Walk through a temperate forest in autumn and you are standing on a ledger. Each year, a hectare of deciduous woodland can drop several tonnes of leaves, twigs, and fallen wood. Within a few years, most of it is gone, broken down molecule by molecule and returned to circulation.

Ecology tends to look upward, toward primary producers and the consumers that eat them. Yet in most terrestrial ecosystems, herbivores consume only a small share of plant production. The great majority, often more than 90 percent, passes uneaten into the detrital pool. What happens there governs the supply of nitrogen, phosphorus, and other nutrients that plants need for the next round of growth.

From a systems perspective, decomposition is the return arm of the nutrient loop. If it runs too slowly, nutrients stay locked in dead matter and productivity stalls. If it speeds up, stored carbon escapes to the atmosphere. Understanding what sets its pace means looking at three things: who does the work, what conditions constrain them, and how this hidden process scales up to the global carbon budget.

Decomposition is less a single process than a relay. Soil invertebrates such as millipedes, isopods, earthworms, springtails, and mites shred and ingest litter, breaking it into smaller pieces. Bacteria and fungi then do the chemical work. They secrete extracellular enzymes that cleave complex polymers into molecules small enough to absorb. Each guild changes the substrate for the next, so the community operates as a processing chain rather than a collection of independent actors.

Within the microbial tier, fungi and bacteria occupy different niches. Bacteria thrive on labile, nitrogen-rich material and turn over quickly. Fungi dominate tougher substrates. Their hyphae penetrate plant tissue, bridge dry gaps between resources, and can move nitrogen from the surrounding soil into nutrient-poor litter. White-rot fungi are among the few organisms able to break down lignin efficiently, which makes them gatekeepers of woody carbon.

Invertebrates matter more than their biomass suggests. Fragmentation increases the surface area available for microbial attack. Passage through a gut also inoculates material with microbes and mixes it into mineral soil. Litterbag experiments use fine mesh to exclude soil fauna, and they consistently show slower decay when fauna are kept out. The effect is strongest in warm, humid ecosystems where invertebrate activity is high.

Interactions among decomposers add another layer. Fungi colonising dead wood compete intensely, and experiments show that the order in which species arrive, a priority effect, can substantially change how fast the wood decays. Litter also often decomposes faster beneath the plant species that produced it. This pattern, called home-field advantage, suggests local communities become specialised on familiar inputs. The rate we measure is a product of these interactions, not of the litter alone.

Takeaway

A decomposition rate is an emergent property of a community, not a fixed property of the material. The same leaf can decay at different speeds depending on who arrives first and who they compete with.

Ecologists often describe litter mass loss with a simple exponential decay model, in which a constant k sets how quickly material disappears. Values of k vary enormously. Leaf litter on a lowland tropical forest floor can largely vanish within a year, while needles in a boreal forest may persist for a decade or more. Three factors account for most of that variation: litter quality, temperature, and moisture.

Litter quality describes how easily decomposers can use the material. Nitrogen-rich, low-lignin leaves decompose quickly, while tough, lignin-rich tissue resists attack. Carbon-to-nitrogen and lignin-to-nitrogen ratios are reliable predictors of decay rate. Quality also shapes the timing of nutrient release. Microbes decomposing nitrogen-poor litter must pull nitrogen from the soil to meet their own needs, which temporarily immobilises it. Net release begins only once enough carbon has been respired to lower the ratio.

Temperature speeds up enzyme activity, and decomposition rates commonly roughly double for every 10°C of warming. Moisture behaves differently, because its effect is hump-shaped. In dry conditions, microbes become inactive and solutes cannot diffuse. In waterlogged conditions, oxygen is excluded, which forces slower anaerobic pathways. This is why peatlands accumulate organic matter, even though they are often cool and nutrient-poor.

The critical insight is that these controls interact rather than add. Warming a drought-limited soil changes little, because moisture is the binding constraint. Warming a cold, moist tundra soil can change a great deal. Pulses matter too. When dry soils are rewetted, a burst of respiration known as the Birch effect releases carbon and nutrients within hours. Predicting decomposition therefore requires identifying which constraint is limiting at a given place and time.

Takeaway

Decomposition behaves like a pipe with several valves in series: opening one fully does little while another stays nearly closed. Predicting how an ecosystem responds to change starts with finding the valve that binds.

Soils hold roughly 1,500 billion tonnes of organic carbon in their upper metre, more than the atmosphere and all living vegetation combined. That stock is constantly being exchanged. Each year, soil respiration returns on the order of 90 billion tonnes of carbon to the atmosphere, close to ten times the carbon released by fossil fuel combustion. The soil carbon pool is the small difference between this outflow and the inputs from plants.

Our understanding of what keeps carbon in soil has shifted. The older view held that persistent organic matter consisted of chemically resistant compounds. Current evidence suggests persistence depends more on context. Carbon bound to mineral surfaces or locked inside soil aggregates is physically protected from microbial enzymes. Microbes are also important builders of this pool, because their dead cells and by-products stick to minerals and can persist for centuries.

This sets up a potential climate feedback. If warming speeds decomposition more than it boosts plant inputs, soils release carbon, which drives further warming. The risk is concentrated in cold, wet ecosystems where decomposition has long been suppressed. Permafrost-region soils, measured to several metres depth, hold a comparable quantity of carbon. Much of it accumulated over millennia because freezing halted microbial activity, and thawing removes that constraint.

Yet the feedback is not straightforward. Microbial communities can acclimate to warmer conditions, readily available substrates can run out, and drying may offset the effects of warming. Fresh plant inputs can also stimulate the decomposition of older soil carbon, a phenomenon called priming, which complicates predictions under rising CO₂. These competing mechanisms help explain why Earth system models still disagree widely on how much soil carbon will respond to warming.

Takeaway

Soil carbon is the narrow balance between two enormous fluxes, so even a modest, persistent shift in decomposition can move more carbon than many deliberate interventions.

Viewed as a system, decomposition closes the nutrient loop and regulates how much carbon stays on land. Its rate emerges from three interacting layers: the organisms present, the material they receive, and the physical conditions they work under.

For managers, this framing points to practical levers. Species selection alters litter quality and the timing of nutrient release. Retaining coarse woody debris sustains fungal communities. Protecting soil fauna, avoiding compaction, and rewetting drained peatlands all shift decomposition toward desired outcomes. Standardised litterbags and tea bag indices make the process measurable.

Most of an ecosystem's production ends up on the ground, handled by organisms we rarely see. How fast they work determines what the system has to build with next.