratio $\mathrm{TCR}/\mathrm{ECS}$ can be as low as 0.5, compared to 0.85 in models without deep-ocean heat uptake (Lembo et al., 2019).
The slow mode is responsible for “warming in the pipeline” post-CO₂ stabilization, causing surface temperatures to converge to ECS only over multi-century timescales (Royce et al., 2013, Bauer et al., 21 Jul 2025).

Ocean heat uptake (OHU) is a key diagnostic: its Green’s function has an instantaneous spike (initial radiative imbalance) followed by a slow decay, structurally limiting further surface warming until equilibrium is restored (Lembo et al., 2019).

5. Impact on Climate Projections and Integrated Assessment

The magnitude and uncertainty of TCR directly control projections of 21st-century warming and the social cost of carbon (SCC):

In climate-economic emulators, varying TCR between 1.3–2.3 K (spanning observed GCMs) yields a factor-of-four difference in SCC under quadratic damages, and a factor-of-eight under cubic damages (Folini et al., 2021).
CMIP-based projections of high-emission scenarios (e.g., RCP8.5) see year-2100 warming ranging from ~3 K (low-TCR realizations) to ~5 K (high-TCR), even with identical ECS (Folini et al., 2021)
The TCR is much more rapidly constrained than ECS in Bayesian learning frameworks, because contemporary and near-term observations predominantly constrain the fast mode (Bauer et al., 21 Jul 2025).

The policy relevance of TCR is thus immediate: it governs the scale and pace of mitigation needed to avoid breaching temperature targets within the next one to two generations, as warming on TCR timescales is “locked in” by both physical inertia and current policy commitments.

6. Model Structural Choices and Time-Scale Dependent Sensitivity

Some frameworks abandon the notion of a unique equilibrium sensitivity in favor of time-scale dependent or frequency-dependent climate response (“fractal” or long-range-memory models). In these models:

TCR is interpreted as the short-to-intermediate time-scale response, growing with $\tau^{\beta/2}$ in long-memory models, where $\beta$ is the memory exponent (Rypdal et al., 2013).
The fractal response function allows decomposition of warming into contributions realized over $0.5-2$, $8-32$, and $128-512$ year time scales, clarifying why TCR can differ significantly from ECS (Hateren, 2011).
In such models, adopted for both empirical fit and theoretical reasons, the ultimate equilibrium is either undefined or realized only on millennium scales, reinforcing the primacy of TCR in policy-focused time windows (Rypdal et al., 2013).

7. Controversies, Limitations, and Ongoing Challenges

Simple energy-balance models highlight that uncertainties in feedback parameters ( $\lambda$ ), effective heat capacity, and the inclusion (or neglect) of key physical processes (latent/sensible heat exchange, cloud-radiative interactions) propagate into TCR estimates. Kramm & Dlugi argue that the uncertainty in such parameters can exceed the anthropogenic perturbation itself, placing limits on what can be inferred from global-mean energy-balance alone (Kramm et al., 2010).

Recent developments emphasize narrowing TCR uncertainties via:

Emergent constraints using independent lines of evidence (e.g., solar-cycle response) (Li et al., 2023)
Assimilating near-term observations (surface temperature, ocean heat content) and accounting for serially correlated internal variability to rapidly learn TCR, even as ECS remains uncertain (Bauer et al., 21 Jul 2025).
Unified response-operator frameworks predicting TCR for arbitrary forcing pathways, based on Green’s functions from a single step experiment (Lembo et al., 2019).

Nonetheless, parameter uncertainty, aerosol-forcing ambiguity, and structural limitations in 1D models all remain active frontiers, and different methodologies yield central TCR values ranging from $\sim$ 1.1 K to $>2.2$  K depending on assumptions and data periods.

In sum, the Transient Climate Response is mathematically and operationally well defined, traceable to rigorous physical and empirical methodologies, and remains the key metric for quantifying multidecadal climate risk under anthropogenic forcing. Its value and uncertainty delimit both physical projections and the economic calculus of mitigation, and future progress in constraining TCR hinges on refining both the theoretical understanding of climate system response and the integration of new observational constraints.