Chapter 22 Litter decomposition: measuring an ecosystem process

Skills this lab builds. 4DEE: Core Ecological Concepts → Ecosystems. Cross-Cutting Themes → Pathways & Transformations of Matter and Energy. Ecology Practices → Fieldwork; Quantitative reasoning and computational thinking. BioSkills: Process of Science → “Design controlled experiments, including plans for analyzing the data”; Quantitative Reasoning → “Interpret the biological meaning of quantitative results.”

This lab is deployed early and retrieved late. You put litterbags out around week 3 and collect them around week 12. Between those two days it needs nothing from you. Read this chapter when you deploy, and again when you retrieve.

Everything a forest builds, it eventually takes apart. A ponderosa drops its needles, and somewhere between the forest floor and the atmosphere those needles stop being needles and become carbon dioxide, fungal tissue, and soil. Decomposition is the return half of the carbon cycle, and it is the process that determines whether an ecosystem stores carbon or releases it.

The species diversity lab opens with a figure from Swift et al. (1979) showing the three things that control how fast that happens: the biota doing the decomposing, the physico-chemical environment they work in, and the resource attributes of the litter itself. That lab measures the biota. This lab measures the rate — and lets you manipulate whichever arm of that diagram you choose.

22.0.1 Downloads for this lab

22.1 Objectives

  1. Measure mass loss from litter in the field over a known interval.
  2. Convert mass loss into a decomposition rate constant, \(k\).
  3. Test whether \(k\) differs between two levels of one factor you choose.
  4. Use a standard substrate as a positive control, and explain why that is necessary here.

22.2 The model

Litter does not lose mass at a steady rate. It loses a constant fraction of what remains, which means the curve is a negative exponential — the same shape as radioactive decay, and the same shape as the exponential growth from the population ecology lab with the sign flipped:

\[\frac{M_t}{M_0} = e^{-kt}\]

where \(M_0\) is the starting mass, \(M_t\) the mass remaining after time \(t\), and \(k\) the decomposition rate constant. Rearranged for the thing you want:

\[k = \frac{-\ln(M_t / M_0)}{t}\]

Report \(k\) in per day if your interval is a semester, or multiply by 365 for the per-year figure the literature uses. A \(k\) of 0.2 yr⁻¹ means about 18% of the litter disappears in a year; \(k\) = 1.0 yr⁻¹ means about 63% does.

Why a rate constant and not just “percent lost”? Because percent lost depends on how long you left the bags out, and \(k\) does not. Your bags and a published study’s bags were out for different numbers of days, and \(k\) is what makes the two comparable. This is the same reason the population ecology lab insists on a per-day lambda.

22.3 Part 1 — Choose your factor

Pick one arm of the Swift diagram. Your TA may assign it.

Factor What you vary What you hold constant The question
LitterType Ponderosa needles vs Gambel oak leaves One location Does resource quality control decomposition rate?
Microhabitat Under closed canopy vs open ground; or Sinclair Wash vs upland One litter type Does the physical environment control it?
MeshSize Fine mesh (~0.5 mm, excludes invertebrates) vs coarse (~4 mm, lets them in) One litter, one location How much of decomposition is done by the soil fauna?

The mesh-size option is the most direct link to the diversity lab: the animals you extract in a Berlese funnel are the ones the fine mesh is keeping out.

22.4 Part 2 — Prepare the bags

You need six bags per level — twelve bags for a two-level comparison. Six feels like a lot for one afternoon. It is the minimum that will detect anything, because between-bag variation in the field is large and the signal here is small.

22.4.1 Step 1: Collect and air-dry the litter

Collect freshly fallen litter — on the ground, not off the tree, and not already crumbling. Spread it somewhere dry indoors for at least a week.

22.4.2 Step 2: Work out the air-dry to oven-dry correction

This is the step people skip, and it is the one that will otherwise wreck your numbers.

Air-dry litter still holds water. Oven-dry litter does not. If you weigh your litter air-dry going in and oven-dry coming out, some of your “mass loss” is just the water you never drove off in the first place — and it can easily be larger than the real signal.

You cannot oven-dry the litter you deploy, because heating it changes how it decomposes. So instead:

  1. Take five extra subsamples of the same litter, about 5 g each. Weigh each air-dry, to 0.01 g.
  2. Oven-dry them at 60 °C for 48 hours, then reweigh.
  3. Calculate the mean ratio: \(\text{correction} = \overline{(\text{oven-dry} / \text{air-dry})}\).
  4. Your true starting mass for every bag is its air-dry mass × that correction.

Record the correction factor. It goes in your methods.

22.4.3 Step 3: Fill and label

  1. Weigh about 5.00 g of air-dry litter into each bag. It does not need to be exactly 5.00 — it needs to be recorded exactly, to 0.01 g.
  2. Seal the bag. Staples, or stitch it.
  3. Label with a plastic tag and pencil or an aluminium tag — not a marker on masking tape, which will be an unreadable grey smear in November.
  4. Record the BagID, the Level, and the InitialMass_g in your datasheet before the bag leaves the room.

