Skip to content
NatureHQ

Ecologyphenomenon

Soil carbon

The leaf that rots fastest may leave the most carbon behind.

Carbon in soil, mostly as the remains of things that lived — and the part that lasts is substantially what microbes made, bound to mineral surfaces, rather than the plant material they could not digest.

The intuition is that the tough, indigestible parts of plants are what accumulate — that soil carbon is the residue nothing could eat. The evidence points somewhere less obvious. Litter that decomposes readily is efficiently assimilated by microbes, and microbial residues bind strongly to mineral surfaces, so easily decomposed material can contribute more to the long-lasting fraction than recalcitrant material does. The leaf that rots fastest may leave the most behind. Recalcitrant litter tends to persist instead as particulate organic matter: recognisable fragments, comparatively unprotected, and turning over faster than the mineral-associated pool. Distinguishing those two fractions is the practical core of modern soil carbon science, because they respond to disturbance and to warming quite differently, and a measurement of total soil carbon conceals which one is changing. Two things follow that are worth holding onto. First, mineral surfaces are finite: there is only so much surface for organic matter to bind to, so the capacity to stabilise carbon in this way has a ceiling that varies with a soil’s mineralogy. Second, the persistence is conditional rather than permanent. Carbon protected by aggregates or by mineral association can be released when the structure is broken, and describing soil as storing carbon "permanently" mistakes a circumstance that has held for centuries for a property that will hold regardless of what happens to the soil.

Early coverage · 39% complete · reviewed 2026-09-04

What this page covers

Soil carbon derives from plants, from the microbes that process them, and from animals. The relative contributions are an active research question rather than a settled split.

Often confused with: Humus as a single stable substance, which modern analysis does not support; Undecomposed plant material, which is a smaller part of the persistent fraction than assumed; A permanent store, when much of it is turning over continuously

Quick facts

Mostly microbial
The persistent fraction is largely what microbes made, not what they refused
Two fractions
Mineral-associated, which lasts; particulate, which turns over faster
Protection, not chemistry
Persistence is about reachability
And it has a ceiling
Mineral surface area is finite, and varies between soils

Where soil carbon comes from

Less of it is undigested plant than you would guess.

The leaf that rots fastest can contribute more lasting soil carbon than the one that resists rotting — because what persists is largely what the microbes made of it, bound to mineral surfaces.

Well supported

Good evidence backs this, though some details remain open.

Readily decomposed, high-quality litter can contribute disproportionately to persistent mineral-associated organic matter, because efficient microbial assimilation produces residues that bind to mineral surfaces. Recalcitrant litter contributes more to particulate organic matter with shorter residence time.

Who this applies to
Terrestrial soils; the balance depends on mineralogy and on the litter and microbial community present.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Plantae, Bacteria, Fungi
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

Supported by isotopic and biomarker evidence and now widely adopted, while partitioning plant-derived from microbe-derived carbon carries method-dependent assumptions.

How far it can be extended

A framework supported in a growing set of systems rather than a universal result; contributions vary with soil mineralogy.

Caveats

  • This concerns which fraction persists, not total carbon input, and a system can gain carbon by either route.
  • Mineral surface availability is finite, so the mechanism has a ceiling that varies between soils.

Still unanswered

  • How saturable mineral-associated carbon is in practice, and what happens to inputs once that capacity is approached.

Last reviewed 2026-09-04

The evidence (2 studies)

The route is worth following. A microbe consumes an easily digested compound, uses part of it for energy and part to build itself, and eventually dies. Its remains — cell wall fragments, proteins, the chitin and melanin of fungal walls — are chemically sticky and bind to clay and mineral surfaces, where they become very difficult for the next organism to reach. What persists is not the leaf. It is a succession of microbes that ate the leaf.

Two pools that behave differently
FractionWhat it isHow it behaves
Mineral-associatedSmall molecules, largely microbial, bound to mineral surfacesLong residence times; limited by available surface
ParticulateRecognisable plant fragments, relatively unprotectedFaster turnover; responds quickly to disturbance and warming

Why it lasts, and what ends that

A circumstance, not a permanent property.

