173 Present and Future of Modeling Global Environmental Change: Toward Integrated Modeling, Eds., T. Matsuno and H. Kida, pp. 173–185. © by TERRAPUB, 2001. A Multi-layered Integrated Numerical Model of Surface Physics—Growing Plants Interaction, MINoSGI Toshihiko HARA 1 , Tsutomu WATANABE 2 , Masayuki YOKOZAWA 3 , Seita EMORI 4 , Kumiko TAKATA 5 and Akihiro SUMIDA 1 1 The Institute of Low Temperature Science, Hokkaido University, Sapporo 060-0819, Japan 2 Forestry and Forest Products Research Institute, Tsukuba 305-8687, Japan 3 National Institute of Agro-Environmental Sciences, Tsukuba 305-8604, Japan 4 National Institute for Environmental Studies, Tsukuba 305-0053, Japan 5 Frontier Research System for Global Change, Tokyo 105-6791, Japan INTRODUCTION It has long been recognized that the geographical distribution of functional types of terrestrial ecosystems is determined by climatic conditions such as temperature, precipitation, radiation (e.g., Holdridge, 1947). This can be viewed as the result of competition among plants with different strategies in each climatic regions. On the other hand, climate is influenced by terrestrial ecosystems through its control of exchanges of energy, moisture, momentum, and carbon dioxide (CO 2 ) at the ground surface. That is, the processes between climate and terrestrial ecosystems construct a closed feedback loop, or interaction, which should be elucidated for an extensive understanding of the past, present and future development of both climate and terrestrial ecosystems. Until recently, however, little has been studied regarding this interaction, partially due to a lack of long-term observations and also a lack of theoretical and numerical modeling framework. In particular, this interaction is totally missed in most of the future climate change projection studies (IPCC, 1996). Among climate modelers, the significance of vegetation processes as a lower boundary condition of the atmosphere has been recognized because the land- surface schemes for climate models including vegetation canopy, e.g., SiB (Sellers et al ., 1986) and BATS (Dickinson et al., 1986), were developed and their effectiveness was shown. In those schemes, the vertical structure of vegetation canopy is integrated into one or two layers, and the stomatal control of transpiration is treated as if the canopy were composed of one “big leaf”. Multi-layered schemes, which represent the vertical structure of the canopy and canopy air explicitly, have also been developed mainly for application to studies of boundary- layer meteorology (e.g., Kondo and Watanabe, 1992). Recently, some of the land- surface schemes began to include more sophisticated ecosystem feedbacks. Some ecosystem models originally developed for more biogeochemical purposes, e.g.,