Simulating oxygen isotope ratios in tree ring cellulose using a dynamic global vegetation model

Sonja G. Keel,Matthias Saurer,Fortunat Joos,Rosemarie B. Weigt,Renato Spahni,Stefan Klesse

doi:10.5194/bg-13-3869-2016

Abstract

Abstract. Records of stable oxygen isotope ratios in tree rings are valuable tools to reconstruct past climatic conditions and investigate the response of trees to those conditions. So far the use of stable oxygen isotope signatures of tree rings has not been systematically evaluated in dynamic global vegetation models (DGVMs). DGVMs integrate many hydrological and physiological processes and their application could improve proxy-model comparisons and the interpretation of oxygen isotope records. Here we present an approach to simulate leaf water and stem cellulose δ18O of trees using the LPX-Bern DGVM (LPX-Bern). Our results lie within a few per mil of measured tree ring δ18O of 31 different forest stands mainly located in Europe. Temporal means over the last 5 decades as well as interannual variations for a subset of sites in Switzerland are captured. A sensitivity analysis reveals that relative humidity, temperature, and the water isotope boundary conditions have the largest influence on simulated stem cellulose δ18O, followed by all climatic factors combined, whereas increasing atmospheric CO2 and nitrogen deposition exert no impact. We conclude that simulations with LPX-Bern are useful for investigating large-scale oxygen isotope patterns of tree ring cellulose to elucidate the importance of different environmental factors on isotope variations and therefore help to reduce uncertainties in the interpretation of δ18O of tree rings.

Highlights

IntroductionModel results are evaluated against stable isotope ratios in precipitation (Joussaume et al, 1984), snow (Jouzel et al, 1987), ground water (Hoffmann et al, 1998), water vapor (Werner et al, 2011), and ice core δ18O data (e.g., Risi et al, 2010)
Annual mean soil water δ18O values simulated by ECHAM5-JSBACH range between −1 and −21 ‰ (1960– 1990; Fig. 1) and are in the same range as reconstructions of δ18O in precipitation from the Global Network for Isotopes in Precipitation (GNIP) database (Bowen and Revenaugh, 2003)
Formulations to describe δ18O in leaf water and stem cellulose are implemented in the LPX-Bern dynamic global vegetation models (DGVMs) and a compilation of tree ring data of δ18O in stem cellulose, mainly for Europe, is established

Summary

Introduction

Model results are evaluated against stable isotope ratios in precipitation (Joussaume et al, 1984), snow (Jouzel et al, 1987), ground water (Hoffmann et al, 1998), water vapor (Werner et al, 2011), and ice core δ18O data (e.g., Risi et al, 2010). Because none of these models describe δ18O in stem cellulose, a direct model–data comparison is not yet possible for tree rings and global-scale models. The importance of individual factors such as rising atmospheric CO2 could be examined

Objectives

Methods

Results

Conclusion