|
dynamic vision
Go away and read about Resilience.
fractured vista
Yale's Fractal Geometry course notes cover a lot of ground. The Cellular Automata and Fractal Evolution section integrates fractals, evolutionary computation, 1/f noise and more within a cellular automata framework; topics I am keen to look further into later. Their Panorama of Fractals and Their Uses is a fantastic playground encompassing such disparate fields as EEG patterns, percolation, cathedrals, Jorge Luis Borges, internet traffic, history, etc! The paper Fractals in the Biological Sciences (Kenkel & Walker, 1996) is a broad review. A major theme is the use of fractal dimension as a quantitative measure of complexity. They note, "Wherever a chaotic process has shaped an environment, a fractal structure is left behind." Fractals are shown to be relevant in multi-species communities, populations and diseases; in their composition, movement, dispersal and long-term dynamics. Also in landscapes, habitats and ecotonal boundaries. At smaller scales, in soil structure, shoots and roots, even in tissues, cells, proteins and DNA. In other words, fractals can be usefully applied to all levels of the ecological heirarchy. But can't we just smooth out these details when modelling? Not necessarily; this is where chaos comes in. Such a seemingly harmless approximation can change the path of the system due to sensitive dependence on initial conditions. The authors give an example: "Turbulent regions (e.g. interfaces between warm and cold water) have high phytoplankton productivity due to increased contact with resources (nutrients and light), which in turn 'feeds' higher trophic levels. This cascade effect implies that spatial patterns at fine spatial scales determine patterns at broader scales." One promising area where this could apply is rainfall modelling. I don't want to wander into the big topic of ecosystem complexity, but the authors refer to a finding that "species coexistence increased as landscape fractal dimension increased." This clearly is associated with habitat complexity, which Krummel et al. (1987) found to be decreased by "anthropogenic activity". This parallels findings in physiology: based on studies of fractal heart rate, some researchers "defined aging as a progressive loss of complexity in the dynamics of all physiological systems." A. Mary Selvam has a grand theory about atmospheric flows, referred to colourfully with such terms as "Cantorian Fractal Spacetime, Quantum-like Chaos and Scale Relativity" / "Long-range spatiotemporal correlations (non-local connections)" / "a superstring theory" / "self-organized criticality" / etc... a cell-dynamical model is developed using the self-similar structures of everything from turbulence through clouds to climate. By the same author (the theory is apparently generally applicable) are papers on the human brain, circadian rhythms, C60 Penrose/Fullerine molecules, the distribution of primes, electric fields, roots of the Riemann zeta function, the dow jones index, and the drosophila genome. Wow! To round this out, fractal analysis is starting to be used to study social systems, even in linguistics. Fractal music (sometimes) sounds good, perhaps because it echoes nature; MIDI libraries here and here. Mysterious things lurk in the perpetually novel structure of fractals (e.g. pi). There is much left to be explored.
blood on god's hands
i was thinking... if god flooded the world, apart from being guilty of genocide, He/She would have condemned the remaining pairs of animals on Noah's ark to likely extinction. the natural genetic diversity of the populations would have been lost. since variation is a prerequisite for evolution, without it any change in the physical or ecological environment could be fatal -- there is no flexibility in the population. even in the short term, organisms with two versions (allelles) of genes generally do better than those with two copies of the same gene; this is known as hybrid vigour. this fate seems to have befallen cheetahs. all cheetahs today were descended from less than 7 individuals, a genetic bottleneck which probably occurred about 10,000 years ago at the end of the last ice age. they are highly inbred, which means their long-term survival is very fragile. read about it here. meanwhile humans continue to destroy their habitat. that particular loss of genetic diversity was probably originally caused by pressure from climate change (many other mammal species went extinct or were severly reduced), although act of god has not been ruled out. clearly the god(s) are crazy, and must be stopped.
hive phylogeny
eusociality is a type of altruism, which is not normally a stable strategy, so its evolution requires an explanation. there have been many factors suggested as being important in its emergence and maintenance. this paper by Klaus Jaffe reports on an agent-based simulation investigating this. he looked at the relative importance, as well as the interactions of: haplo-diploidy (changes relatedness and provides direct exposure of allelles in the haploid sex; see eusociality link), assortative mating (sexual selection--also increases relatedness, thus promoting 'kin selection' for altruism), and 'social economics' (evolutionary game theory--the advantages of group living to the individual). all are significant, but the economics/energetics turned out to be the most important in isolation. he ends with the statement, "[...] sociality increases the energetic efficiency in ants, in termites and in human settlements, supporting the results of the simulations presented here."
hive psychology
the distributed nature of these systems lends itself to agent-based modelling. one of the general frameworks available for this is starlogo, developed at the MIT media lab. they have simulated some aspects of ant, bee and termite behaviour (projects tab on the site). swarm is another framework; some applications are here. Alexis Drogoul's manta aims to find the necessary conditions for the emergence of various social behaviours in ants. such simulation is increasingly used in testing biological hypotheses. my insect is better than yours...
the blog begins... i envision this as occasional thoughts on, and experiments with, digital biology. in practice it will be mainly links to interesting things i come across (as most blogs are), and an occasional simulation applet. the other major theme here will be the open source movement (as well as open standards and open content). specifically projects related to complex system modelling, science in general, or other cool stuff. just to make it even less coherent, you may find chunks of atheism, conservationism and socialism. please contribute anything remotely relevant.
|
posted by floybix on Mon Aug 2nd, 2004 at 02:04:35 AM EAST
oops, long time no post. i'm not used to having a pet blog, and often forget to feed it.
the |
webmaster@neurofractal.org
OpenPGP public key |
© 2002 Felix Andrews |
|