Recent increase in forest mortality events worldwide and their relationship with drought episodes highlight the importance of understanding tree resilience to a changing climate. Empirical models of forest mortality have been typically used and were focusing on carbon related variables such as growth to predict tree death. Recent efforts have shifted toward a more mechanistic modeling of mortality. Mechanistic approaches use tree traits and climate as inputs to model processes and represent carbon and water fluxes, all necessary to plant life. The advantage of mechanistic approaches is their ability to account for potential adaptation of trees to climate change, but also to physically explore the causes of vulnerability and resilience to droughts. Mechanistically, the atmospheric demand for moisture at the canopy level is communicated to the tree through stomata, creating a water gradient between the leaves and the roots, and resulting in the ascent of sap via the plant hydraulic structure. Depending on the climate (temperature, atmospheric dryness, light, precipitation), different architectures will perform differently at maintaining the gradient. For example, deep roots can access deep water in dry regions and shallow roots can access rare precipitation events whereas larger leaf area increases the atmospheric demand for moisture. In very harsh conditions such as extreme or lasting droughts, the hydraulic structure might suffer from a steep water gradient. Protection against excessive gradients can be achieved either through an investment in a stronger structure (denser wood) or through a regulation of the pulling force at the top of the canopy (closing leaf stomata). Evolution of structures and physiological strategies have resulted in fitness advantages and partially explain the diversity of species architectures across climates. More importantly, this diversity is at the core of the vulnerability and resilience of each species to increased aridity and frequency of extreme events, and therefore its mortality.
This dissertation investigates the resilience to droughts of two co-occuring species in common woodlands of New Mexico, USA. This location is of specific interest because drought conditions (high temperature and/or low precipitation) have become more frequent as a result of global warming and because these ecosystems have suffered extensive mortality in the last decades. The two species, Pinus edulis and Juniper monosperma have very different physiological strategies, which allows for an extra level of vulnerability comprehension. To further test their resilience to extreme drought and possibly future climatic conditions, I studied trees that were subject to a six-year rain-reduction experiment.
In the first part we develop a mechanistic model of the tree functioning that includes water and carbon fluxes and is based on their respective supply-demand balances. We use this simplified mechanistic model to study the sensitivity of mortality to hydraulic structure variations and to the physiological strategy of each species. We find that for both species death resulted from an irreversible damage of tissues transporting water. Despite P. edulis’s ability to close stomata to reduce the atmospheric demand, they died first because of their vulnerable tissues. In the second part, I specifically investigate P. edulis’s structural response to drought at the canopy level. By dissecting branch anatomy at an annual resolution, I find that during droughts this species increase relatively more leaf area (water demand) compared to transport area (water supply). I suggest that the structural adjustments that occur at the branch level do not contribute to the protection of the tissues transporting water. In the third part, I analyze the anatomy of these tissues in branches of P. edulis. I find that in response to long-lasting drought the trees built tissues more efficient at transporting water but also more vulnerable to future drought. By contrast, a short-intense drought decreases efficiency without changing vulnerability. I hence show that during lasting drought the anatomical adjustment of branch tissues increase the vulnerability of the piñons.
This study shows the importance of considering climate responses of structure and physiology together to compare resilience across species. It also shows that adjustments of hydraulic elements in response to drought tend to decrease hydraulic resilience and could favor a run-away scenario. If the population of Pinus edulis - a dominant species of the Southwest US - were to decline, major shift should be expected in related ecosystems.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8CJ9XB1 |
| Date | January 2018 |
| Creators | Guerin, Marceau |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0021 seconds