The tree water isoscape of a central Pennsylvania catchment: ecohydrologic patterns and processes

Abstract:

The connections between vegetation and catchment hydrology are important for tree physiology, plant geography, stream flow, and transport of solutes within a watershed. While water isotopes from tree stems have been studied extensively to examine source-water differences at a small scale, there has been little emphasis on modeling of plant stem water isotopes at larger scales, due to the expensive and laborious extraction and analysis processes. We characterized the tree stem water for stable isotopes over a landscape (isoscape) at a first-order catchment in central Pennsylvania in order to address the following questions: 1) How does tree water isotopic composition relate to catchment topography and tree characteristics? 2) What are the underlying hydrologic processes that are revealed by tree water isotopes?

We used 267 observations of tree xylem water δ¹⁸O from 121 trees to build a statistical model with candidate variables related to topography and tree characteristics. We then applied the final model to predict the tree xylem water δ¹⁸O composition during the growing season of the remaining trees defined as > 18-cm diameter (at breast height; DBH) in the catchment. The final model included tree canopy height and slope magnitude as predictors, and explained about 56% of variance in tree water δ¹⁸O composition in the catchment. Tree canopy height and degree of slope were both negatively related to tree water δ¹⁸O suggesting the tallest trees and trees on the steepest slopes had tree water isotopic compositions most depleted in heavy isotopes. Each of these suggested the influence of cool-season isotopic inputs. On the valley floor, where tree canopy heights were tallest, the tree water δ¹⁸O composition was likely due to early growing season soil saturation from a shallow ground water table. Conversely, the steep hill slope δ¹⁸O composition may be a result of tree water use of tightly-bound soil water originating from cool season precipitation. The model highlighted the feedbacks between catchment topography and vegetation that may relate to contrasting hydrologic drivers. The independent variables included in the model were derived from widely available LiDAR data to support validation and applications to other ecosystems.