Science1 distinct publisher2 min readPublished
Redwood National and State Parks hold nearly half the world's tallest trees. One whiplash winter took the over-100-metre count from 950 to 825, mostly by breaking crowns rather than trunks.
The Scientist · Science desk

Compiled by The ScientistSomething wrong?How this is made
The 100 metre line is a measuring tape's verdict, not a mortality count. Between the 2022 peak and the post-storm survey, 125 trees dropped out of the class, about 13 percent of the tally [1]. The study attributes only 11 percent of documented losses to treefall [8], which puts roughly 14 trees on the ground and around 111 still rooted and simply shorter [2]. Redwood answers damage by resprouting, and the authors expect damaged crowns to recover into vigorous reiterated tops [9].
The biomass side is harder to shrug off. Priced in tonnes rather than years, the Humboldt County losses come to roughly 20 to 104 tonnes per hectare [5], against a pre-storm accumulation rate of 10 to 13 tonnes per hectare per year through 2007 to 2022 [10]. On the worst hectares the estimate exceeded 500 tonnes per hectare, more than the entire standing biomass of forests counted as high-biomass elsewhere [12]. At the pre-storm growth rate, that is 38 to 50 years of accrual gone in one winter [3].
The wood did not leave the site. What came down was mostly decay-resistant heartwood, which keeps storing carbon as woody debris on the forest floor, and the authors reckon the crown damage roughly rivalled the live-tree biomass lost to whole trees falling [14]. That is the useful distinction: the maps lost a lot of live biomass, the atmosphere gained much less, and the carbon mainly changed pools.
Then there is siting. Tall forests exist where they do partly because strong winds are infrequent there [16]. The traits that let this species reach such heights, extreme decay resistance, survival of high-severity fire, shade tolerance and prolific trunk reiteration [17], are answers to rot, flame and competition for light. None of them is an answer to sustained wind on a crown carrying a rare snow load after a run of atmospheric rivers [3]. Fire and logging shaped how these reserves were drawn and defended; wind loading on a saturated slope was treated as background noise.
The geography leaves no room to spread the risk. Primary forest inside the parks covers 16,402 hectares against 38,379 hectares of secondary forest regrowing from 20th-century logging [2], roughly 30 percent of the wooded area [4], and nearly half the world's living trees over 100 metres stand inside that fraction [1]. A hazard that arrives from the sky and lands on the tallest crowns hits the concentrated part first, which is the opposite of how a fire perimeter or a timber sale boundary behaves.
Ranked by verification strength, evidence, and original report placement.
RNSP contains 16,402 hectares (40,500 acres) of high-biomass primary old-growth forest plus 38,379 hectares (94,800 acres) of much shorter, low-biomass secondary forest regenerating from 20th-century logging.
A new study in Forest Ecology and Management combined ground-based measurements with repeated airborne laser scanning from 2007 to 2025 to inventory the tallest redwoods and quantify forest biomass.
Field measurements of 1,076 trees associated with 400 laser-derived tree objects were used to develop biomass estimation equations across RNSP.
Only 11 percent of documented losses from the tall-tree population were treefall events.
Widespread reductions in live-tree biomass occurred across the Humboldt County portion of RNSP, with losses equalling 2 to 8 years of pre-storm biomass accumulation.
California's Redwood National and State Parks protect nearly half the global population of living trees more than 100 metres (328 feet) tall.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Peer-reviewed repeat-survey measurement, self-qualified and single-sourced
The core numbers rest on a named journal study with a stated design: repeated airborne laser scanning from 2007 to 2025, segmentation into tree objects, and biomass equations calibrated on 1,076 field-measured trees across 400 objects. Counts (950 to 825), the 11 percent treefall share, and pre-storm growth of 10-13 t/ha/yr are specific and attributable. Evidence is held below high because only one publisher account exists, no uncertainty ranges are given, and the source itself flags that its biomass-change estimates may be exaggerated where crowns broke but trunks survived.
No adoption signal in supplied material
The cluster contains one research-findings write-up. Beyond the study's publication, the supplied material reports no deployments, programme rollouts, procurement, pricing, licensing or third-party uptake of the monitoring approach or of any management response, so adoption cannot be scored without inferring facts the sources do not contain.
Broadly aligned; headline arithmetic runs ahead of the mechanism
The framing of 125 redwoods 'dropping out' and of losses exceeding 500 t/ha is dramatic relative to what was measured: 89 percent of the losses were crown breakage on trees whose massive trunks remain standing, the downed heartwood still stores carbon, and the source concedes its laser-derived biomass losses may be somewhat exaggerated. The gap is small and positive rather than large because the qualifications travel with the claim in both the dek and the body, and the counts themselves are directly measured.
Author-framed research summary with disclosed limitations
The only account presents the study team's own framing, including an unmeasured expectation that damaged crowns will recover quickly and a conclusion supporting the premise of ongoing restoration logging inside the reserve - positions the authors have a professional stake in. Incentive load is moderate rather than high because the same account volunteers the countervailing methodological caveat about exaggerated biomass loss, and the supplied material discloses no commercial, funding or advocacy interest.
Solid on counts, thin on corroboration and consequence
Confidence is moderate: the tall-tree inventory change, forest areas and primary/secondary biomass contrast are specific and method-backed, so the central factual spine is dependable. It is held down by single-publisher sourcing, absent uncertainty ranges, a self-flagged bias in the biomass-change metric, no adoption or management-response evidence, and a recovery expectation that remains a forecast.
science
What you expect from your own old age shows up a decade later in who you still see1 distinct publisher
science
The sharp crystal seed was never there: 8,000 nuclei imaged atom by atom, all gradients1 distinct publisher
science
Eastern US extreme rain is pooling into fewer, wider storms, and station records hide it1 distinct publisher
science
Women's work friendships grow by introduction, men's by selection, Durham study reports1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · August 25, 2026