You cannot run an experiment on the Amazon. So ecologists do two things instead: they shrink an ecosystem down until it fits in a sealed jar, and they measure the real thing from satellites and calculate what is changing. This page covers both — how to build a mesocosm properly, and how to handle the percentage-change maths that comes with deforestation data.
📘 What you need to know
A mesocosm is an experimental container in which a naturally occurring ecosystem is simulated.
Mesocosms let you control every factor except the one being tested, which a real ecosystem never does.
A mesocosm needs a transparent container, autotrophs, and at most a few primary consumers. Never secondary consumers.
Sealed systems are more controlled than open ones, so they are more useful experimentally.
Every mesocosm experiment needs a control mesocosm that is identical except for the altered variable.
IB guidelines mean animals should be left out where possible, and treated carefully and returned if used.
Deforestation reduces transpiration, which changes local rainfall and temperature and could push the Amazon past a tipping point.
You must be able to calculate percentage change in forest cover.
Mesocosms: an ecosystem you can actually experiment on
The stability of an ecosystem can be investigated using a model ecosystem called a mesocosm. Because it is enclosed, you can hold everything constant and alter one known factor — light level, temperature, nutrient concentration — and then watch what happens.
Mesocosms come in every size. Water tanks on land can be used to study the effect of sewage pollution on ponds or lakes. Underwater enclosures in coastal waters or lakes can test how temperature change or dissolved carbon dioxide affects ocean ecosystems. Trees have even been planted in huge greenhouse-like buildings to replicate a rainforest and follow carbon through it.
The obvious weakness
Mesocosm experiments can be criticised as unrealistic, precisely because of what makes them useful: they are enclosed, and the level of control is far higher than anything in nature. Realism can be improved by building larger mesocosms that share more features with the real thing — for example, one big enough to allow layers of water to mix as they would in the open ocean.
This is the classic validity-versus-control trade-off, and it is worth being able to argue both sides. More control means cleaner results but weaker relevance to the real world. Examiners like answers that name the trade-off rather than just criticising the method.
Building one in the lab
Small mesocosms are easy to build. Getting them to survive is the hard part, and the design choices all follow from the same logic: energy must get in, and the food chain must not be longer than the container can support.
The container must be transparent so sunlight can reach the producers inside.
Autotrophs must be included so light energy can be converted into chemical energy inside the mesocosm.
Small primary consumers such as zooplankton or other small invertebrates can be included — but only if the mesocosm is large enough to support them.
Do not include secondary consumers. There will not be enough energy in the food chain to sustain them for long, and it would be unethical to let the primary consumers be eaten in that way.
Sealed systems prevent organisms and substances entering or leaving, so they are more controlled than open ones and more useful for experiments.
A control mesocosm must be set up at the same time, identical in every way except that the variable under test is not altered. Without it you cannot show that any change was caused by your factor rather than something else.
The layers in the terrestrial version are not decoration. Gravel drains, charcoal keeps mould down, the filter layer stops soil washing into the gravel, and the compost carries the micro-organisms that do the nutrient cycling.
🧩 Terrestrial mesocosm, in order
Drainage layer — gravel in the bottom of a clear container.
Charcoal on top of the gravel, which helps prevent mould growth.
Separation layer — sphagnum moss or filter paper, keeping the base layers apart from the organic matter above.
Soil or compost, providing organic material and the micro-organisms that drive nutrient cycling.
Slow-growing producers such as healthy mosses and ferns, planted into the growth medium.
Water, then seal. It may need watering while it establishes, but avoid over-watering — once stabilised, the plants release enough water vapour to keep the moisture level up.
Place it somewhere light, at a stable temperature.
🧩 Aquatic mesocosm, in order
Organic substrate from the bottom of a lake or pond, which supplies naturally occurring nutrients and microorganisms.
Lake or pond water, which contains the microscopic organisms you need and avoids the chemicals found in tap water.
Healthy aquatic plants, to produce carbohydrates and oxygenate the water.
A few small animals such as water fleas or water snails — primary consumers only, and never more than the mesocosm can support.
Place it somewhere light, at a stable temperature.
The ethics: IB guidelines on animals
The IB policy on animals in schools says investigations should only involve animals where there is no alternative, must not be cruel, and must include measures that remove potential causes of distress.
