Friday, August 6, 2010 - 8:40 AM

COS 107-3: Self-thinning of annual grass seedlings accounts for spatial and temporal variation in the availability of multiple nutrients

Valerie T. Eviner, Naomi M. Clark, and Lillian Hayden. University of California Davis

Background/Question/Methods

In California grasslands, researchers have limited ability to explain site-to-site and year-to-year variability in forage production. Similarly, nitrogen budgets in these grasslands have indicated that nitrogen supplies through litter decomposition, soil organic matter mineralization, atmospheric deposition, and nitrogen fixation are not adequate to account for yearly plant uptake, suggesting that nitrogen is made available from another source. Previous work demonstrated that naturally high densities of seeds result in 50-90% of germinated seedlings dying during the growing season, and that the annual flux of nitrogen released from dying seedlings is 60-170% of that released from the decomposition of senesced litter.

This study’s objective was to determine the extent to which N release from senesced litter vs. dying seedlings explains variability in soil nutrient availability seasonally, and across sites that differ in environmental and management conditions. We also explored the extent to which small mammals (the system’s primary granivores) influence nutrient availability. We monitored seasonal nutrient availability, litter decomposition, and seedling thinning dynamics at 12 sites that differ in soil fertility, temperature, plant species composition, and history of grazing. At each of these sites we monitored a pair of plots- a small mammal exclosure as well as a control plot.

Results/Conclusions

We found that both seedling thinning and litter decomposition are important drivers of soil nutrient availability, with different nutrients varying in how strongly they are affected by either plant input. Seasonal patterns of nitrogen availability were tightly correlated with seedling thinning at all sites. In contrast, seasonal patterns in sulfur availability strongly correlated with litter mass loss. Potassium availability was impacted by both inputs, while neither litter mass loss, nor seedling thinning correlated with seasonal patterns of phosphorus and calcium availability.

Variability in nutrient availability across sites showed weaker correlations with litter and seedling thinning than did seasonal variability. Early in the growing season, site-to-site variability of nitrogen and phosphorus availability correlated with seedling thinning, while litter mass loss explained spatial variability in sulfur and potassium. Spatial variability in early season calcium availability related to both inputs. Later in the growing season, litter mass loss accounted for site variability of all nutrients except for potassium, which correlated with seedling thinning.

The presence of small mammals significantly decreased early season availability of calcium, sulfur, phosphorus, and potassium, but not nitrogen. This is likely because seed predation by small mammals decreased the early season flush of seedling thinning.