Comparing two ground-based seed count methods and their effect on masting metrics

Abstract

Masting, i.e. interannually variable and synchronized seed production, plays a crucial role in forest ecosystems, influencing wildlife dynamics, pathogen prevalence, and forest regeneration. Accurately capturing masting variability is important for effective forest management, conservation efforts, and predicting ecosystem responses to environmental changes. The adoption of low-cost methods facilitates the large-scale data acquisition needed in this time of unprecedented environmental upheaval, but it is important to understand the reliability of such methods. We investigated the relationship between the timed count method and the quadrat-based method for monitoring seed production in European beech (Fagus sylvatica). The timed count method is fast, cost-effective, and suitable for areas with public access. These characteristics make time counts a practical choice for large-scale seed monitoring. However, the method has not been cross-calibrated with more traditional ground-based methods like quadrat sampling, which involves exhaustive seed collection from designated plots under tree canopies. Our research reveals a loglinear relationship between seed counts obtained by the two methods, and shows that the timed count is an effective method of estimating seed production. We also found that seed production exhibits greater dispersion in patchiness at lower levels of seed fall, which explains why the timed count method, covering a larger area, captures greater variability in seed fall compared to the quadrat method in such contexts. This highlights the importance of choosing an appropriate sampling strategy to accurately assess seed fall. The differences between the two methods introduce variability into derived masting metrics, such as the coefficient of variation and synchrony, with individual-level seed production variability metrics being more affected than population-level ones. The findings underscore the importance of understanding how different sampling methods can impact long-term ecological studies, particularly those focused on masting behaviour.