You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Spec mapping: AQA 7037, Paper 1 (Physical), §3.1.1 — feedbacks within and between the water and carbon cycles, the concept of dynamic equilibrium, and the implications of change for the climate system. This depth lesson assumes familiarity with the individual stores/fluxes and develops the systems theory of feedback (positive vs negative, gain, non-linearity), the major climate feedbacks quantitatively, and the tipping-point literature. AOs exercised: AO1 (precise feedback mechanisms; tipping-element thresholds), AO2 (reasoning about how feedbacks couple the water and carbon cycles and determine stability), AO3 (interpreting feedback-strength data, sea-ice and AMOC time series). Synoptic links run to Hazards (abrupt change, cascading risk) and Global systems (governance of irreversible risk).
A feedback is a loop in which the output of a process alters its own driver. Feedbacks decide whether the climate system, when perturbed, settles back to equilibrium or accelerates towards a new state — and they are the single most important reason the response to a given CO₂ rise is uncertain. The depth treatment requires distinguishing feedback sign from feedback gain, recognising that the water and carbon cycles are coupled through feedback, and understanding non-linearity — the property that makes tipping points possible.
Positive (amplifying) feedback reinforces the initial change, driving the system further from its starting state. It is destabilising and can produce rapid, self-sustaining change.
Negative (damping) feedback opposes the initial change, restoring the system towards its starting state. It is stabilising and maintains dynamic equilibrium.
flowchart LR
subgraph POS[Positive feedback - destabilising]
A1[Initial warming] --> A2[Process amplifies driver]
A2 --> A3[More warming]
A3 --> A1
end
subgraph NEG[Negative feedback - stabilising]
B1[Initial warming] --> B2[Process opposes driver]
B2 --> B3[Cooling tendency]
B3 --> B4[Return towards equilibrium]
end
A more advanced framing treats each feedback as a gain factor. If the no-feedback warming for doubled CO₂ is ΔT0≈1.2∘C (the Planck response), the realised warming is:
ΔT=1−fΔT0
where f is the net feedback factor (sum of individual gains). For f≈0.6:
ΔT=1−0.61.2=3.0∘C,
which is exactly the IPCC AR6 best-estimate equilibrium climate sensitivity (ECS). As f→1 the warming diverges — the mathematical signature of a runaway. This equation shows why the same forcing can yield very different warming depending on the feedback balance, and it is a strong AO3/AO2 device.
Key point: Real systems combine positive and negative feedbacks. Their net sum (the overall f) determines stability. The danger of contemporary warming is that several large positive feedbacks (ice-albedo, water vapour, carbon-cycle) are pushing f upward faster than the slow negative feedbacks (weathering thermostat) can respond.
The most frequently examined climate feedback.
Quantified evidence:
The feedback also runs in reverse: in glacial inceptions, growing ice raises albedo and amplifies the cooling initiated by Milankovitch forcing — explaining how small orbital changes produce large ice ages.
This feedback directly links the water cycle to the energy/carbon climate.
Significance: the water-vapour feedback roughly doubles the warming from CO₂ alone (Held & Soden, 2000) and is the single largest positive feedback in the climate system — the clearest demonstration that the water and carbon cycles cannot be understood in isolation.
More vapour means more cloud, but cloud feedback is mixed:
The net cloud feedback is assessed as slightly positive but remains the largest single uncertainty in climate sensitivity (IPCC AR6) — and the main reason ECS still spans 2.5–4.0 °C.
Quantified risk: permafrost holds ~1,400–1,700 GtC (≈ twice the atmosphere); high-warming pathways could release ~150–200 GtC by 2100, plausibly adding ~0.1–0.3 °C beyond current projections — an amount largely outside standard emissions budgets and effectively irreversible.
Methane hydrates (clathrates) — methane caged in ice within marine continental-margin sediments and beneath permafrost — are a further, more uncertain reservoir. Their destabilisation could release large methane pulses, though the consensus is that a rapid, catastrophic "clathrate gun" this century is unlikely; the slower terrestrial permafrost feedback is the nearer-term concern. Stating this measured assessment (rather than overclaiming catastrophe) is itself an evaluation mark.
Evidence: severe droughts in 2005, 2010, 2015–16 and 2023–24 each raised mortality and fire. In 2010 the basin briefly flipped to a net source, releasing on the order of 1.6 GtC instead of its usual modest uptake (Lewis et al., 2011). Models place a possible dieback tipping zone around 20–25% deforestation combined with 2–3 °C warming; roughly 17–20% of the original forest has already been cleared — uncomfortably close. This is the textbook example of a feedback that couples the hydrological and carbon cycles.
Evidence and assessment: the AMOC may have weakened by ~15% since the mid-twentieth century (Caesar et al., 2018, based on a sea-surface-temperature "fingerprint"). IPCC AR6 judges further weakening this century very likely, but a complete collapse before 2100 of low likelihood — while stressing it cannot be ruled out and would be high-impact. A shutdown would disrupt European climate, shift tropical monsoons and alter Atlantic ecosystems. The combination of low probability and very high, irreversible impact makes AMOC a defining example of deep-uncertainty risk.
