The overall goal of this project is to identify the functional traits and environmental conditions that drive variation in fitness (λ) and vital rates (survival, growth, and reproduction) for native, perennial graminoids in the ponderosa pine-bunchgrass ecosystems in northern Arizona. Using our long-term (17 years) data set, vegetation maps from 1-m2 quadrats, we have quantified individual growth and survival for these perennial graminoids, yet reproduction data, a crucial element of any demographic and population model, remains elusive. This study aims to quantify fecundity kernels for each of 10 graminoid species. These kernels require information on probability of flowering, fecundity (seeds/adult), and establishment as a function of the continuous state variable, size. In perennial graminoids, fecundity is challenging to quantify because of the large quantity of seeds produced by each plant. Therefore, we quantified plant size, number of flowering stalks, inflorescence height, basal leaf height, and seeds per plant around permanent quadrats to predict seed production using correlated traits that are easier to measure than seed production itself. Additionally, we modeled the probability of an individual graminoid species flowering as a function of plant size using regression techniques, producing another element of the fecundity kernel.
Results/Conclusions
We focused on one cool season (C3) graminoid species, Poa fendleriana (mutton bluegrass), to demonstrate the proof of concept for the remaining nine species. Poa fendleriana seeds per spikelet follow a predictable distribution averaging 3.5 seeds per spikelet and 1 spikelet per glume. Number of seed stalks was the best predictor of total number of glumes across 30 individuals of Poa fendleriana (R² = 0.97). In contrast, individual basal area, flowering stalk height and basal leaf height were not good predictors of glume production. These results suggest that we can predict individual fecundity for Poa fendleriana using number of flowering stalks, which is a simple and efficient data collection technique. We found that the probability of an individual Poa fendleriana plant flowering increased linearly with plant size because even small Poa fendleriana plants were flowering at a high probability (intercept coefficient = 0.92). These fecundity measurements will fill the gap in our long-term data set, and future work will use these data to build fecundity kernels to be utilized in density-dependent Integral Projection Models for 90 populations of 10 graminoid species in northern Arizona pine forests.