Old soil carbon is not made of indestructible molecules. It persists because decomposers cannot reach it — locked in aggregates or bound to minerals. Break the soil apart and it becomes available again.

Well supported

Good evidence backs this, though some details remain open.

Soil organic matter is a continuum of progressively decomposing fragments rather than a set of distinct stable humic macromolecules. Persistence arises from physical protection within aggregates, association with mineral surfaces and consequent inaccessibility to decomposer enzymes, rather than from intrinsic molecular recalcitrance.

Who this applies to
Soil organic matter generally; the relative weight of each mechanism varies by soil.
Studied in
Bacteria, Fungi, Plantae
Why we rate it this way, and what the caveats are
Well supportedHigh confidence

The shift is well established in the current soil-science literature, though quantifying each protection mechanism in a given soil remains difficult.

How far it can be extended

Supported by spectroscopic and isotopic evidence across soil types, and by the response of soils to physical disturbance.

Caveats

  • The older extraction-based literature is large, and some disagreement about terminology and mechanism continues.
  • Chemistry has not become irrelevant: it influences how readily material is used, just not how long it ultimately survives.

Still unanswered

  • How to quantify the relative contribution of mineral association, aggregate occlusion and other protection in a given soil.

Last reviewed 2026-09-04

The evidence (2 studies)

Because persistence depends on inaccessibility rather than on chemistry, anything that restores access ends it. Physical disruption breaks aggregates open. Erosion moves protected material into conditions where it is exposed. Changes in moisture and temperature alter which organisms are active and how fast they work. None of this requires the carbon to be different from the carbon that stayed put for a century.

That is also the reason to be careful with the phrase "carbon storage". Soil does hold enormous quantities of carbon, and the holding is conditional. Describing it as stored implies a vault; describing it as protected implies a condition that can be removed, which is closer to what the evidence supports.

Related

  • Soil

    The structure that does the protecting

  • Decomposition

    The process that produces all of this

  • Detritus

    What arrives at the top of the profile

What is not settled

  • How saturable is mineral-associated carbon in practice?

    Why it matters: If mineral surfaces are finite and many soils are near capacity, then adding carbon inputs to soils will produce far less lasting storage than a linear assumption implies. Estimates of how much additional carbon soils can hold depend directly on this.

    What would settle it: Long-term input manipulations on soils of differing mineralogy, measuring the fractions separately rather than total carbon.

  • How much of the persistent fraction is microbial rather than plant?

    Why it matters: The two imply different levers. If persistence runs through microbial efficiency, what matters is what microbes are fed and how well they grow; if through undigested plant material, litter chemistry matters more.

Claims about this, checked

Things people have heard, and what the evidence actually supports.

The research behind this page

3 studies, newest first. Each one has a page explaining what it found and what it could not show.

2015Nature

The contentious nature of soil organic matter

Soil organic matter is better described as a continuum of progressively decomposing fragments than as discrete humic macromolecules.

2013Global Change Biology

The Microbial Efficiency-Matrix Stabilization (MEMS) framework integrates plant litter decomposition with soil organic matter stabilization: do labile plant inputs form stable soil organic matter?

The easily digested parts of litter — sugars and simple compounds — are converted efficiently into microbial tissue, and it is that microbial material which ends up bound to mineral surfaces and persisting.

2004Soil and Tillage Research

A history of research on the link between (micro)aggregates, soil biota, and soil organic matter dynamics

Aggregation operates hierarchically: mineral particles and organic matter bind into microaggregates, which are held into macroaggregates by transient binding agents including roots and fungal hyphae.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 39% completeness against what we would call a finished subject, and was last reviewed on 2026-09-04. It carries 3 claims and answers 3 mapped search questions.

  • 2 high-priority search intent(s) not yet covered
  • no research from the last few years is attached — check for newer work
  • more experiments could be explained in plain English
  • Global soil carbon stock figures are deliberately not quoted; they vary with method and depth convention.
  • Charcoal and pyrogenic carbon are not covered.
  • Agricultural carbon management is out of scope.