1998 Fiscal Year Final Research Report Summary
A Study on Modelling of Stomatal Conductance
| Project/Area Number |
09650649
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Architectural environment/equipment
|
|---|
| Research Institution | KYOTO UNIVERSITY |
|---|
Principal Investigator |
HIRAOKA Hisashi Kyoto University, Graduate School of Engineering, Instructor, 工学研究科, 助手 (80115922)
|
|---|
| Project Period (FY) |
1997 – 1998
|
| Keywords | heat and water budget model within vegetation / stomatal conductance / neural networks / Jarvis's model / semi-empirical model / photosynthetic model |
|---|
| Research Abstract |
The goal of this research is to make a model that estimates a micro-climate of a green treat of land. A heat and water budget model within vegetation is required for such a model. A stomatal conductance model is included in the budget model as a sub-model. The object in this study is to make a model for stomatal conductance. The following is a summary of the results of this research.
1. We reviewed the researches of stomatal conductance models. We clarified that stomatal conductance model were classified into two groups : a phenomenological method and a semi-empirical one. The phenomenological methods were still more classified into three groups : a linear multi-regression, Jarvis's model, and a nonlinear multi-regression using a neural network.
(1) A linear multi-regression model is not applicable, since stomata respond extremely non-linearly to environment factors.
(2) Jarvis's model is not applicable to incorporate an interaction of environmental factors.
(3) Though the semi-empirical model utilizes a photosynthetic model, the model can not take account of the effect of leaf water potential on a stomatal response, at the present stage.
2. We made a stomatal conductance model by using a neural network, We compared the model with Jarvis's model, which was usually used. We clarified that the present model was superior to Jarvis's model in accuracy.
3. We investigated the cause of the superiority of the present model over Jarvis's one. It was that the present model could incorporate the interaction of environmental factors though Jarvis's model could not.
|