22.4.4 Step 4: Add the standard substrate

Alongside your litter bags, deploy two commercial tea bags at each location: one green tea, one rooibos. Same depth, same day.

This is not a gimmick. It is a positive control, and here is why you need one.

Ponderosa needles are tough, waxy and low in nitrogen, and in a dry montane fall they may lose only 3–5% of their mass in ten weeks — which is close enough to your weighing error that a null result would be uninterpretable. You would not know whether the needles resisted decomposition or your site was simply too dry and cold for anything to happen at all. The tea tells you which. Green tea loses roughly half its mass in the same period almost anywhere that decomposition happens. If your green tea decomposed and your needles did not, that is a real result about needles. If nothing decomposed, that is a result about your site.

This use of standard tea as a comparable substrate is a published method — the Tea Bag Index (Keuskamp et al. 2013) — and if you follow the standard protocol your numbers join a global dataset.

Check the bag material before you buy. The Tea Bag Index depends on the mesh itself not decomposing. Some manufacturers have switched from nylon to plant-based mesh that breaks down in soil, which destroys the measurement. Your instructor will confirm what is currently available; if the mesh is not synthetic, use fine-mesh bags of your own filled with the tea leaves.

22.5 Part 3 — Deploy

  1. Choose your locations using the randomization approach from the sampling lab. Do not place bags where the litter looks interesting.
  2. Clear the surface litter aside, lay the bag flat on the soil surface, and replace the litter on top. Tea bags go 8 cm deep in the soil — that is the Tea Bag Index standard and changing it makes your numbers incomparable.
  3. Pin each bag with a landscape staple or a wire flag so it does not blow away.
  4. Mark the spot so you can find it in ten weeks under snow. Flag it, and record a GPS point or a written description with two distances to fixed objects.
  5. Record the date. This is \(t\), and getting it wrong makes every \(k\) wrong.

Expect to lose bags. Animals dig them up, snow buries the flags, grounds crews tidy. Deploying twelve and recovering ten is a normal outcome — one more reason six per level and not three.

22.6 Part 4 — Retrieve and weigh

  1. Record the date and calculate Days.
  2. Lift each bag carefully. Roots will have grown into it and soil will have washed in.
  3. Clean the outside — brush off soil, clip away roots growing through the mesh. Do this gently and consistently; how hard you brush is now part of your method.
  4. Oven-dry at 60 °C for 48 hours.
  5. Weigh to 0.01 g, and record as FinalMass_g.

The bias you cannot brush away. Fine soil and mineral grit work through the mesh and stay there. That inflates your final mass, which makes your litter look less decomposed than it was. Every litterbag study has this problem. The proper fix is ash-free dry mass — combust a subsample in a muffle furnace at 500 °C, weigh the ash, and subtract the mineral fraction. If your instructor has a muffle furnace available, do it. If not, say so in your methods and state which direction the bias pushes your result. Naming the direction of a bias you could not remove is what a careful methods section looks like.

22.7 Analysis

Run LitterDecomposition.R. It calculates percent mass remaining and \(k\) for every bag, compares your levels, and plots them.

Two things to look at before you conclude anything:

  • The spread within a level. Six bags of identical litter in the same place will not give six identical numbers. If the within-level spread is as big as the between-level difference, you have not shown anything, regardless of what the mean says.
  • The tea. If green tea lost less than about 20% of its mass, decomposition barely happened at your site over your interval, and your litter comparison has no power to detect anything. That is a legitimate finding and it is the one to report.

22.8 Assignment

Submit one document containing:

  1. Your completed datasheet, including the air-dry to oven-dry correction factor and how you calculated it.
  2. A table of mean percent mass remaining and mean \(k\) (per day and per year) for each level, with the standard deviation.
  3. A figure showing \(k\) for each level, with the individual bags visible as points. Legend included, no title on the figure.
  4. Your result: did \(k\) differ between your two levels, and by how much?
  5. The control: what did the tea do, and what does that let you say about the interpretation in item 4?
  6. One paragraph on the assumption most likely to be wrong in your study — soil contamination, bags lost non-randomly, the correction factor, mesh excluding fauna you did not mean to exclude — stating which direction it would push your result and whether your conclusion survives it.

22.9 Sources

Olson, J.S. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44(2): 322–331. — the negative exponential decay model and \(k\).

Keuskamp, J.A., B.J.J. Dingemans, T. Lehtinen, J.M. Sarneel, and M.M. Hefting. 2013. Tea Bag Index: a novel approach to collect uniform decomposition data across ecosystems. Methods in Ecology and Evolution 4(11): 1070–1075. https://besjournals.onlinelibrary.wiley.com/doi/10.1111/2041-210X.12097. Protocol and current guidance at https://www.teatime4science.org.

Swift, M.J., O.W. Heal, and J.M. Anderson. 1979. Decomposition in Terrestrial Ecosystems. Blackwell, Oxford. — the three-factor framework, and the source of the figure in the species diversity chapter.