-------
=========================================
== UK Nonlinear News ==
== Issue 5 : July 1996 ==
== ASCII edition, Part (1/3) ==
=========================================
---------------
My Book Cases
---------------
by Carsten Knudsen
Department of Applied Mathematical Studies,
University of Leeds.
email: carsten@amsta.leeds.ac.uk.
The explosive combination of a strong interest in nonlinear science and a
lack of common sense in economical decisions has left me with a considerable
collection of books related to nonlinear science and dynamical systems in
particular. After becoming a pop-subject the field of dynamical systems was
inundated with new text books, monographs, reprint collections etc.
Consequently the book market is now a jungle and newcomers may want some
advice before investing too much in fancy, but colourful, books. On the
other hand they probably won't. Too late, because here is my $0.02's worth.
* Introduction
* The Classics
* Textbooks
* More Specialised Books
* Numerical Aspects
* Concluding Remarks
* Appendix: Contents of Books
* Appendix: Bibliography
Introduction
------------
The first I heard about nonlinear science was something about strange
pictures which under magnification appeared very different; apparently these
pictures could be magnified repeatedly and keep showing new interesting
structures. In the university book shop I saw some of these pictures in the
book by Peitgen and Richter[25]. Of course, in those days, being a student,
I had plenty of money for food, rent and other inessentials. Needless to
say, I bought the book.
In the following I attempt to give a superficial, but unfair, survey of some
of the books in my collection. For good measure I should mention that I am
unreasonably biased for dissipative systems and somewhat indifferent toward
subjects such as Hamiltonian systems, complex dynamical systems, and
fractals. So if you have a problem with that, well, that is your problem.
The Classics
------------
Few books in the general area of nonlinear science have risen to the status
of undisputed classics. One that have achieved this fame is undoubtedly (and
deservedly) Guckenheimer and Holmes[12]. Their coverage of dynamical systems
is thorough, although brief, making it demanding to read. Don't read this
book from cover to cover (few books are designed to be used this way).
However, if you need to look up a definition or find an example, chances are
you will find it here. As with most of the older books this book is mainly
concerned with differential equations, but most results, eg the centre
manifold theorem, are also stated for maps. A second edition, containing no
major alterations, has appeared. It remains good value for money.
Slightly pre-dating Guckenheimer and Holmes is Lichtenberg and
Lieberman[19]. Their emphasis is very different, focusing on mechanics, in
particular Hamiltonian systems. Classical subjects such as perturbation
theory are also covered. Only the final chapter deals with dissipative
dynamical systems; the coverage is, however, quite decent even by todays
standards (or possibly because of), eg the description of Melnikov's method
is not bad.
I should also mention Thompson and Stewart[32]. Although very different
stylistically from Guckenheimer and Holmes, it is a nice first text in
dynamical systems. As the subtitle of the book suggests, it presents a
geometrical view on dynamical systems, and gets away with it quite
successfully. It helped me understand global bifurcations, such as crises,
and in the process taught me the importance of a good geometric intuition.
The book takes the differential equation view, but contains two chapters on
bifurcations and chaos in maps. These chapters are non-rigorous but neatly
intuitive; a good read. Some comments on Sarkovskiis theorem is a bit, hmmm
(p. 170, 1st ed.). A second edition (supposedly available) includes a useful
glossary of geometrical dynamics.
One of my favourites is Devaney[8]. This deals mainly with iterated maps and
the presentation is very rigorous. The textbook starts from scratch with
virtually no prerequisites and very rapidly arrives at an analysis of
chaotic motion. This is a winner with students (at least it was with me).
Having said this, the book is not an easy read; some commitment and
determination are required. The second edition is divided into three main
parts: the first deals with one-dimensional dynamics; the second treats
higher dimensional theory, though not beyond the third dimension; and in the
final part complex dynamical systems, ie primarily Julia sets and the
Mandelbrot set, are covered.
If I had to spend time on an isolated desert island with a single book from
my collection, my choice would probably be Abraham and Shaw[1] (previously
published in four separate volumes at an obscene price). This book is quite
close to being ingenious. It presents a very intuitive approach to dynamical
systems, explained through a mother lode of drawings which develop from from
simple two-dimensional trajectories to the complex structure of stable and
unstable manifolds in three-dimensional phase space. This book also contains
a large amount of historical information, a particularly interesting part is
the account of the origin of bifurcation analysis. However, though the style
is very relaxed and non-rigorous you shouldn't be fooled: it is not a
completely straightforward read, and it covers a lot of ground including
structural stability, explosive bifurcations and Birkhoff signatures.
