Disentangling the effects of geographical location and local climate on masting in Fagus crenata using a comprehensive population model

Masting, the intermittent and highly temporally variable production of seed crops, is exhibited by many plant genera, including Fagus. Although patterns of masting have been investigated over a wide range of areas, very few studies have comprehensively examined the various related factors. First, we developed a new comprehensive population-level process model of seed production and tested its ability to reproduce interannual variation in Fagus crenata seed production at a test site for which detailed quantitative flower and seed production data were available. Second, to explore hypotheses for the mechanisms of spatial variation in masting in Fagus and disentangle the effects of geographical and climate variables, we fitted coarse qualitative data (seeding rank) for a 20-year period for 102 populations of Fagus crenata across 550 km in northern Japan to the process model above, encompassing variable average climate conditions and latitudinal genetic divergence. We found good agreement between observed and predicted numbers of flowers and seeds at the test site. A simulation demonstrated that the process model reproduced the observed masting pattern well. Fitting data for the 102 sites allowed us to develop new hypotheses. The indicator of the effect of higher summer temperatures as a weather cue for masting decreased with latitude, possibly reflecting latitude-associated genetic divergence. The indicator of the evasion of insect predation was positively correlated with winter precipitation, possibly due to poor survival of the primary seed predator of F. crenata in heavy-snowfall areas. These hypotheses related to spatial variation in masting patterns should be tested in future studies. The proposed process model will be useful for forecasting climate-change impacts on masting patterns in Fagus and other genera, improving the efficacy of forest ecosystem management.