A tipping point is a critical threshold beyond which a system shifts, often abruptly and irreversibly (on human timescales), to a qualitatively different state, with the change becoming self-sustaining once triggered. Tipping points arise from non-linearity: near the threshold a small additional forcing produces a disproportionately large, often unstoppable response (in feedback terms, f briefly exceeds 1 for that element).
| Tipping element | Estimated threshold (above pre-industrial) | Potential impact |
|---|---|---|
| Greenland Ice Sheet | ~1.5–3.0 °C | ~7.4 m sea-level rise over millennia |
| West Antarctic Ice Sheet | ~1.5–3.0 °C | ~3.3 m sea-level rise over centuries |
| Amazon dieback | ~2.0–3.0 °C + 20–25% deforestation | ~150–200 GtC released; biodiversity collapse |
| Boreal forest dieback | ~3.0–5.0 °C | Sink → source switch |
| AMOC collapse | ~3.0–5.0 °C (deeply uncertain) | NW Europe cooling; monsoon disruption |
| Permafrost (abrupt thaw) | ~1.5–2.0 °C (already underway) | 150–200+ GtC by 2100 |
| Warm-water coral reefs | ~1.5 °C (70–90% loss); ~2.0 °C (~99% loss) | Marine biodiversity and livelihood crisis |
Tipping cascades. Elements can interact, so crossing one threshold raises the odds of crossing others (Lenton et al., 2019). For example, Greenland melt freshens the North Atlantic → weakens AMOC → shifts the tropical rain belt → stresses the Amazon. Cascading, self-amplifying risk is the most serious AO2 point in the whole topic, and it is why "every fraction of a degree matters".
Not all feedbacks amplify; the stabilising ones are why Earth has remained habitable.
The crucial AO2 insight is the timescale mismatch: the dangerous positive feedbacks act on years-to-decades, whereas the strongest negative feedback (weathering) acts on millennia — so the system has little fast-acting self-correction against the current forcing.
Suppose individual feedback factors (gains, dimensionless) for doubled CO₂ are estimated as: water vapour +0.50, lapse-rate −0.20, surface-albedo +0.10, cloud +0.20.
Manipulate. Net feedback factor:
f=0.50−0.20+0.10+0.20=0.60.
Realised warming, with ΔT0=1.2 °C:
ΔT=1−0.601.2=3.0∘C.
If the (most uncertain) cloud term were instead +0.35, then f=0.75 and ΔT=1.2/0.25=4.8 °C.
Explain and evaluate. A change of just 0.15 in one uncertain feedback raises projected warming from 3.0 to 4.8 °C — illustrating both why ECS is uncertain and why the cloud feedback dominates that uncertainty. The non-linear 1/(1−f) form means uncertainty grows rapidly as f approaches 1, which is precisely the mathematical reason high-sensitivity outcomes cannot be dismissed. Quoting the values as estimates and identifying clouds as the key uncertainty is what earns the evaluation marks.
The geological record provides real evidence that feedbacks and thresholds operate, which strengthens (and disciplines) projections of the future.
Glacial–interglacial cycles. Ice cores show that over the past 800,000 years, temperature and CO₂ moved together, with CO₂ oscillating between ~180 and ~280 ppm. The orbital (Milankovitch) trigger is small, so the large temperature swings require amplifying feedbacks — chiefly ice-albedo and the CO₂ feedback (a warming ocean and thawing land release CO₂, which warms further). This is direct palaeo-evidence that the ice-albedo and carbon-cycle feedbacks are real and powerful, and it is why models that omit them cannot reproduce the ice ages.
The Palaeocene–Eocene Thermal Maximum (PETM, ~56 million years ago). A massive injection of carbon (thousands of GtC) over a few thousand years raised global temperatures by ~5–8 °C and severely acidified the oceans (recorded as a sharp carbon-isotope excursion and a dissolution horizon in deep-sea sediments). The PETM is studied as the closest geological analogue to a large, relatively rapid carbon release — and as evidence that carbon-cycle feedbacks (possibly including methane hydrates and permafrost-equivalent stores) can amplify an initial perturbation. Crucially, the PETM unfolded over millennia, whereas the modern release is unfolding over centuries — so it is, if anything, a conservative analogue. Deploying deep-time evidence like this signals an unusually sophisticated grasp of why feedbacks matter for the future.
Most A-Level answers treat the carbon cycle as a driver of warming, but a depth treatment recognises it as a feedback. As warming proceeds:
The net effect is that the land and ocean sinks weaken as the planet warms, so a larger fraction of each tonne emitted stays airborne — an amplifying feedback that the IPCC now represents explicitly. This is why the same emissions produce more warming in a high-warming world than a low-warming one, and why "carbon-climate feedback" appears as an additional uncertainty band on long-term projections. Framing the carbon cycle as both driver and feedback is one of the most powerful synoptic moves available in this topic, binding the carbon-store and feedback lessons together.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.