Textbooks
---------
New textbooks keep appearing, and in the following I shall comment on a few
of these, as well as some not so new ones. The presentation is not
chronological, but by subject matter covered.
Wiggins[33] has been very productive and now has five books to his name,
only three of which will be mentioned here (I only have the first four).
Wiggins follows the tradition of Guckenheimer and Holmes in material, but
uses a more modern style. The treatment is very rigorous, though exclusively
centered around differential equations. The section on symbolic dynamics is
short but nicely written. The final chapter deals with global perturbation
methods, something not often considered in our educational system. The level
of difficulty is quite high (roughly post-postgraduate level) and to solve
this problem Wiggins wrote another, more entry level, dynamical systems
text[34]. This book remains one of the best dynamical systems texts; it is
very complete and yet fairly easy to read. Certain sections are virtual
copies from[33], but never mind, as they were good in the first place. Note
that there is quite a few typos, eg Fig. 1.1.2, so don't take everything at
face value.
Arrowsmith and Place[2] provide a very thorough account of most standard
aspects of dynamical systems theory. There is a considerable amount of
material on Hamiltonian systems; the last chapter (over 90 pages) deals with
area-preserving maps and their bifurcations. Each chapter ends with a large
number of exercises (hints are given towards the end of the book). This is a
good textbook on dynamical systems, although it is far from elementary and
biased towards Hamiltonian systems.
Hale and KoIak[14] have written a clear exposition of introductory dynamical
systems theory. It covers standard material such as stability and local
bifurcations of flows and maps, global bifurcations are only touched upon.
The book is well-written and, dare I say, very pedagogical. There are many
good examples and illustrative figures (though many of these are clearly
produced from low resolution screen dumps). This may well be the best
introductory text considered in this review, assuming that you can forgive
the authors the last three chapters which are too superficial to benefit
students.
Perko[26] presents a very clear exposition of the differential equations
side of dynamical systems. The book is very complete and almost all proofs
are included, which makes it an excellent reference book, as well as a good
textbook.
Glendinning[9] has written a book centered around differential equations.
(This was reviewed in UK Nonlinear News 1.) Quite a bit of the book is spent
on local analysis and one chapter covers classical subjects, such as
perturbation theory. Only the final two chapters (out of 12) deals with
chaos and global bifurcations. The former primarily deals with
one-dimensional iterated maps. The exposition of the latter is a bit
non-standard and frankly is not my first choice.
Gulick[13] is very similar in material, and partly presentation, to Devaney;
I very much prefer the original. The best part of Gulick is the section on
iterated function systems, where Hausdorff distance, completeness, the
contraction mapping theorem, and Cauchy sequences etc are treated.
Another Devaney-look-alike is Holmgren[15]. Holmgren considers virtually the
same basic material as Devaney, ie one-dimensional and complex maps. The
level is more introductory than Devaney's, and many of the more advanced
subjects considered by the latter are not mentioned. The general
presentation is very good, and there are many illustrative figures. There is
a good section on Newton's method for cubic polynomials.
Crownover[6] places an emphasis on fractals, but includes quite a bit of
chaos theory as well. The dynamical systems part contains material on both
one-dimensional maps (a traditional treatment) and on maps in more general
metric spaces. An example of one of the more abstract mapping covered is
iterated function systems. Some recent results relating to the definition of
chaos are mentioned. Complex dynamics are mentioned, although rather
superficially.
In Robinson[27] chapters two and three are Devaney-like in subject matter,
except that more concepts are considered, often in a setting of a general
metric space. This is followed by standard material on differential
equations and a smallish section on bifurcations. Later, a large amount of
non-standard material is covered, topological entropy receives considerable
attention. In general, the text considers the abstract side of dynamical
systems. On the whole Robinson gets away with it, and although at a very
high level of difficulty it is commendable.
Moon[21] makes a very informal presentation of basic nonlinear dynamics such
as period-doubling cascades, intermittency and fractals (including fractal
basin boundaries). The most technical aspect is the mentioning of Melnikov's
method. A nice aspect of the book is that it very often compares theory with
experimental results. A new and improved version is available, but since I
wasn't particularly impressed by the first version I haven't checked it out,
aside from the usual glance in the book store.
