Comparative demography of two declining oak species in a Mediterranean-climate savanna

Background/Question/Methods

Previously we reported long-term experimental results demonstrating strong control of cattle, deer and rodents on the demography of California valley oak (Quercus lobata), a deep-rooted deciduous tree.  Using matrix models, we showed that valley oak demography in our study area is more sensitive to rates of sapling recruitment and growth than to acorn production or seedling establishment. Here, using a decade of data from the same experiment, we compare the demography of valley oak to that of coast live oak (Quercus agrifolia), an evergreen, relatively shallow-rooted tree. The two species co-dominate and are declining at our study site in central coastal California. We ask: 1) how sensitive is coast live oak population growth rate to variation in acorn production, seedling establishment, and sapling survival and growth? 2) How important is browsing by cattle and deer to the species’ demography? 3) How do population growth rates and vital rate sensitivities compare for the two species? We construct matrix models for coast live oaks sown as acorns in 4 planting  years under no protection and three different levels of protection: no cattle; no cattle and no deer; no cattle, no deer, and no rodents).

Results/Conclusions

Similar to valley oak, coast live oak survivorship was more strongly related to plant height than plant age after 2 years. Six functional stages for matrix models include age < 1 yr, age 1-2 yr, established small seedling (> 2 yrs, height < 20 cm), small sapling (>2 yrs, height 20-140 cm), non-reproductive tree (ht > 140 cm, age < 35 yrs), and reproductive tree (ht > 140 cm, age > 35 yrs). Effects of ungulate browsing were more pronounced for coast live oak than valley oak. Year 1 and year 2 survivorship rates were low and weaky affected by excluding cattle, but roughly doubled when both deer and cattle were excluded. In the presence of cattle or cattle and deer, no plants transitioned out of the small sapling stage. Calculated population growth rate exceeded 1 when plants were protected from both cattle and deer, and was significantly higher under protection from rodents. In summary, coast live oak demography is similar to valley oak, strongly influenced by vertebrate consumers, with some differences in vital rate sensitivities.  We compare species’ demography in light of functional trait differences, and consider implications of our findings for regional oak restoration efforts.