When it comes to understanding the why's and wherefores of climate, there is an infinite amount one needs to know, but life affords only a finite time in which to learn it; the time available before one's fellowship runs out and a PhD thesis must be produced affords still less. Inevitably, the student who wishes to get launched on significant interdisciplinary problems must begin with a somewhat hazy sketch of the relevant physics, and fill in the gaps as time goes on. It is a lifelong process. This book is an attempt to provide the student with a sturdy scaffolding upon which a deeper understanding may be hung later.
The climate system is made up of building blocks which in themselves are based on elementary physical principles, but which have surprising and profound collective behavior when allowed to interact on the planetary scale. In this sense, the "climate game" is rather like the game of Go, where interesting structure emerges from the interaction of simple rules on a big playing field, rather than complexity in the rules themselves. This book is intended to provide a rapid entree into this fascinating universe of problems for the student who is already somewhat literate in physics and mathematics, but who has not had any previous experience with climate problems. The subject matter of each individual chapter could easily fill a textbook many times over, but even the abbreviated treatment given here provides enough core material for the student to begin treating original questions in the physics of climate.
The Earth provides our best-observed example of a planetary climate, and so it is inevitable that any discussion of planetary climate will draw heavily on things that can be learned from study of the Earth's climate system. Nonetheless, the central organizing principle is the manner in which the interplay of the same basic set of physical building-blocks gives rise to the diverse climates of present, past and future Earth, of the other planets in the Solar system, of the rapidly growing catalog of extrasolar planets, and of hypothetical planets yet to be discovered. A guiding principle is that new ideas come from profound analysis of simple models - thinking deeply of simple things. The goal is to teach the student how to build simple models of diverse planetary phenomena, and to provide the tools necessary to analyze their behavior.
This is very much a how-to book. The guiding principle is that the student should be able to reproduce every single result shown in the book, and should be able to use those skills as a basis for explorations that go beyond the rather limited display of results that can be presented in a printed tome of reasonable size. Similarly, the student should have access to every data set used to produce the figures in the book, and ideally to more comprehensive data sets that draw the student into further and even original analyses. To this end, I have set as a ground rule that I would not use reproductions of figures from other works, nor would I show any results which the student would not be able to reproduce. With the exception of a very few maps and images, every single figure and calculation in this book has been produced from scratch, using software written expressly for the purposes of this book and provided as an online software supplement. The computer implementation have pedagogy as their guiding principle, and readability of the implementation has been given priority over computational efficiency . A companion to this philosophy is what I call "freedom to tinker." The code should all be in a form that can easily be modified for other purposes. The goal is to allow the student to first reproduce the results in the book, and then use the tools immediately as the basis for original research. In this, I have been much inspired by what the book Numerical Recipes did for numerical analysis. This book does not sell fish. Instead, it teaches students how to catch fish, and how to cook them. As gastronomical literature goes, the book before the reader is somewhat in the spirit of one of Elizabeth David's extended pedagogical discourses on food (with recipes interspersed), whereas Numerical Recipes is somewhat more in the spirit of a traditional cookbook like Joy of Cooking.
The software underlying this book was implemented in the open-source interpreted language Python, because it lends itself best to the design principles annunciated above. It has a versatile and powerful syntax but nonetheless is easy to learn. In my experience, students with no previous familiarity with the language can learn enough to make a substantial start on the problems in this book in only two weeks of self-study or computer labs. Python also teaches good programming style, and is a language the student will not outgrow, since it is easily extensible and provides a good basis for serious research computations. I do hope that the student and instructor will fall for Python as madly as I have, but I emphasize that this book is not Python-specific. The text focuses on ideas that are independent of implementation, and even in the Workbook section of each chapter, Python-specific advice is isolated in clearly demarcated Python tips. The instructor who wishes to make use of some other computer language in teaching the course will find few obstacles. The transparency and readability of Python is such that the Python implementations should provide a convenient aid to re-implementation in other languages. It is envisioned that Matlab versions of most of the software will ultimately be made available.
In this book I have chosen to deal only with aspects of climate that can be treated without consideration of the fluid dynamics of the Atmosphere or Ocean. Many successful scientists have spent their entire careers productively in this sphere. The days are long gone when leading-edge problems could be found in planetary fluid dynamics alone, so even the student whose primary interests lie in atmosphere/ocean dynamics will need to know a considerable amount about the other bits of physics that make up the climate system. There are many excellent textbooks on what is rather parochially known as "geophysical fluid dynamics," from which the student can learn the fluid dynamics needed to address that aspect of planetary climate. That does not prevent me from entertaining a vision of adding one more at some point, as a sequel to the present volume. This sequel, entitled Things that Flow would treat the additional phenomena that emerge when fluid dynamics is introduced. It would continue the theme of taking a broad planetary view of phenomena, and of providing students with the computational tools needed to build models of their own. It would take a rather broad view of what counts as a "flow," including such things as glaciers and sea ice as well as the more traditional atmospheres and oceans. We shall see; for the moment, this is just a vision.