Schuster[29] presents a large number of dynamical systems results from a
non-rigorous perspective. He covers the standard routes to chaos, eg
period-doubling cascades and intermittency. This is the sort of book that
contains a large number of pictures. A new edition is on the market,
containing new appendices.
A newer book following the same tradition is Ott[22]. Ott covers many,
relatively new, subjects that are often not mentioned in dynamical systems
texts: control of chaos, crisis phenomena, nonattracting chaotic sets,
multifractals, quantum chaos. Some of these sections are on the short side.
Well-written and amply illustrated (sorry, no colour pics). It is available
in a reasonably cheap paperback version, just what our students need. Good
value for money and an excellent first text if mathematics is not of the
essence.
The first half of Marek and Schreiber[20] presents an informal description
of dynamical systems theory, including fractals and chaos theory. A section
on numerical methods is included, and an appendix contains Fortran code for
continuation. In the second half a large number of examples are provided,
often with some bifurcation analysis involved. Most of the examples are
chemical systems, and many experimental results are mentioned. Each chapter
contains a large number of references.
Bai-lin[3] presents an unusual, but interesting, description of dynamical
systems, with primary focus on symbolic dynamics. This book contains a great
deal of valuable information on symbolic dynamics. In the second part of the
book more standard subjects are treated, eg dimensions, entropies, Lyapunov
exponents and multifractals. The first part is definitely the best. A plus
point is that practical numerical aspects are discussed several times in the
book, but do not expect state-of-the-art methods. The book is not intended
as a textbook, but even so the presentation and organisation is lacking. I
have not seen many of the subjects treated in this book in books before.
More Specialised Books
----------------------
Iooss and Joseph[16] treat stability and bifurcations of solutions to
differential equations. The exposition is very detailed and complete, but is
too technical for my taste; it contains very few good examples. As a
reference book on local bifurcations it may serve a purpose. The vey short
final chapter outlines stability and bifurcations in conservative systems.
Why the title of this text contains the word `elementary' is beyond me; the
text is very demanding, for example, on page one the general evolution
problem is stated in nonlinear operator form. A much more comprehensive
treatment of bifurcation theory from the point of view of singularity theory
can be found in Golubitsky and Schaeffer[10] and Golubitsky, Stewart and
Schaeffer[11].
Kuznetsov[18] (see Alan Champney's review in UK Nonlinear News 2) has
written a quite comprehensive text on bifurcation theory with some emphasis
on practical aspects. This book contains a wealth of valuable information
about global and codimension-two bifurcations. Another positive aspect is
the final chapter which addresses numerical analysis of dynamical systems; I
would have liked this chapter to be even longer, but never mind as it is the
best found in a standard text (so far). The book contains some of the best
illustrations I have seen for a long time, with the possible exception of
Abraham and Shaw, but that is a somewhat high standard. It is an excellent
reference book that you should have access to, but it should not be read
cover to cover.
Ruelle[28]. This book is in the French tradition, need I say more? The
treatment of just about everything in this book is very brief, however, it
does cover a lot of ground. Only consult this book if you know the subject
already.
If you should ever find yourself with a lack of purpose in life, then try to
understand every detail in Collet and Eckmann[5]. That should keep you busy
for a while. Collet and Eckmann focus entirely on the dynamics of
one-dimensional mappings, apart from the first part which is merely
motivation. The two main parts of the book treat kneading theory for
individual iterated maps and for families of maps. Ergodic theory is often
ignored or mentioned superficially in dynamical systems texts, here it is
addressed in some detail. If you find reading this book easy going you are
ready for de Melo and van Strien[7]. This is the authoritative source on
one-dimensional dynamical systems. This covers the works: combinatorial,
topological, ergodic as well as smooth theory. Everything is covered from
the (relatively) simple circle diffeomorphisms to renormalisation theory.
Taking on this book may well earn you the Purple Heart. Excellent for
references, but reading requires a solid background in maths. Block and
Coppel[4] is a slightly earlier book dealing with similar material. This
book also deals exclusively with one-dimensional dynamics, but from a
topological viewpoint, containing no information on ergodic theory. The
exposition is very complete including almost all proofs.
Palis and Takens[23] present a wealth of results on homoclinic bifurcations,
including some of the more recent results on homoclinic bifurcations and
strange attractors. Beware, this is a pure mathematics book and is extremely
challenging.
