20
STEPHEN GRIFFIES
The remainder of this section introduces some basic concepts, themes,
and questions, some of which are revisited later in the lectures. We
present some philosophical notions which motivate a focus on fundamental concepts and notions when designing, constructing, and analyzing ocean models.
1.1
Model environments
The field of ocean model design is presently undergoing a rapid growth
phase. It is arguable that the field has reached adolescence, with further
maturation likely taking another 10-20 years as we take the models to
a new level of integrity and innovation. Many applications drive this
evolution, such as studies of climate change, operational oceanography,
and ultra-refined resolution process studies.
One goal of many developers is that the next decade of model evolution will lead to a reduction in code distinctions which presently hinder
the ability of modelers to interchange algorithms, make it difficult to
directly compare and reproduce simulations using different codes, and
increase the burdens of model maintenance in a world of increasingly
complex computational platforms and diverse applications. Notably,
the distinctions will not be removed by all modelers using a common algorithm. Such is unreasonable and unwarranted since different scientific
problems call for different algorithmic tools. Instead, distinctions may
be removed by the development of new codes with general algorithmic
structures flexible enough to encompass multiple vertical coordinates,
different horizontal grids, various subgrid scale (SGS) parameterizations,
and alternate numerical methods.
The word environment has recently been proposed to describe these
highly flexible and general codes. As yet, no model environment exists
to satisfy the needs and desires of most modelers. Yet some models are
moving in this direction by providing the ability to choose more than
one vertical coordinate. This is a critical first step due to the central
importance of vertical coordinates. The present set of lectures formulates
the fundamental equations using generalized vertical coordinates, and
these equations form the basis for generalized vertical coordinate ocean
models. Ideally, the advent of general model environments will allow
scientists to use the same code, even though they may use different
vertical coordinates, horizontal grids, numerical methods, etc.
Many of the ideas presented here are an outgrowth of research and
development with the Modular Ocean Model of Griffies et al., 2004, as
well as the MITgcm (Marshall et al., 1997, Adcroft and Campin, 2004).
The MITgcm provides for a number of depth-based and pressure-based
STEPHEN GRIFFIES
The remainder of this section introduces some basic concepts, themes,
and questions, some of which are revisited later in the lectures. We
present some philosophical notions which motivate a focus on fundamental concepts and notions when designing, constructing, and analyzing ocean models.
1.1
Model environments
The field of ocean model design is presently undergoing a rapid growth
phase. It is arguable that the field has reached adolescence, with further
maturation likely taking another 10-20 years as we take the models to
a new level of integrity and innovation. Many applications drive this
evolution, such as studies of climate change, operational oceanography,
and ultra-refined resolution process studies.
One goal of many developers is that the next decade of model evolution will lead to a reduction in code distinctions which presently hinder
the ability of modelers to interchange algorithms, make it difficult to
directly compare and reproduce simulations using different codes, and
increase the burdens of model maintenance in a world of increasingly
complex computational platforms and diverse applications. Notably,
the distinctions will not be removed by all modelers using a common algorithm. Such is unreasonable and unwarranted since different scientific
problems call for different algorithmic tools. Instead, distinctions may
be removed by the development of new codes with general algorithmic
structures flexible enough to encompass multiple vertical coordinates,
different horizontal grids, various subgrid scale (SGS) parameterizations,
and alternate numerical methods.
The word environment has recently been proposed to describe these
highly flexible and general codes. As yet, no model environment exists
to satisfy the needs and desires of most modelers. Yet some models are
moving in this direction by providing the ability to choose more than
one vertical coordinate. This is a critical first step due to the central
importance of vertical coordinates. The present set of lectures formulates
the fundamental equations using generalized vertical coordinates, and
these equations form the basis for generalized vertical coordinate ocean
models. Ideally, the advent of general model environments will allow
scientists to use the same code, even though they may use different
vertical coordinates, horizontal grids, numerical methods, etc.
Many of the ideas presented here are an outgrowth of research and
development with the Modular Ocean Model of Griffies et al., 2004, as
well as the MITgcm (Marshall et al., 1997, Adcroft and Campin, 2004).
The MITgcm provides for a number of depth-based and pressure-based