Do Earth System Models Equate Photosynthesis with Wood Growth? The Claim Is an Overstatement of a Real Flaw.

The claim holds that many Earth system models equate photosynthesis rates — technically, gross primary production (GPP) — with wood growth. The verdict is partially false. There is a genuine, well-documented flaw in how ESMs handle carbon allocation, but the claim dramatically overstates what that flaw actually is.

The strongest evidence against the literal claim comes from Fatichi et al. (2019) in Nature Plants, which reviewed 15 land surface models and found that not one equated GPP directly with wood growth. Every model in the review subtracted autotrophic respiration from GPP to produce net primary production (NPP), then divided NPP among leaves, roots, and wood — with wood receiving roughly 20–40% of NPP. That means wood growth, in these models, draws on somewhere around 10–20% of total photosynthesis at most. Calling that an equation between photosynthesis and wood growth is not a simplification; it is a different claim entirely.

The steelman version of the argument points to two real problems. First, Collalti and Prentice (2019) in New Phytologist documented that many ESMs assume autotrophic respiration is a fixed ~50% of GPP, producing a near-constant NPP:GPP ratio. Second, De Kauwe et al. (2014) in Global Change Biology showed that models including CLM4, JULES, and ORCHIDEE use fixed allocation fractions rather than dynamic ones, meaning wood's share of NPP does not respond flexibly to drought, nutrient stress, or atmospheric CO2 the way real forests do. Trugman et al. (2018) in Geophysical Research Letters confirmed this pattern and explicitly flagged it as a misrepresentation of observed forest behavior. These are legitimate scientific criticisms.

But here is precisely where the claim breaks down: fixed allocation fractions and a simplified NPP:GPP ratio are not the same as equating photosynthesis with wood growth. Friedlingstein et al. (2006), reviewing CMIP3-era models in the Journal of Climate, found wide variation in how models partitioned GPP but confirmed that all included autotrophic respiration terms. Pugh et al. (2019) in Nature Geoscience, while noting that ESMs overestimate forest carbon uptake partly due to inadequate representation of carbon costs, explicitly distinguished GPP, NPP, and net biome production — the conceptual chain that the claim says models collapse. The models are oversimplified; they are not broken in the way the claim describes.

What is genuinely conceded: ESMs do a poor job of capturing the dynamic, flexible carbon allocation that real forests use to respond to environmental stress. The fixed-fraction approach means models can misattribute how much carbon ends up in long-lived woody biomass versus short-lived tissues or respiration, which matters enormously for projecting forest carbon sinks. Pugh et al. (2019) specifically linked this to systematic overestimates of forest carbon uptake. That is a serious modeling limitation worth scrutinizing.

The manipulation pattern here is compression: a nuanced technical critique — fixed allocation fractions misrepresent flexible carbon partitioning — gets collapsed into a more dramatic and quotable assertion that models treat all photosynthesis as wood growth. Watch for this move whenever modeling criticism travels from specialist literature into broader commentary. The original critique is usually accurate; the simplified version often swaps a real flaw for a fictional one, making it harder to fix the actual problem.