In Sparrow[31] you will find a quite good exposition of one of the most
classical dynamical systems: The Lorenz equations. Although focused on an
extensive case study, you may still learn much of the general theory from
this book; stability, Lyapunov functions, global bifurcations and symbolic
dynamics are all covered. Don't let it be your first book on dynamical
systems though.
The last Wiggins book to receive a mention in this exclusive review is his
book on chaotic transport[35]. In this Wiggins takes the view that in many
dynamical systems the transient behaviour is as important, if not more so,
than the steady state behaviour. The text is very specialised and
consequently not casual reading. Quite a lot on Hamiltonian systems and
examples from fluid mechanics (sigh!).
Numerical Aspects
-----------------
The before mentioned books all suffer from the complete lack of any
description of numerical aspects of dynamical systems. Luckily, a few books
have appeared that deal with numerical methods for dynamical systems.
KubYcek and Marek[17] was one of the first texts to consider numerical
bifurcation analysis. Their treatment is not exactly modern, but quite
detailed and a large amount of Fortran code can be found in appendices. Many
of their examples are drawn from chemistry (surprise, surprise) and many
bifurcation diagrams are presented.
Parker and Chua[24] present a wealth of algorithms for the standard tasks
faced when carrying out a numerical study of a dynamical system. Among the
tools described are determination of invariant tori, continuation methods,
and methods for calculating stable and unstable manifolds. Pseudo-code is
included, so that programming in languages such as C, Fortran, C++ or Pascal
is relatively straightforward; however, beware, there are many errors in the
code (this may be corrected in the second edition; in the first edition even
the horseshoes are drawn incorrectly).
The second edition of Seydel[30] is a very careful and extremely detailed
text, to be recommended to anyone truly interested in the numerical analysis
of dynamical systems. It is aimed at bifurcation analysis, and does not
cover invariant manifolds etc. Following Parker and Chua, Seydel has
included a section on chaos theory. It not quite successful, and should have
been left out; just as Parker and Chua should have left out certain
superficially covered subjects. Nevertheless, the latter two books are
definitely handy to have within reach.
Concluding Remarks
------------------
A few questions naturally arise when attempting a review such as this: are
there a sufficient number of quality textbooks books around?; are there any
subjects that are under represented or neglected?; what is needed in future
textbooks? There are a number of excellent textbooks in dynamical systems,
several of the above mentioned texts fall into this category. The choice of
a particular one depends entirely on the purpose. Considering that the vast
majority of dynamical systems activity lies in practical applications, the
numerical analysis of dynamical systems is shamefully neglected in most
texts. This is a great pity, as there are interesting theoretical aspects to
cover here as well; strange as it may sound, treating numerics does not
necessarily imply thousands of lines of Fortran code. The final question
mentioned above cannot be answered in a simple way, as it depends on future
developments. It could, however, be the topic of an interesting discussion.
Related to this is the question of how to teach the subject, at
undergraduate, as well as postgraduate, level. UK Nonlinear News would be a
natural place to take up such a discussion.
Appendix: Contents of Books
---------------------------
The following table is intended to give the reader an overview of the
contents of the various books mentioned above. A plus indicates that the
subject is considered in the book. If a subject is mentioned in a footnote
or in an insultingly short section I have not included it.
The following abbreviations have been used:
GB: global bifurcations;
SD: symbolic dynamics;
NA: numerical analysis of dynamical systems;
HD: Hamiltonian dynamics;
CD: complex dynamical systems;
F: Fractals.
Contents of Reviewed Books
GB SD NA HD CD F
Lichtenberg & Lieberman + +
Guckenheimer & Holmes + + + +
Wiggins[33] + + + +
Wiggins[34] + + + +
Wiggins[35] + +
Ruelle +
Arrowsmith & Place + + +
Hale & KoIak + + +
Perko + +
Glendinning + +
Devaney + + +
Gulick + + +
Holmgren + +
Crownover + + +
GB SD NA HD CD F
Robinson + + + +
Thompson & Stewart + + + +
Moon + +
Schuster + + +
Ott + +
Marek & Schreiber + + + +
Bai-lin + + + +
Iooss & Joseph
Golubitsky & Schaeffer
Golubitsky & Stewart & Schaeffer
Kuznetsov + + +
KubYcek & Marek +
Parker & Chua + +
Seydel +
----------------------------------------------------------------------------
Appendix: Bibliography
1. R.H. Abraham and C.D. Shaw, Dynamics: The geometry of behavior,
Addison-Wesley, Redwood City, 1992.
