Large underestimation of intraspecific trait variation

Background/Question/Methods: Intraspecific trait variation (ITV) widely exists in natural communities and has gained increasing attention due to its large ecological effects on community dynamics and ecosystem functioning. However, little attention has been paid to the estimation of ITV per se. It remains unclear whether our ITV estimation is accurate enough to make reasonable ecological inferences. Therefore, we focused on the most commonly used ITV estimator, ITV₁, in existing studies and tested its accuracy using three large real trait datasets from local to regional scales. To reach a more accurate ITV estimator, we proposed two new composite estimators to consider the skewness and kurtosis of trait value distribution and compared their performance with four existing ITV estimators across various sample sizes using both simulated and the three real trait datasets. We asked which is the best ITV estimator and what is the minimum sample size that could reach a ±5% and a ±10% accuracy of the ITV estimate for a completely random sample scheme.

Results/Conclusions: Our results consistently showed the ITV₁ is often underestimated, and this underestimation varies largely among traits/species and can reach 20%. This bias arises from an oft-ignored fact that ITV₁ is a biased estimator and the extent of bias depends on the sample size, skewness and kurtosis of the trait value distribution. Our new composite estimator ITV₅ largely reduced the underestimation of ITV and had the lowest total absolute bias across samples sizes from 10 to 400. It reached ±5% and ±10% accuracies once the sample sizes are greater than 70 and 45 in most of situations, respectively. These results suggest that the real extent of ITV in natural communities may be underestimated in past studies, and can be more accurately estimated by our new ITV estimator. Thus our estimator not only provides a solid foundation for robustly quantifying ITV, but also helps to more confidently inferencing ecological mechanisms linked with ITV.