SelfOrganising Systems in Geomorphology

1
Self-Organising Systems in Geomorphology

• A Fixed-Term Working Group
• of the British Society for Geomorphology
• Co-Convenors Dave Favis-Mortlock and Brian
  Whalley
• (Queen's Belfast)

2
This session

• Introduction (Brian)
• Self-organising systems? (Dave)
• The SOS Fixed-Term Working Group (both)
• Questions/discussion

3
(No Transcript)
4
This session

• Introduction (Brian)
• Self-organising systems? (Dave)
• Background
• SOS in geomorphology
• A personal view
• The SOS Fixed-Term Working Group (both)
• Questions/discussion

5
Self-Organising Systems Background 1

• Self-organisation is a process in which the
  internal organisation of a system, normally an
  open system, increases in complexity without
  being guided or managed by an outside source.
  Self-organising systems typically (though not
  always) display emergent properties.
• From http//en.wikipedia.org/wiki/Self-organisatio
  n
• In other words, the dynamics of a system can, by
  themselves, tend to increase the inherent order
  of a system.

6
Self-Organising Systems Background 2

• Self-organising probably first introduced in
  1947 by cyberneticist W. Ross Ashby.
• Self-organisation associated with general
  systems theory in the 1960s.
• Term only became commonplace following adoption
  by (mainly) physicists, working on complex
  systems, in 1970s and 1980s.
• Still somewhat controversial re. definition,
  limits etc.
• See recent/ongoing discussion on
  complex-science_at_necsi.org

7
Self-Organising Systems Background 3

• Most rigorous work on self-organisation to date
  by Ilya Prigogine, on dissipative structures in
  far-from-equilibrium thermodynamic systems far
  from equilibrium (won him 1977 Nobel Prize in
  Chemistry).
• Thermodynamically open and closed systems
• Closed systems become simpler and more disordered
  over time they go to thermodynamic equilibrium
  (cf. Davisian peneplains).
• Thermodynamically open systems however are
  capable of maintaining or increasing their
  internal organization, whilst a flow of
  matter/energy is maintained through the system
  (e.g. living systems).

8
Self-Organising Systems in Geomorphology 1

• Many geomorphological systems may similarly be
  seen as thermodynamically open systems, which
  increase their internal organization as long as a
  flow of matter/energy is maintained through the
  system.
• Distinguish between more complex SOS systems
  which learn (i.e. which store information for
  subsequent re-use, modifying subsequent initial
  boundary conditions, in a form such as DNA or
  writing) and less complex SOS systems (e.g.
  geomorphological systems) which do not.
• OK, but is this notion of any use to us, as
  geomorphologists?

9
Self-Organising Systems in Geomorphology 2

• The components of a self-organising system may be
  (but are not necessarily) relatively simple.
• Local interactions between these components,
  re-enacted a large number of times, give rise to
  more complex (and often qualitatively different)
  responses from the whole system (i.e.
  globally).
• Scale-crossing
• Such higher-level responses are termed
  emergent.
• Emergent behaviour cannot be simply inferred
  from the behaviour of the components

10
Self-Organising Systems in Geomorphology 3

• The search for simplicity...
• Interactions between the components of a complex
  system need not themselves be complicated to
  produce a complex, whole-system, emergent
  response.
• Could it be that the complexities of (some)
  geomorphological systems are determined by simple
  underlying rules?

11
(No Transcript)
12
SOS in Geomorphology a personal view 1

• A simple CA model was constructed (RillGrow 2),
  based on a SOS view of soil erosion by water on a
  hillslope.
• All flow and sediment transport relationships are
  applied locally (at the scale of microrelief)
• The only spatially explicit input parameter is
  microtopography.
• This model is able to realistically predict the
  initiation and spatial development of rills.

13
RillGrow 2 validation 1
X11 (calibrated, 15) experiment on left,
simulation on right
X14 (15) experiment on left, simulation on right
14
RillGrow 2 validation 2
X13 (10) experiment on left, simulation on right
X16 (10) experiment on left, simulation on right
15
RillGrow 2 validation 3
X15 (5) experiment on left, simulation on right
X12 (5) experiment on left, simulation on right
16
X13 (10) experiment on left, simulation on right
17
SOS in Geomorphology a personal view 2

• It is rather easy to generate complex behaviour
  by starting from simple initial conditions and
  then following simple sets of rules. But the
  point is that if one starts from some particular
  piece of behaviour there are in general no such
  simple rules that allow one to go backwards and
  find our how this behaviour can be produced.
• p551 of Wolfram, S. (2002). A New Kind of
  Science, Wolfram Media Inc., Champaign, IL, USA.
  1197 pp.

18
(No Transcript)
19
SOS in Geomorphology a personal view 3

• Thus the problem is one of model equifinality
  (cf. Keith Beven) there may be many sets of
  underlying rules which would produce similar
  results (i.e. rill patterns etc.), if
  incorporated into a model such as RillGrow.
• How do I know that the rules I have used even
  come close to saying anything about reality?
• Test the model again... and again... and again...

20
SOS in Geomorphology a personal view 4

• Have gained simplicity here, by abandoning the
  idea of working at ONE particular spatial scale.
• The cells of a CA (and the rules which govern
  their interactions) represent one spatial scale
  the rills which form (emergent structures) are at
  a LARGER scale.
• Similarly for temporal scale (Courant condition)
  the rules which govern the interactions between
  cells operate at one timescale, but development
  of the the rills (the emergent structures) is
  slower.
• To study self-organising geomorphological
  systems, do we need to re-think the scales at
  which we look at processes/patterns in the
  field/laboratory???

21
This session

• Introduction (Brian)
• Self-organising systems? (Dave)
• The SOS Fixed-Term Working Group (both)
• Meetings
• Website
• Publications
• Questions/discussion

22
The SOS Fixed-Term WG 1

• Meetings
• June 2006 Initial 'welcome' meeting at the
  Loughborough ESS Meeting
• Autumn/Winter 2006 Meeting/workshop
• April 2007 Special topic session at EGU Meeting
• Autumn/Winter 2007 Meeting/workshop
• Summer 2008 Final meeting/workshop
• At some, will be invited guest speakers (both
  geomorphologists and non-geomorphologists)
• E.g. Jonathan Phillips has already expressed an
  interest

23
The SOS Fixed-Term WG 2

• Website at http//www.qub.ac.uk/geog/sos/
• Summer 2006 Establishment of on-line discussion
  group (for all BGRG members and perhaps those
  outside the BGRG) this would allow discussion
  without attendance at meetings
• Additionally, we would look to produce on-line
  papers for discussion (perhaps 'closed' in the
  first instance) as we progress.

24
The SOS Fixed-Term WG 3

• Publications
• For the end of the FTWG period, we envisage an
  edited book (Wiley have already indicated an
  interest in such a volume) which would present
  the state of the art in this topic, and be of
  international interest.
• Special issues of journal(s)

25
This session

• Introduction (Brian)
• Self-organising systems? (Dave)
• The SOS Fixed-Term Working Group (both)
• Questions/discussion

26
BSG SOS FTWG Questions/discussion

• To express interest in this working group, please
  send an email to
• d.favis-mortlock_at_qub.ac.uk
• b.whalley_at_qub.ac.uk
• Over to you!