Carbon cycle box model

Emit carbon dioxide for a while and follow where it goes: how much stays in the air, how much the land and the ocean take up, and how long the extra CO₂ lingers after emissions stop.

Scenario

1 GtC (gigatonne of carbon) = 3.67 Gt of CO₂.
Then emissions drop to zero.
0 means plants take up no extra carbon when CO₂ rises.
Larger means surface water resists taking up more CO₂.

Results

Atmospheric CO₂ (ppm)
Where the extra carbon is (change in each reservoir, GtC)

How it works

Four boxes of carbon (in GtC) start in balance, as before industrial times: atmosphere 590, land vegetation and soil 2,500, surface ocean 900 and deep ocean 37,100. Round numbers of the size reported in IPCC assessments are used for teaching. Carbon moves between boxes each year:

  • Atmosphere ⇄ land. Photosynthesis takes 120 GtC/yr × (CO₂ / start CO₂)β, with β your fertilisation slider. Respiration and decay return 120 GtC/yr × (land carbon / start land carbon).
  • Atmosphere ⇄ surface ocean. Gas exchange takes 90 GtC/yr × (CO₂ / start CO₂) into the ocean and returns 90 GtC/yr × (ocean carbon / start)R, with R the Revelle factor. Because R is about 10, a 1% rise in dissolved carbon raises the ocean's CO₂ pressure by about 10%, which is why the ocean takes up less than its size suggests.
  • Surface ⇄ deep ocean. 90 GtC/yr each way in balance, proportional to each box's carbon.
  • Emissions are added to the atmosphere. 1 ppm of CO₂ is about 2.12 GtC. The equations are integrated with a fourth-order Runge–Kutta method.

This is a teaching box model, not a climate projection. It leaves out ocean carbonate chemistry in detail, ocean circulation changes, warming feedbacks, rock weathering and sediment burial, so it recovers faster than the real cycle does; real extra CO₂ persists for thousands of years. The default settings were chosen so that the share staying in the air lands in the range often reported for the industrial era (roughly 40–45%), but the model is not fitted to observations.