Climate Sense

[

] 144

bserving

, P

redicting

and

rOjecting

limate

OnditiOns

A good example is the structure and results of MIROC-

ESM, the GCM-based ESM developed by Japan Agency

for Marine-earth Science and Technology (JAMSTEC),

under collaboration with Center for Climate System

Research, University of Tokyo, and National Institute

for Environmental Studies.

The target resolution for the model development is

280 kilometres with 80 vertical levels for the atmos-

phere and 50-140 kilometres (spatially variable) with

44 levels for the ocean. The resolution is rather coarse,

particularly considering the fact that JAMSTEC runs

the Earth Simulator, once the world’s fastest computer

and still one of the fastest available for earth scientists.

Coupling many components with a GCM is itself a chal-

lenging task for model developers. Furthermore, some

of the component modules are quite expensive in terms

of computer resources. Plunging into a high-resolution

ESM from the start is not a wise choice. A unique aspect

of MIROC-ESM is its sophisticated treatment of the

stratosphere, leading to reproduction of self-sustained

Quasi-Biennial Oscillation and realistic distributions of

chemical particles such as ozone.

MIROC-ESMwas involved in a coordinated experiment

under C4MIP, in which interactions between climate and

carbon cycle were examined. Results showed that, when

interactions are taken into account, the estimated CO

concentration for 2100 has to be elevated by 120ppm

compared to the case where no interactions are consid-

ered. The mainmechanisms are enhancement of microbial

degradation of soil carbon, and lower uptake of excess CO

by oceans due to lower solubility of CO

, both of which

In order to fill the gap between conceptual models and GCMs,

earth system models with intermediate complexity (EMICs) are

now being vigorously developed. EMICs greatly simplify equations

of motion for the atmosphere and ocean and radiation processes,

retaining the minimal ability to reproduce realistic properties such

as geographical temperature distribution and deep water formation.

EMICs require fewer computer resources, and can be integrated

without supercomputers.

Using the advantage of computational efficiency, EMICs have

often been applied to paleoclimate studies, in which model integra-

tion of tens of thousands of years is common. Embedding vegetation

dynamics and carbon cycle processes in EMICs is an active field.

Indeed, four EMICs with the carbon cycle made a significant contri-

bution to the IPCC Fourth Assessment Report (AR4) by participating

in the Coupled Climate Carbon Cycle Model Intercomparison

Project (C4MIP) together with seven GCMs. EMICs are also suit-

able for tackling problems addressed in the coming Paleo Carbon

Modelling Intercomparison Project.

General circulation models

EMICs constitute, with carefully designed experiments, an impor-

tant element for future projection and interpretation of past events

with timescales longer than a few hundred years. However, they

have the same drawback of a high degree of abstraction as pointed

out for conceptual models, although to a lesser extent. It is desirable

to make a parallel use of GCM-based earth system models (ESMs)

and EMICs in a complementary manner within the allowance

of computer resources. Many of the institutes involved in global

warming projection are developing elaborate ESMs by coupling

biogeochemical modules (vegetation dynamics, carbon cycle and

others) with GCMs.

Components in MIROC-ESM, a GCM-based earth system model

Increasingly elaborate earth systemmodels are being developed for global warming projection by coupling biogeochemical modules with general circulation models

Source: JAMSTEC