For a mesocosm, the most ethical approach is usually to leave animals out entirely — not every mesocosm needs them to be sustainable. If animals are needed, use only a limited number of herbivores in carefully controlled conditions: enough food available, not too hot or too cold, the investigation not run for too long, and the animals returned to their natural environment at the end.
An ethics answer that scores. Do not just write “it is unkind”. Name the specific measure: excluding animals where possible, limiting numbers, keeping conditions within tolerance, limiting duration, and returning them afterwards.
Deforestation of the Amazon
The Amazon is the standard example of human activity threatening ecosystem sustainability, and the mechanism is worth following carefully because it is a chain, not a single step.
Transpiration releases water vapour into the air above the rainforest. That has a cooling effect, and it affects air movement and rainfall. So changing the number of trees carrying out transpiration changes local temperature and rainfall. Temperature and rainfall in turn control the rates of photosynthesis and nutrient cycling — two of the requirements for stability from the previous page. Deforestation therefore has a knock-on effect far beyond the trees actually cut.
Scientists are concerned that the Amazon could reach a tipping point: so many trees removed that temperature and rainfall patterns change significantly, pushing climatic factors beyond the tolerance levels of some species. There is real uncertainty about how much forest would have to be lost for this to happen, so nobody knows how close the Amazon actually is.
Two different questions can be asked of this graph. “How much has been lost?” is a subtraction. “Is the situation improving?” is about the gradient, which is getting shallower.
The percentage change calculation
Learn this arrangement
percentage loss of forest = (change in area ÷ original area) × 100
The change is the initial forest cover minus the final forest cover. The denominator is always the original value, never the final one — that single detail is where most marks are lost.
Worked examples
WORKED EXAMPLE 1
A region had 100 million hectares of forest in 1990 and 72 million hectares in 2020. Calculate the percentage loss of forest over this period. [2]
Step 1: find the change100 − 72 = 28 million hectares lostStep 2: divide by the ORIGINAL area28 ÷ 100 = 0.28Step 3: multiply by 1000.28 × 10028 % lossdividing by 72 instead of 100 gives 38.9 % — the single most common error in this calculation
WORKED EXAMPLE 2
Explain how deforestation could reduce the rate of photosynthesis in the trees that remain. [4]
Step 1: the direct effect
Fewer trees means less transpiration, so less water vapour is released into the air above the forest.
Step 2: the climate effect
Less water vapour reduces cloud formation and rainfall, and removes the cooling effect, so the area becomes hotter and drier.
Step 3: the effect on the remaining trees
With less water available, stomata close to reduce water loss, which limits carbon dioxide uptake.
Step 4: the conclusionThe rate of photosynthesis in the surviving trees falls, so productivity drops furtherthis is a feedback loop — the loss of trees makes conditions worse for the trees that are left
WORKED EXAMPLE 3
A student sets up a sealed aquatic mesocosm and adds a small fish to feed on the water fleas. Suggest two reasons why this is a poor design choice. [2]
Reason 1: energy
A fish is a secondary consumer. There will not be enough energy transferred along the food chain to sustain it, because energy is lost at each trophic level.
Reason 2: ethics
Allowing the primary consumers to be eaten in a sealed container would breach IB guidelines on avoiding animal distress.
Include producers and, at most, a few primary consumersthe safest design of all contains no animals — a mesocosm does not need them to be sustainable
💡 Exam tip
Define a mesocosm precisely: an experimental container in which a natural ecosystem is simulated.
Always mention the control mesocosm when asked about experimental design.
Justify design choices with a reason: transparent so light reaches producers, sealed so nothing enters or leaves.
For percentage change, always divide by the original value, and show the substitution.
When evaluating a mesocosm, give the trade-off — high control but low realism — and the fix, which is a bigger mesocosm.
For ethics questions, cite specific measures rather than general kindness.
⚠ Common mix-up
Dividing by the final value in a percentage change. It is always the original.
Adding a predator to a mesocosm to make it “more realistic”. Energy losses make it unsustainable.
Forgetting the control. Without it, nothing can be attributed to the variable you changed.
Saying deforestation only removes trees. The bigger issue is what it does to rainfall and temperature.
Stating the tipping point as a known figure. There is genuine uncertainty about where it lies — say so.
Confusing an open with a sealed mesocosm. Sealed gives more control; open is more realistic but less useful experimentally.
Up next: Keystone Species — why removing one species from an ecosystem can bring far more than that one species down with it.
Want this explained one-to-one?
Book a free session with an experienced IB Biology tutor and get your trickiest topics made simple.
