Recruitment is a limiting life history stage in many forests, and recruitment limitation has the potential to increase in the context of climate change due to the particular vulnerability of early life stages to changing environmental conditions. Recruitment limitation may result from constraints on any stage of recruitment. The demographic transitions extending from seed production to germination are poorly understood components of recruitment owing to the difficulty of observing these stages directly, but they may be particularly sensitive to spatiotemporal variation in environmental conditions. The objective of this study was to investigate climate sensitivity among unobserved, early recruitment stages (e.g. pollination, fertilization, seed maturation, and dispersal) by quantifying lags in the effects of climate on seedling germination in naturally regenerated conifer forests in the Pacific Northwest. To address this objective, we monitored seedling germination over three years in 600 quadrats stratified across five long-term monitoring plots that span a broad climatic and elevational gradient in Oregon. Hierarchical Bayesian Poisson regression models were used to quantify the effects of climate during the germination year, one year prior to germination, and two years prior to germination, and models representing different lag effects were compared to assess climate sensitivity in unobserved recruitment stages.
Results/Conclusions
For both focal species, the best-fitting models included climate in the year of germination. Air temperature and precipitation were more strongly related to recruitment than soil temperature and moisture for germination years and lag years. Air temperature and precipitation had a stronger influence on Tsuga heterophylla recruitment than climate in the summer, winter, or average annual climate, and annual climate was most strongly related to Pseutosuga menziesii recruitment. Across individual sets of seasonal covariate combinations and excluding the overall best-fitting models for each species, models utilizing climate in the year prior to germination or two years prior to germination showed equivalent fit, and performed better than models utilizing climate in the year of germination. When considered as a whole, these findings indicate that climate has a particularly pronounced effect on seedling germination, but that earlier recruitment stages related to seed production still reflect a substantial climate influence.