Ecological Modelling 220 (2009) 2272–2280 Contents lists available at ScienceDirect Ecological Modelling journal homepage: www.elsevier.com/locate/ecolmodel Foliage proﬁles of individual trees determine competition, self-thinning, biomass and NPP of a Cryptomeria japonica forest stand: A simulation study based on a stand-scale process-based forest model Motomu Toda a,∗ , Masayuki Yokozawa b,1 , Akihiro Sumida c,2 , Tsutomu Watanabe c,2 , Toshihiko Hara a,c,2 a Pan-Okhotsk Research Center, The Institute of Low Temperature Science, Hokkaido University, Sapporo 060-0819, Japan b National Institute for Agro-Environmental Sciences, Ibaraki 305-8604, Japan c Cryosphere Science Research Section, The Institute of Low Temperature Science, Hokkaido University, Sapporo 060-0819, Japan article info Article history: Received 19 September 2008 Received in revised form 28 April 2009 Accepted 11 May 2009 Available online 17 June 2009 Keywords: MINoSGI Size structure Competitive asymmetry Carbon gain of an individual tree Biomass allocation pattern Maximum carboxylation velocity abstract A simulation study was carried out to investigate simultaneously the effects of eco-physiological param- eters on competitive asymmetry, self-thinning, stand biomass and NPP in a temperate forest using an atmosphere–vegetation dynamics interactive model (MINoSGI). In this study, we selected three eco- physiological relevant parameters as foliage proﬁles (i.e. vertical distribution of leaf area density) of individual trees (distribution pattern is described by the parameter ), biomass allocation pattern in indi- vidual tree growth () and the maximum carboxylation velocity (V max ). The position of the maximal leaf area density shifts upward in the canopy with increasing . For scenarios with  < 4 (foliage concentrated in the lowest canopy layer) or  > 12 (foliage concentrated in the uppermost canopy layer), a low degree of competitive asymmetry was produced. These scenarios resulted in the survival of subordinate trees due to a brighter lower canopy environment when  < 4 or the generation of spatially separated foliage proﬁles between dominant and subordinate trees when  >12. In contrast, competition between trees was most asymmetric when 4 ≤  ≤ 12 (vertically widespread foliage proﬁle in the canopy), especially when  = 8. In such cases, vertically widespread foliage of dominant trees lowered the opportunity of light acquisition for subordinate trees and reduced their carbon gain. The resulting reduction in carbon gain of subordinate trees yielded a higher degree of competitive asymmetry and ultimately higher mortality of subordinate trees. It was also shown that 4 ≤  ≤ 12 generated higher self-thinning speed, smaller accumulated NPP, litter-fall and potential stand biomass as compared with the scenarios with  < 4 or  > 12. In contrast, our simulation revealed small effects of  or V max on the above-mentioned variables as compared with those of . In particular, it is notable that greater V max would not produce greater potential stand biomass and accumulated NPP although it has been thought that physiological parameters relevant to photosynthesis such as V max inﬂuence dynamic changes in forest stand biomass and NPP (e.g. the greater the V max , the greater the NPP). Overall, it is suggested that foliage proﬁles rather than biomass allocation or maximum carboxylation velocity greatly govern forest dynamics, stand biomass, NPP and litter-fall. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Over the past decade, many atmosphere–vegetation dynamics model studies have been conducted, and one of the signiﬁcant aims of the model development is to elucidate the feedback responses of terrestrial ecosystems to climate systems (e.g. Cox et al., 2000; Moorcroft, 2003; Sitch et al., 2003; Watanabe et al., 2004; Sato et al., 2007). The intercomparison of representative models has shown ∗ Corresponding author. Tel.: +81 11 706 3356. E-mail address: todam@ees.hokudai.ac.jp (M. Toda). 1 Tel.: +81 29 838 8435. 2 Tel.: +81 11 706 5455. large uncertainties regarding the carbon storage potential of forests in the future (Cramer et al., 2001; Friedlingstein et al., 2006). The uncertainties possibly remain due to our inability to accurately cap- ture ecological dynamics in forest models (Purves and Pacala, 2008), therefore, suggesting the importance of accurately assessing forest dynamics that inﬂuence the carbon cycle. In the present study, we assess how the differences in eco-physiological parameters affect the dynamic changes of plant size structure (frequency distribu- tion of individual plant size (plant height, stem diameter, mass) in a plant population) due to competition between individual plants and the carbon cycles such as NPP (net primary production) in a forest stand. We dealt with the eco-physiological parameters such as foliage proﬁles of individual trees, biomass allocation pattern in individual plant growth and maximum carboxylation velocity. 0304-3800/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ecolmodel.2009.05.011