Five reasons to last that theory course

This group of texts addresses a wide range of audiences and there is no question of any of them competing with each other for space on the bookshelf. In some cases, however, there are alternatives already in print.

The new edition of Contemporary Climatology , by Peter Robinson and Ann Henderson-Sellers, is a great improvement. In the 13 years since it first appeared, so much has happened in the field of climate science and this is well reflected in the updated content (even if, as the authors point out, "the fundamental physics of climate have not changed").

The content falls into three sections (14 chapters). The first discusses the fundamental climate variables, the second, regional climate and the last, climate change. Figures are very clear and a modest list of websites is provided in an appendix as a potential kick-off for further research.

A particular strength is that the text incorporates a good meteorological foundation that provides a useful scientific feel, rather than a purely geographical treatment. If I taught an undergraduate course in climatology, I would happily use this as a broad-ranging text with which to touch base, supported by other more focused literature. I hope it will not be quite as long again before another revision is published.

Environmental Physics by Egbert Boeker and Rienk van Grondelle is less satisfying. The content of both editions (the first was published in 1995) is curious and it does not hang together that well. To some extent this is explained by the fact that the text was written for undergraduate and postgraduate physics students to act as a bridge to the environmental sciences.

The authors state in the preface that they pick and choose from the book's eight chapters for the courses that they teach. There are chapters on the essentials: spectroscopy, global climate, energy, pollution, transport and noise and societal implications. In the last chapter, the authors offer opinions rather than allow the student to weigh up the evidence. Most figures in the text are small to keep a lid on the total size.

In summary, the text falls well short of communicating the excitement of physical environmental science (even if environmental physics is not precisely the same thing).

The two texts by L. D. Danny Harvey should be considered as a pair because Climate and Global Environmental Change is a simple subset of Global Warming .

The former belongs to the Understanding Global Environmental Change series and, in pocket-book style, it attempts to cover a bewildering array of topics in three sections: the climate system, climate change and the science-policy interface. The result is that headings appear thick and fast with a frustratingly limited amount of material in between. An undergraduate reader, who would probably not be a climate specialist, would have trouble keeping pace with the rapid twists and turns.

Global Warming: The Hard Science , on the other hand, is a gem that I shall keep to hand for regular reference and that could form the basis of a graduate-level course. It is, to some extent, a rival to John Houghton's Global Warming: The Complete Briefing.

The text covers the material, especially the climate model hierarchy, in an authoritative yet accessible way that is hard to find elsewhere. The figures are of good quality and a substantial reference list and modest website appendix are included.

Statistical Analysis in Climate Research is a reference text that one might look to for advice or reassurance on a particular topic. I am not sure I agree with the authors' suggestion that it is used as the main text for a graduate course in, for example, statistics for climate. At Pounds 65 for the hardback version, there will not be much danger of that anyway.

Although the authors provide examples of real-life applications of the techniques covered, the material would feel less dry if it could be used in support of a parallel course in which there was more time to explore the applications and put them into context. Nonetheless, the importance of sound statistical methodology comes across strongly. In addition to basic statistical concepts, statistical models, time-series analysis, Eigen techniques and forecast performance are usefully covered.

In the preface, the authors draw attention to the fact that Bayesian statistics and geostatistics do not appear. This is a pity because these are two new areas open to use and abuse for which guidance and attention are surely needed. The common illegibility of the output from statistical analyses and climate models has, for the most part, been overcome and figures are generally of reasonable clarity. Overall, this is an excellent addition to the subject literature and I can imagine its being a valuable library resource in support of graduate dissertation work.

All five texts include cutting-edge material underpinned by long-established scientific theory. There is a tremendous opportunity to motivate the reader to "last the theory course" by providing exciting and topical case-study material. To succeed, in this rapidly evolving subject arena, regular revisions are needed. Writing this sort of text seems, increasingly, to be a long-term commitment not just a short-term fix.

Stephen Dorling is lecturer in atmospheric sciences, University of East Anglia.