Analysis of six-decadal seed mass and emergence records in mast species shows little inter-annual variability

Abstract

Patterns of crop production in mast species do not track crop-year climate, but instead are regulated by climate cues in prior-years. Whether the pattern of year-to-year seed mass variation is coupled in time with mast seeding, maintaining seed mass-number trade-offs, and coherently driven by similar climate cues as other seed traits (e.g. seed germination) remains unknown. Using ca. 6,000 long-term seed inventory data over the years 1955-2015 in conifers, this retrospective study revealed the temporal patterns of mast species seed mass and its associated trait, seed germination. To pinpoint their ecological drivers, pairwise correlation analysis was performed between each trait and seasonal climates in crop year and four prior-years. Using climate variables key to each trait, regression models were constructed to project trait values. Findings showed minor seed mass variation among years, which rejects the generality of seed mass-number trade-offs in many plant species. This result reasonably arises as the economies of scale (compensating benefits) theory are often used to account for mast seeding but not for seed mass. Moreover, final germination fraction also varied little over time, but exhibited an increasing tendency. In addition, we found that temperature-based climate variables drive seed mass, number, and germination variation, but these variables in different seasons of crop year or prior-years did not have equal influences on trait variability. Finally, regression models showed that the number of frost-free days and evapotranspiration are crucial to the three traits and climate in autumn is a critical season, followed by summer and winter. This study holds considerable promise for explaining reproductive strategies of taxonomic groups with mast seeding characteristics in allocating reproductive resources to different life-history traits using ecological signals.