Widely Used Variants of the Farquhar-von-Caemmerer-Berry Model Can Cause Errors in Parameter Estimation

A bstract The Farquhar-von-Caemmerer-Berry (FvCB) model is the most widely-used mechanistic model of C 3 net CO 2 assimilation, and it plays a significant role in plant physiology, ecology, climate science, and Earth system modeling. As use of the model has grown, multiple variants have appeared across publications. Although many of these are commonly used, there has not been a detailed investigation of existing variants and their impacts on results and interpretations. Here we summarize the types of variants and their prevalence in the literature, and we present a comprehensive comparison of differences between them. A key finding is that a common variant that uses the minimum of assimilation rates rather than the minimum of carboxylation rates, which we call the “min- A variant,” makes different predictions than the original “min- W variant,” yet appears in approximately half of highly-cited publications and software tools that use the FvCB model. Another concern is that although leaf biochemistry restricts the range of CO 2 partial pressures where limitations due to triose phosphate utilization (TPU) can occur, this restriction is commonly omitted from the model’s equations. Among other potential issues, these variations can introduce errors exceeding 20% when estimating photosynthetic parameter values from CO 2 response curves. It is therefore important to be aware of this source of error when fitting the model, to avoid using the min- A variant, and to include the biochemically-derived CO 2 threshold for TPU limitations. C entral T heme of the M anuscript The Farquhar-von-Caemmerer-Berry model of CO 2 assimilation plays a key role in plant research, but many publications use variants of the model that differ from the original and can potentially introduce errors in photosynthetic parameter estimates. N ovel R esults , I deas, or M ethods Using a literature survey, FvCB model variants are categorized, and some are found to make contradictory predictions. Comparisons against A - C i curves show that the “min- W variant” exhibits the best performance, especially at low CO 2 concentrations.