2. D.K. Arrowsmith and C.M. Place, An introduction to dynamical systems,
Cambridge University Press, Cambridge, 1990.
3. H. Bai-lin, Elementary symbolic dynamics and chaos in dissipative
systems, World Scientific, Singapore, 1989.
4. L.S. Block and W.A. Coppel, Dynamics in one dimension, Lecture Notes in
Mathematics, vol. 1513, Springer-Verlag, Berlin, 1992.
5. P. Collet and J.P. Eckmann, Iterated maps on the interval as dynamical
systems, Birkhouser, Basel, 1980.
6. R.M. Crownover, Introduction to fractals and chaos, Jones and Bartlett
Publishers, Boston, 1995.
7. W. de Melo and S. van Strien, One-dimensional dynamics,
Springer-Verlag, Berlin, 1993.
8. R.L. Devaney, An introduction to chaotic dynamical systems,
Benjamin/Cummings, Menlo Park, 1985.
9. P. Glendinning, Stability, instability and chaos, Cambridge University
Press, Cambridge, 1994.
10. M. Golubitsky and D.G. Schaeffer, Singularities and groups in
bifurcation theory I, Springer-Verlag, New York, 1985.
11. M. Golubitsky, I. Stewart, and D.G. Schaeffer, Singularities and groups
in bifurcation theory II, Springer-Verlag, New York, 1988.
12. J. Guckenheimer and P. Holmes, Nonlinear oscillations, dynamical
systems and bifurcations of vector fields, Springer-Verlag, New York,
1983.
13. D. Gulick, Encounters with chaos, McGraw-Hill, New York, 1992.
14. J. Hale and K. KoIak, Dynamics and bifurcations, Springer-Verlag, New
York, 1992.
15. R.A. Holmgren, A first course in discrete dynamical systems,
Springer-Verlag, Berlin, 1994.
16. G. Iooss and D.D. Joseph, Elementary stability and bifurcation theory,
Springer-Verlag, 1980.
17. M. KubYcek and M. Marek, Computational methods in bifurcation theory
and dissipative structures, Springer-Verlag, Berlin, 1983.
18. Y.A. Kuznetsov, Elements of applied bifurcation theory,
Springer-Verlag, New York, 1995.
19. A.J. Lichtenberg and M.A. Lieberman, Regular and stochastic motion,
Springer-Verlag, New York, 1983.
20. M. Marek and I. Schreiber, Chaotic behaviour of deterministic
dissipative systems, Academia, Prague, 1991.
21. F.C. Moon, Chaotic vibrations, John Wiley & Sons, Chichester, 1987.
22. E.Ott, Chaos in dynamical systems, Cambridge University Press,
Cambridge, 1993.
23. J. Palis and F. Takens, Hyperbolicity and sensitive chaotic dynamics at
homoclinic bifurcations, Cambridge University Press, Cambridge, 1993.
24. T.S. Parker and L.O. Chua, Practical numerical algorithms for chaotic
systems, Springer-Verlag, New York, 1989.
25. H.O. Peitgen and P.H. Richter, The beauty of fractals, Springer-Verlag,
Berlin, 1986.
26. L.Perko, Differential equations and dynamical systems, Springer-Verlag,
New York, 1991.
27. C. Robinson, Dynamical systems, CRC Press, Boca Raton, 1995.
28. D. Ruelle, Elements of differentiable dynamics and bifurcation theory,
Academic Press, San Diego, 1989.
29. H.G. Schuster, Deterministic chaos, VCH Verlagsgesellschaft mbH,
Weinheim, 1989.
30. R. Seydel, Practical bifurcation and stability analysis,
Springer-Verlag, New York, 1994.
31. C. Sparrow, The Lorenz equations: Bifurcations, chaos, and strange
attractors, Springer-Verlag, New York, 1982.
32. J.M.T. Thompson and H.B. Stewart, Nonlinear dynamics and chaos, John
Wiley & Sons, Chichester, 1986.
33. S. Wiggins, Global bifurcations and chaos, Springer-Verlag, New York,
1988.
34. S. Wiggins, Introduction to applied nonlinear dynamical systems and
chaos, Springer-Verlag, New York, 1990.
35. S. Wiggins, Chaotic transport in dynamical systems, Springer-Verlag,
New York, 1992.
============================================================================