succeeded in reconstructing the current climate, independently tried to use the disruption of the
radiative budget calculated by Milankovitch to simulate the last entry into glaciation 115,000
years ago. In both cases, it was a total fiasco. The changes induced by the variation of the
orbital parameters in these models were far too small to generate perennial snow. The components and feedbacks related to the ocean and terrestrial vegetation needed to be included.
Developing a model that couples all three of these components is what modelers have been
striving to achieve over the last 20 years, and these are the models that now contribute to the
international IPCC effort.
Today, the so-called Earth system models that incorporate aspects from the atmosphere–
ocean–terrestrial and marine biosphere, chemistry, and ice caps are used to explore the climate
of the future and the climates of the past. Spatially, they are increasingly precise, they involve
a very large number of processes and are run on the largest computers in the world. But, the
flip side of this complexity is that they can only explore a limited number of trajectories
because of the considerable computing time they require. Also, from the beginning, climate
modelers armed themselves with a whole range of models. From behemoths like the ‘general
circulation models’ to conceptual models, with models of intermediate complexity in-between.
From this toolbox, depending on the questions raised by the paleoclimatic data, they choose
the most appropriate tool or they develop it if it does not exist. With the simplest models, they
can explore the possible parameter variations and, by comparing them with the data, try to
establish the most plausible scenario. All of these modeling strategies are described in detail in
volume II, which constitutes the last and fourth part of this book.
This investigative approach at each step of the research work, dating, reconstruction,
modeling, and the back and forth between these stages allows us to develop and refine the
scenarios to understand the evolution of the past climates of the Earth. We are certain that this
approach also allows us, by improving our understanding of the phenomena that govern the
climate of our planet and through continuous improvement of the models, to better predict
future climate change. This comparison between models and data, which makes it possible to
validate numerical simulations of the more or less distant past, is an essential step toward the
development of climate projections for the centuries to come, which will, in any case, involve
an unprecedented transition.
Introduction
xi
radiative budget calculated by Milankovitch to simulate the last entry into glaciation 115,000
years ago. In both cases, it was a total fiasco. The changes induced by the variation of the
orbital parameters in these models were far too small to generate perennial snow. The components and feedbacks related to the ocean and terrestrial vegetation needed to be included.
Developing a model that couples all three of these components is what modelers have been
striving to achieve over the last 20 years, and these are the models that now contribute to the
international IPCC effort.
Today, the so-called Earth system models that incorporate aspects from the atmosphere–
ocean–terrestrial and marine biosphere, chemistry, and ice caps are used to explore the climate
of the future and the climates of the past. Spatially, they are increasingly precise, they involve
a very large number of processes and are run on the largest computers in the world. But, the
flip side of this complexity is that they can only explore a limited number of trajectories
because of the considerable computing time they require. Also, from the beginning, climate
modelers armed themselves with a whole range of models. From behemoths like the ‘general
circulation models’ to conceptual models, with models of intermediate complexity in-between.
From this toolbox, depending on the questions raised by the paleoclimatic data, they choose
the most appropriate tool or they develop it if it does not exist. With the simplest models, they
can explore the possible parameter variations and, by comparing them with the data, try to
establish the most plausible scenario. All of these modeling strategies are described in detail in
volume II, which constitutes the last and fourth part of this book.
This investigative approach at each step of the research work, dating, reconstruction,
modeling, and the back and forth between these stages allows us to develop and refine the
scenarios to understand the evolution of the past climates of the Earth. We are certain that this
approach also allows us, by improving our understanding of the phenomena that govern the
climate of our planet and through continuous improvement of the models, to better predict
future climate change. This comparison between models and data, which makes it possible to
validate numerical simulations of the more or less distant past, is an essential step toward the
development of climate projections for the centuries to come, which will, in any case, involve
an unprecedented transition.
Introduction
xi