Lindsay A Turnbull, Cloe Paul-Victor, Bernhard Schmid, and Drew W Purves
Relative growth rate (RGR) is currently the most commonly used method for measuring and comparing species’ intrinsic growth potential. Comparative studies have, for example, revealed that small-seeded species have higher RGR, leading to the common belief that small-seeded species possess physiological adaptations for rapid growth that would allow them to outgrow large-seeded species, given sufficient time. We show that, because RGR declines as individual plants grow, it is heavily biased by initial size, and does not measure the size-corrected growth potential that determines the outcome of competition in the long term. We develop a daily growth model which includes a simple mechanistic representation of above- and belowground growth and its dependency on plant size and environmental factors. Intrinsic growth potential is encapsulated by the size-independent growth coefficient, G. We parameterized the model using repeated-harvest data from 1724 plants of nine species growing in contrasting nutrient and temperature regimes. Using information theoretic criteria, we found evidence for interspecific differences in only three of nine model parameters: G, aboveground allocation, and frost damage. With other parameters shared between species, the model accurately reproduced above- and belowground biomass trajectories for all nine species in each set of environmental conditions. In contrast to conventional wisdom, the relationship between G and seed size was positive, despite a strong negative correlation between seed size and average RGR, meaning that large, rather than small-seeded species, have higher size-corrected growth potential. Further, we found a significant positive correlation between G and frost damage which, according to simulations, causes rank reversals in final biomass under daily temperature changes of ±5 °C. We recommend the wider use of this new kind of plant growth analysis as a better way of understanding underlying differences in species’ physiology; but we recognise that RGR is still a useful metric if considering the potential rate of population increase in empty habitats.