Book zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGF Reviews zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPON 111 the attendant difficulties for integration with bio- physical models at the global scale. The final chapter by Skole reviews current approaches to acquiring and analysing data on land cover and land cover conversion. Conven- tional and modern methods for mapping and documentation (statistics, maps, remote sensing) over time and space are explored. The use of modern information systems for handling spatial and temporal data (Geographic Information Sys- terns-GIS) is introduced and an analysis of defor- estation in Brazil using natural and social and demographic factors is presented as an example of an interdisciplinary study at regional scale. As a whole the book is recommended reading for all researchers involved in regional or global environmental science. The various chapters pre- sent an excellent overview with detailed refer- ences to many aspects of global environmental issues and the reports of the working groups have already provided guidelines for ongoing research. As such, it has amply achieved the organisers’ goal. What does the book miss out? Given the com- plexities of the problems and the needs to deal with large amounts of multivariate, multiscale spatial and temporal data it is remarkable that GC14 did not include contributions from spatial statisticians or from mathematicians versed in chaos theory or complexity theory, as well as experts in using these skills together with spatial information systems. On p. 386, the editors pose the question “how can factors so diverse and seemingly unpredictable and unconnected be brought into the modeling and forecasting of global land cover changes?“. We do not yet know whether the current inability to obtain satisfac- tory model results is caused by (a) an inability to understand the basic processes (b) a lack of ade- quate mathematical models of those processes, (c) insufficient or inappropriate spatial and tem- poral data at the correct levels of resolution to ensure that the models work or (d) how uncer- tainties in these data and in model parameters propagate uncertainties to the results. Such an interdisciplinary research program as this would surely benefit from (a) incorporation of geostatis- tical methods for extending deterministic mod- elling to deal with probabilities and for upscaling the understanding of natural phenomena from one level to another, (b) chaos theory for under- standing how instability in even simple mathemat- ical models can lead to unstable responses and (c) complexity theory the better to understand how collections of simple systems at one level can lead to structured, coherent behaviour at another. On the other hand, this is too much to expect for one highly productive two-week workshop and could properly form the basis of a future GCI. P.A. Burrough, Utrecht 0012-8252(95)00005-4 zyxwvutsrqponmlkjihgfedcbaZYXWVUT Wind-wave modelling G.J. Komen, L. Cavaleri, M. Donelan, K. Hassel- mann, S. Hasselmann, and P.A.E.M. Janssen, 1994. Dynamics and Modelling of Ocean W aves, Cambridge University Press, Cambridge, U.K. Hardcover, XXI + 532 pp. Price: f40.00. ISBN 0-521-47047-l. We have all seen publications in experimental nuclear physics or space science with a tele- phone-book list of authors, and we have come to realize that progress in those logistically demand- ing fields can only happen through extensive col- laboration. This book is clear evidence that col- laboration also pays off in other scientific endeav- ours. After more than two centuries of intense study, ocean waves continue to pose challenges which require extensive international collabora- tion. This book is an account of the most signifi- cant recent progress in the practical understand- ing and prediction of wind-generated ocean waves, being, in fact, the final report of Working Group 83 of the Scientific Committee for Scien- tific Research (SCOR). The authorship of the book is actually much wider than the list of names on the cover: 35 contributors in all. Wave modelling consists in expressing the characteristics of the sea state in terms of those of the wind that causes it: its speed and direction as well as its spatial distribution and time history.