Leaf-trait variation of tundra plants along a climatic gradient: an integration of responses in evergreen and deciduous species

Citation
G. Kudo et al., Leaf-trait variation of tundra plants along a climatic gradient: an integration of responses in evergreen and deciduous species, ARCT ANTARC, 33(2), 2001, pp. 181-190
Citations number
43
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Multidisciplinary
Journal title
ARCTIC ANTARCTIC AND ALPINE RESEARCH
ISSN journal
1523-0430 → ACNP
Volume
33
Issue
2
Year of publication
2001
Pages
181 - 190
Database
ISI
SICI code
1523-0430(200105)33:2<181:LVOTPA>2.0.ZU;2-I
Abstract
To understand response patterns of leaf traits in tundra plants against dec reasing annual season length comprehensively, a graphic model based on carb on balance theory was presented. The model predicted that leaves with very short life-span and high nitrogen concentration (N-mass) or leaves with ver y long lift-span and small N-mass will be dominate under the conditions of a short growing season. To test this prediction, leaf life-span and other l eaf traits of 26 tundra species were compared among four sites selected alo ng a gradient of climatic harshness: a subalpine site at Abisko in northern Sweden, two nearby mid-alpine sites at Latnjajaure with early and late sno wmelt, and a site at Ny-Alesund (Svalbard) in the High Arctic. In herbaceou s and deciduous shrub species, leaf life-span and/or leaf mass per unit are a (LMA) tended to decrease, and leaf N-mass tended to increase along the cl imatic gradient with decreasing growing season and lower temperatures. In e vergreen shrub species, both leaf life-span and leaf N-mass tended to incre ase under harsh conditions, but the response pattern of LMA was less clear. Deciduous species produced short-lived leaves with a low construction cost and probably high photosynthetic potential, whereas evergreen species prod uced long-lived leaves. The contrastive patterns of leaf-trait variation be tween the deciduous and evergreen plants observed in this study were concor dant with the prediction of the model.