Incorporation of plant traits in a land surface model helps explain the global biogeographical distribution of major forest functional types. Issue 3 (13th November 2016)

• Record Type: Journal Article
• Authors: Lu, Xingjie; Wang, Ying‐Ping; Wright, Ian J.; Reich, Peter B.; Shi, Zheng; Dai, Yongjiu
• Other Names: Enquist Brian checker.
• Is Part Of: Global ecology & biogeography. Volume 26:Issue 3(2017)
• Page Start: 304
• Page End: 317
• Publication Date: 2016-11-13
• Subjects: Ensemble simulations -- global biogeography -- plant functional types -- plant strategy -- plant traits -- trait continua based modelling; Ecology -- Periodicals; Biogeography -- Periodicals; Biodiversity -- Periodicals; Macroevolution -- Periodicals; 577
• DOI: 10.1111/geb.12535
• Languages: English
• ISSNs: 1466-822X

Abstract

Abstract: Aims: Contrasting leaf types with different leaf life spans represent different adaptive strategies in plants. Previous studies explained the adaptive advantages of different strategies on the basis of environmental climatic limits, but could not account for the observed co‐dominance of multiple plant functional types (PFTs) in many regions. Here we used a process‐based model to explore whether observed inter‐ and intra‐PFT variation in plant traits can explain global biogeographical variation in the dominance and co‐dominance of major forest types. Location: World‐wide. Methods: We identified four important plant traits: leaf N concentration, leaf life span, fraction of net primary production (NPP) allocated to leaves and plant basal respiration rate. We incorporated means and variances of these plant traits from trait databases into the Community Atmosphere–Biosphere–Land Exchange model. We then predicted the dominant PFT or PFT mixture for global forested grid cells, using NPP as a proxy for growth rate, and considering three PFTs: evergreen needleleaf forests (ENF), evergreen broadleaf forests (EBF) and deciduous broadleaf forests (DBF). We compared results with: (1) those from simulations that did not account for trait variance; (2) results from an empirical model based purely on mean annual temperature; and (3) data from remotely sensed observations. Results: Our estimates of the fractions of land area covered by major forest types were consistent with observation; i.e. ENFs dominate in boreal regions, EBFs dominate in tropical regions and DBFs are distributed widely across a broad range of environmental conditions. We also showed that co‐dominance of different forest PFTs cannot be reproduced without considering variation in plant traits. Main conclusions: Global trait data are useful for representing underlying plant strategies and functional diversity. Variation in key plant traits explains significant fractions of global biogeographical variation of three major forest types. Future developments in dynamic global vegetation modelling will benefit from the inclusion of plant trait variation.