In the first phase of the project (two-year outgoing at Arizona State University) the experienced researcher (ER) used rainfall manipulation experiments (RMEs) at two different scales (population and plant phenotypic level) to monitor B. eriopoda responses to different rainfall legacies in a gradient of precipitation. Besides, she performed a litter matter decomposition trial to determine to what extent rainfall characteristics and legacies affect soil microorganisms and the carbon and nutrient cycles. The RMEs were implemented at The Jornada Long Term Ecological Research site (JRN LTER, New Mexico). At population scale, two variables were studied: patch size and total B. eriopoda cover. Mean patch size and total cover under extreme drought conditions (80% reduction of naturally occurring rainfall) during 3 and 12 years were significantly lower than in natural precipitation conditions. After 12 years of extreme drought, the total cover was near 0 m2. However, in the short-term drought, only the greatest patch sizes suffered a regression. Under an extreme regime of precipitation (80% increase), these variables were only significantly higher in the long term. These results pointed out the importance of the duration of extreme precipitation and drought events in the B. eriopoda population structure.
At the plant phenotypic level, 18 leaf and shoot traits and 6 sexual reproductive traits were analysed. The plasticity indexes of the traits related to sexual reproduction were located at the extreme ranges of the gradient of values. A three-year extreme drought event stimulated the seed production when summer precipitation returned close to its historical average value. Unexpectedly, this did not occur when after the extreme drought period the contemporaneous rainfall changed to an extreme rainfall event. An extreme precipitation legacy of three years did not cause significant changes in sexual reproduction investment, suggesting that the potential of fitness increase that greater water availability could cause is mediated by the asexual reproduction. The total number of leaves and the increase of alive leaves had low and high values of plasticity, respectively. This indicates that the responses to rainfall extremes related to photosynthetic tissues in this species are mainly driven by leaf span rather than to leaf morphology and quantity. The plastic potential of these traits modulated their resistance responses to extreme rainfalls and drought events. The morphological traits were in general the most resistant to change. The results of the decomposition trial pointed out an adaptation of the soil community to a scenario of lower soil water availability.
In the second phase of the project, another RME was established at the Garraf experimental station (south of Barcelona, Spain) with B. hybridum as the target species. In this experiment, a gradient of drought treatments was imposed (25% and 75% of natural rainfall reduction) through the use of rainfall exclusion shelters. The specific objective of this experiment is to evaluate the changes of response and effect traits caused by the drought treatments. The former will be used to predict how B. hybridum will respond to the projected increasing aridity in the study area, and the latter will be used to determine the impacts of the treatments on ecosystem processes (carbon and nutrient cycling). This projects also pursue to find links between the response traits and changes in the RNA expression of this species to identify the loci of its genome more subjected to selection or epigenetic processes due to drought stress gradient.