Since I have in mind the full variety of planets in our Solar System and in extrasolar systems, there is the question of what kind of terminology to use to emphasize the generality of the phenomenon. Should we create new terminology that emphasizes that we are talking about an arbitrary system, at the risk of creating confusion by introducing new jargon? Or should we adopt terminology that emphasizes the analogy with familiar concepts from Earth and our own Solar System? For the most part, I have adopted the latter approach, which leads to a certain amount of Earth-centric terminology. For example, if I sometimes refer to "the sun" it is to be thought of as referring to whatever star our planet is orbiting, and not necessarily Earth's Sun or even a star like it. In the same spirit, the term solar constant, denoted by the symbol Lq, will refer to the radiation flux from the parent star of the planet we are considering, measured at whatever position in its orbit the planet happens to be in, whatever the planet may be and whatever the star it may be orbiting. The symbol Lq does not, in this book, refer to the specific number giving the annually averaged radiation flux measured at Earth's orbit at this particular era in time. A more proper term would be stellar radiation or stellar constant, but that unfortunately calls to mind starlight from the night sky and seems needlessly confusing (though I will gradually break in the use of the term to help the reader get used to the idea that there are a lot of stars out there, with a lot of planets with a lot of climates). The radiation from a planet's star will also sometimes be referred to as shortwave radiation, to emphasize that it is almost invariably of considerably shorter wavelength than the thermal radiation by which a planet cools to space.
In a similar vein "air" will mean whatever gas the atmosphere is composed of on the planet in question - after all, if you grew up there, you'd just call it "air." When I need to refer to the specific substances that make up our own atmosphere, it will be called "Earth air." All this is a bit like the way one refers to Martian "geology" and "geophysics," so we don't need to refer to Areophysics on Mars and Venerophyics on Venus when we are really talking about the same kind of physics in all these cases. Eventually, we will all need to learn to get used to terms like "periastron" as a generalization of "perihelion," as the focus of the field shifts more to the generality of phenomena amongst planetary systems.
To improve the readability of inline equations, I will usually leave out parentheses in the denominator. For example, a/2n is the same thing as , whereas I would write (a/2)n or na/2 if meant an was intended.
With few exceptions, SI units (based on kilograms, meters and seconds) are used throughout this book. To avoid the baggage of miscellaneous factors of 1000 floating around, when counting molecules kilogram-moles are used, denoted with a capital, i.e. Mole. Thus 1 Mole of a substance is the number of molecules needed to make a number of kilograms equal to the molecular weight - one thousand times Avogadro's number. There are a few cases where common practice dictates deviations from SI units, as in the use of millibars (mb) or bars for pressure when Pascals (Pa) involve unwieldy numbers, or the use of cm-1 for wavenumbers in infrared spectroscopy.
The short exercises embedded in the text are meant to be done "on the spot," as an immediate check of comprehension. More involved and thought-provoking problems may be found in the accompanying Workbook section at the end of each chapter. The Workbook provides an integral part of the course. Using the techniques and tools developed in the Workbook sections, the student will be able to reproduce every single computational and data analysis result included in the text. The Workbook also offers considerable opportunities for independent inquiry launching off from the results shown in the text.
There are four basic kinds of problems in the Workbook. Some calculations are analytic, and require nothing more than pen and paper (or at most a decent pocket calculator). Others involve simple computations , data analysis, or plotting of a sort that can be done in a spreadsheet or even many commercially available graphing programs, without the need for any actual computer programming. Many of these problems involve analysis of datasets from observations or laboratory experiments, and all critical datasets are provided in the online supplement to this book in a tab-delimited text format which can be easily read into software of any type. Students who have competence in a programming language, either from prior courses or because the instructor has integrated programming instruction into the climate sequence (as is done at The University of Chicago) have the option of doing these problems in the programming language of their choice.
They should be encouraged to do so, since these simpler problems make good warm-up exercises allowing students to consolidate and hone their programming skills. The third class of problems requires actual programming, but can easily be carried out from scratch by the student in the instructor's language of choice (perhaps with the assistance of some standard numerical analysis routines). While just about any language would do, I have found that interactive interpreted languages such as Python and Matlab offer considerable advantages, since they provide instant feedback and encourage exploration and experimentation.
There are just a handful of basic computational methods and computer skills needed to do the Workbook problems and to reproduce all the calculations in this book. None of the calculations require any more computer power than is available on any decent laptop computer. The required numerical skills are outlined and exercised in the Chapter 1 Workbook section, which the student should master before proceeding to the rest of the book. I have not provided detailed discussions of basic algorithms like ordinary differential equation integration, interpolation, or numerical quadrature, since they are well described in the book Numerical Recipes, available from Cambridge University Press. Numerical Recipes should be viewed as an essential companion to this book, though only a small part of the material in that opus is actually required for the problems that concern us here.
Was this article helpful?