Permanent, biodiverse pastures in Montado ecosystems - biogeochemical and physiological implications for cork oak trees

Sown biodiverse permanent pastures rich in legumes (SBPPRL) have been implemented in Portugal as a management tool to increase soil fertility, grassland productivity and animal carrying capacity and were later selected as a voluntary land-use change activity towards increased carbon sequestration within the context of the Kyoto protocol. SBPPRL are commonly found in the understory of Mediterranean-type agro-silvo-pastoral systems - Montados - with cork oak as a dominant tree species. However, little is known about the effects of these introduced pastures on co-occurring cork oak physiology and productivity. Understanding the impact of grassland conversion on carbon, water, and nutrient cycling - namely at the tree level - could be of great importance for future management and policy decisions. Cork oak trees growing in an LTER, flux-tower site in Southern Portugal have been selected among two types of understory land-use: natural grassland and sown biodiverse permanent pasture. A suite of leaf-based physiological and morphological parameters were measured in cork oak trees across both land-use scenarios and different seasons. Here we focus on the results from foliar 15δN and 13δC between spring and summer. 13δC ranged from-30.21 to -27.36, with an average value of -28.74 (± 0.12) and no significant differences found between pasture types (natural vs. improved) or time (spring vs. summer). Foliar 15δN on the other hand showed statistically significant differences between cork oaks in different pasture types (-2.96±0.09 natural vs. -2.21±0.17 improved pastures, t-test, p ≤ 0.05), but no differences across time points. Cork oak trees in the permanent pasture have a 15δN signature closer to zero, consistent with a higher percentage of legumes (and N2 fixation) in that system. Using a mixed-model approach we estimated these trees to be using ca. 25% of their nitrogen from legume-fixation in the pasture. Despite the clear signature influence of legume-fixed N in oak trees growing on sown, permanent pasture, this difference was not translated into foliar N content (overall average of 1.43 ±0.02) suggesting that trees may not be taking full advantage of higher levels of N present in the soil. Further data from leaf-level gas exchange and other parameters currently being measured might elucidate these apparently contradicting observations.