Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
Which peer-reviewed papers argue natural variability explains recent warming and what are their main critiques?
Executive summary
Peer‑reviewed papers that emphasize natural variability as a major explanation for recent warming exist, but the mainstream literature and synthesis reports (IPCC/major journals) conclude natural variability cannot fully explain the global warming trend of the late 20th and early 21st centuries [1]. Examples of peer‑reviewed work that attribute substantial roles to internal variability include long‑term analyses of ocean‑driven multidecadal change (PNAS 2009) and recent studies that quantify how natural variability influences the timing of overshoot of policy thresholds (Communications Earth & Environment 2025) [2] [3].
1. The papers that foreground natural variability — who and what they argue
A prominent earlier peer‑reviewed example is “Long‑term natural variability and 20th century climate change” (PNAS, 2009), which presents techniques to separate a multidecadal, ocean‑rooted variability signal from the externally forced trend and shows an externally forced warming signal that is monotonic once that internal component is removed, while arguing that long‑term natural variability could underlie multi‑decadal non‑monotonic behavior in the 20th century [2]. More recent peer‑reviewed work focuses not on overturning anthropogenic attribution but on quantifying natural variability’s role in particular aspects of recent warming — for example, a 2025 Communications Earth & Environment paper uses observed persistence properties to quantify natural variability and assesses how that affects the timing of “overshoot” of Paris targets, implicitly emphasizing that internal variability can shift apparent near‑term trends [3].
2. The core critiques made in those papers
Authors who emphasize natural variability typically make two linked technical points: (a) multidecadal internal modes — e.g., ocean circulation patterns like PDO/AMO — can produce global or hemispheric temperature excursions that mimic forced trends on decadal scales, and (b) estimating and removing that hidden variability changes the perceived shape and timing of warming in observational records [2] [3]. The Communications Earth & Environment paper argues that data‑driven quantification of persistence increases uncertainty in projected timing for policy thresholds because natural variability superimposes on forced warming [3].
3. How major syntheses and other peer‑reviewed work respond
Comprehensive assessments and detection‑attribution studies find that simulations including only natural forcings cannot reproduce the magnitude of observed warming; only simulations including anthropogenic forcings match observed global temperature change, leading to the conclusion that natural causes alone are insufficient (IPCC FAQ material cited here) [1]. Other studies reinforce that while internal variability matters regionally and on decadal scales, its contribution to long‑term global mean warming is limited — for example, multi‑basin analyses find less than ~10% contribution from Atlantic and Pacific internal variability for the second half of the 20th century in some assessments discussed in the review literature [4].
4. Methodological dissents and the limits of inference
Papers stressing natural variability often rely on statistical decomposition of observed records or on single‑model large‑ensemble realizations to estimate the range of unforced trends; critics point out that those methods hinge sensitively on choices of filters, indices, and the representativeness of ocean indices [2]. Conversely, other work shows models sometimes overestimate observed natural variability (tropospheric decade‑to‑decade variability), which would bias attempts to assign more warming to internal variability if models “too noisy” are used as reference [5]. Thus, methodological uncertainty cuts both ways: decompositions can expose hidden variability but model biases complicate how much weight to give those decompositions [2] [5].
5. What the recent literature actually agrees on
Multiple lines of evidence in the recent literature and synthesis reports converge: internal (natural) variability affects regional and decadal fluctuations and increases uncertainty in near‑term projections or attribution of single events, but natural factors and internal variability alone do not reproduce the observed global warming trend — anthropogenic forcings are required to match magnitude and pattern [1] [3]. Studies focused on variability typically do not claim that natural variability explains all recent warming; rather they refine how variability alters timing, regional expression, and uncertainty [3] [2].
6. Hidden agendas and interpretive risk in public debate
The technical nuance — that natural variability can modulate near‑term trends but not replace anthropogenic forcing for the long‑term global mean — creates a communication opening that some actors use to argue the warming is “natural.” The peer‑reviewed papers that highlight variability usually aim to improve attribution and projection methods, not to negate anthropogenic responsibility; treating those papers as disproof of human influence overlooks their own stated limitations and the contrasting conclusions in synthesis assessments [2] [1].
7. Bottom line for readers and policymakers
If your question is “Which peer‑reviewed papers argue natural variability explains recent warming?” — the literature contains papers emphasizing a larger role for internal variability in specific contexts (PNAS 2009; Communications Earth & Environment 2025), but major attribution studies and the IPCC say natural variability alone cannot explain the global trend and that anthropogenic forcings are necessary to reproduce observed warming [2] [3] [1]. Available sources do not mention a peer‑reviewed overturning of that consensus (not found in current reporting).