PS 36-190 - The relationship between tree hydraulics and rock fractures in bedrock cliffs

Background/Question/Methods

In most terrestrial ecosystems, water availability from soils regulates the water movement through plants. Rocks have long been neglected as unsaturated reservoirs because they're difficult to access and are frequently considered saturated with groundwater. However, their influence on vegetation has recently been acknowledged as a key ecohydrological process. Trees growing in fractures on bedrock cliffs provide us with a natural laboratory to study the effects of rock moisture on trees under these, almost ideal, assumptions: no tree competition, direct input of overland flow into the fracture fillings, and no interaction with the water table.This study aims to explore the response of plants to an extremely water-limited environment and determine the primary sources of the water uptaken by the trees. We selected two tree species, Pseudotsuga menziesii and Picea engelmannii, which commonly grow on both soils and limestone cliffs in the Canadian Rockies. We measured sap flow, stem water potential, and fracture 'substrate' moisture for trees growing in rock fractures and glacial till (soil). We also measured weather variables and accumulated runoff water. We parameterized a modified Soil-Plant-Atmosphere Continuum (SPAC) using the collected data. Additionally, we collected water from precipitation, plant, and soil samples for isotopic analysis.

Results/Conclusions

We show that the model predicts the sap flow and water potential relatively well under certain conditions, e.g., at the start of the season where overland flow is driven by snowmelt and the fillings of the fractures have higher moisture values. At the peak of the growing season, where we have almost no water input, trees get progressively more stressed (pass the expected hydraulic conductivity loss for those two species). However, they still move water at a median daily maximum between 0.2 L/h to 2 L/h. Preliminary isotopic analysis for deuterium and oxygen shows that the water inside the trees at this point of the season has a different isotopic signal than the fracture ‘substrate’. These signals can be due to the crack surface evaporation but could also mean trees are uptaking water from rock moisture and plant tissue storage. Overall, our results show that plant regulation coupled with rock water storage in highly water-limited environments is crucial to correctly model water movement through plants. Future work will include a more detailed sampling (to localize the main water sources) and modifications to the mechanistic processes involved to better model tree hydraulics in rock-dominated environments.