Page 37 - Fundamentals of Geomorphology
P. 37

20    INTRODUCING LANDFORMS AND LANDSCAPES


               Box 1.3

               THRESHOLDS

               A threshold separates different states of a system. It  change in an external variable. A prime example is the
               marks some kind of transition in the behaviour, oper-  response of a geomorphic system to climatic change.
               ation, or state of a system. Everyday examples abound.  Climate is the external variable. If, say, runoff were
               Water in a boiling kettle crosses a temperature thresh-  to increase beyond a critical level, then the geomor-
               old in changing from a liquid to a gas. Similarly, ice  phic system might suddenly respond by reorganizing
               taken out of a refrigerator and placed upon a table  itself into a new state. No change in an external vari-
               in a room with an air temperature of 10 C will melt  able is required for a geomorphic system to cross an
                                             ◦
               because a temperature threshold has been crossed. In  internal threshold. Rather, some chance fluctuation
               both examples, the huge differences in state – liquid  in an internal variable within a geomorphic system
               water to water vapour, and solid water to liquid water –  may take a system across an internal threshold and
               may result from tiny changes of temperature. Many  lead to its reorganization. This appears to happen
               geomorphic processes operate only after the crossing  in some river channels where an initial disturbance
               of a threshold. Landslides, for instance, require a criti-  by, say, overgrazing in the river catchment triggers
               cal slope angle, all other factors being constant, before  a complex response in the river channel: a compli-
               they occur. Stanley A. Schumm (1979) made a power-  cated pattern of erosion and deposition occurs with
               ful distinction between external and internal system  phases of alluviation and downcutting taking place
               thresholds. A geomorphic system will not cross an  concurrently in different parts of the channel system
               external threshold unless it is forced to do so by a  (see below).



              environmental changes or random internal fluctuations  change is seen as a simple response to an altered input.
              that cause the crossing of internal thresholds (Box 1.3),  It shows that landscape dynamics may involve abrupt
              a landscape will respond in a complex manner (Schumm  and discontinuous behaviour involving flips between
              1979). A stream, for instance, if it should be forced away  quasi-stable states as system thresholds are crossed.
              from a steady state, will adjust to the change. However,  The latest views on landscape stability (or lack of it)
              the nature of the adjustment may vary in different parts  come from the field of dynamic systems theory,
              of the stream and at different times. Douglas Creek in  which embraces the buzzwords complexity and chaos.
              western Colorado, USA, was subject to overgrazing dur-  The argument runs that steady states in the landscape
              ing the ‘cowboy era’ (Womack and Schumm 1977). It  may be rare because landscapes are inherently unsta-
              has been cutting into its channel bed since about 1882.  ble. This is because any process that reinforces itself
              The manner of incision has been complex, with discon-  keeps the system changing through a positive feed-
              tinuous episodes of downcutting interrupted by phases  back circuit and readily disrupts any balance obtain-
              of deposition, and with the erosion–deposition sequence  ing in a steady state. This idea is formalized as an
              varying from one cross-section to another. Trees have  ‘instability principle’, which recognizes that, in many
              been used to date terraces at several locations. The ter-  landscapes, accidental deviations from a ‘balanced’ con-
              races are unpaired (p. 236), which is not what would  dition tend to be self-reinforcing (Scheidegger 1983).
              be expected from a classic case of river incision, and  This explains why cirques tend to grow, sinkholes
              they are discontinuous in a downstream direction. This  increase in size, and longitudinal mountain valley profiles
              kind of study serves to dispel for ever the simplistic  become stepped. The intrinsic instability of landscapes
              cause-and-effect view of landscape evolution in which  is borne out by mathematical analyses that point to the
   32   33   34   35   36   37   38   39   40   41   42