Thomas,

Wow! your question has really made me think. I cannot answer it very fully, but I can make a start and possibly provoke further discussion as follows:

1) Most responses of landforms to stimuli are transient, it is only the exceptions that are not. In systems theory numerous transient responses collectively define steady state behaviour and I would also argue dynamic equilibrium. These building blocks of systems behaviour are thus simply the time-dependent averages of numerous transient responses;

2) Non-transient responses are very interesting representing major a environmental hiatus and generating landforms or modifications that will be retained in the landscape for significant periods e.g. a major landslide on a slope, a hurricane or tsunami impact on a coast. These types of event typically trigger alternations of system states within meta-stable equilibrium behaviour. Some geomorphologists refer to them as "formative events" and the paper below is highly informative and thought provoking:

Brunsden, D. (1996) Geomorphological events and landform change, Zeitschrift für Geomorphologie,NF40: 273–88.

3) In my specialism of coastal science a typical transient response would be beach profile change over the tidal cycle in response to changing wave conditions with the morphological response rarely being able to "catch-up" with changing wave conditions. Consequently, the concept of the equilibrium beach profile is a most useful, but also a contested concept;

4) So how to define a transient landform? I'm really starting to struggle here. Is it a landform created only by transient responses? probably not as my definition in point 1 suggests that steady state and dynamic equilibrium are the logical outcomes. I think that the key is persistence and stability in time and space. The problem is to define the appropriate temporal and spatial scales;

5) Finally, perhaps I can demonstrate how I dealt with an element of transience back in 1995 when I was writing a paper about littoral cells and sediment budgets. I found that I needed to formalise a classification scheme for littoral cell boundaries. Most were defined by headlands or inlets, but some transport boundaries occurred where there were changes in rates and/or directions of littoral sediment transport due to effects such as wave refraction and partial sheltering by headlands, islands, nearshore banks and tidal streams. The only physical evidence of their existence would be enigmatic zones of shoreline erosion (drift parting) or accretion (drift convergence) and these tended to migrate and disappear/reappear as wave conditions altered. I classified them as "transient boundaries" lacking in stability over a 20 year period. The 20 year criterion was an educated guess influenced by the timescales of coastal engineering as the work was originally intended to inform coastal managers. That paper is available below - check out pages 386 to 389:

https://www.researchgate.net/publication/269393093_Littoral_cell_definition_and_budgets_for_central_southern_England

6) So where does this leave us? are transient responses and transient landforms unrelated? are transient landforms the product of "unusual" behaviour/responses that go beyond the envelope of transient responses, but fall short of truly formative landscape events? Is my thinking constrained rather by my experience in applied coastal science and a concentration upon timescales of days to decades? It would be very interesting to learn of perspectives from specialists in other environments.

I credit my inspiration in these areas to past research advisers Professors Denys Brunsden and David Jones.

I hope that this answer is of some use and I would be most interested to hear what other researchers think about this topic - many thanks for raising it as a question.

Article Littoral cell definition and budgets for central Southern England

Dear Thomas, dear Malcolm,

And what if we use Malcolm's model so that there are triggering phenomena of various scale, duration and frequency, so that the "dynamic equilibrium" is a result (response?) to a certain set of triggers of given spatial scale and given spatial and temporal frequencies (not the same!)? Then the transience is "simply" the change in this set: either the scale or the frequency or the type (e.g.,  glaciation)? This way the decomposition would be possible, if we are able to separate effects (not always  possible).Does this make any sense? This would be then somewhat similar to the deterministic chaos theories. It would also allow the creation and survival of paleosurfaces and existence of "topographic memory".

Would you comment on that?

Kind regards, Balázs

Balazs and Thomas,

Thank you for your discussion. I think that transience occurs when a landform does not persist sufficiently to achieve steady state or dynamic equilibrium adjustment so your suggestion that glacial landforms are often transient is quite perceptive.

Having reflected further on the question asked I think the most interesting concept is the transient landform. A transient landform has been defined as one that has a duration shorter than the recurrence interval of the process that formed it. Conversely, a persistent landform lasts longer than the recurrence interval of the event that created or modified it (Wohl, 2010).

Wohl, E. (2010) 2nd ed. Analysing a natural system. In Clifford, N., French, S. and Valentine, G. (eds) Key Methods in Geography, London: Sage. pp. 253-256.

Related to these definitions is the potentially useful concept of the transient-form ratio of Brunsden and Thornes (1979) see p480. P.S. in my opinion this paper is so important that in its own right it constitutes a formative event in the study of geomorphology.

Brunsden, D., Thornes, J.B., 1979. Landscape sensitivity and change. Trans. Inst. Br. Geogr. NS 4, 463–484.

Some very practical applications of these concepts are set out by Brunsden (2003) in a paper concerning geomorphology, engineering and planning available at the link below.

So if Brunsden and his co-workers have explored these concepts so effectively what opportunities are there for further research? My estimation of potential areas is as follows:

1) Evolution of landforms to encompass variations in the degree of transience (spatial and temporal) and switching between transience and stability;

2) The role of human modifications and the status of human-created landforms. In my coastal science specialism I could identify stabilised cliffs, replenished and artificially controlled beaches, management to halt or accelerate vegetation succession, land reclamation in estuaries, construction of new islands e.g. Dubai; coastal impacts of dams on rivers.

I think that I had better stop there for now otherwise this topic can easily dominate my thinking and disrupt work schedules etc.

Best wishes

http://www.igipz.pan.pl/en/wydaw/GP_76_2.pdf#page=176

Hi Thomas,

I would suggest reading Philip Allen's short essay in Nature (attached) about transient or reactive and steady or buffered landscapes. He phrases the question in terms of the timescales of forcing and reaction. I find it quite enlightening.

Cheers,

Luca

Dear Luca,

Thanks for referring to Allen's essay. By all respect, IMHO geomorphologists should learn some signal processing theory, simple electronics (RLC circuits), solar-terrestrial physics (solar wind-Earth magnetosphere interactions), and then they would understand what an integrative member in an electric circuit does (Allen terms that, IMHO somewhat erroneously, buffered) to various types of transients. Just use electric signals of various frequencies as excitation (as for uplift/postglacial rebound, temporal precipitation pattern, flash floods, catastrophic events, etc.) to an electric circuit with condensers and coils and consider the resulting electronic signal as, e.g., sediment output of a catchment.  I have the impression that what is considered in this essay is nothing new to an electronic engineer. Response to excitation of some biological systems with appropriate feedback mechanisms are also very good analogues.

Kind regards, Balázs

Dear Balazs

Perhaps then, this is the research question - to translate the qualitative descriptions in Brunsden's work into a relevant set of coupled ODEs that produce first-order, quantitative models of the geomorphologcal concepts discussed in the 1970s. Hence, transience can then be quantified. Another useful reference on this is the work of J.R.L. Allen on relaxation times in landscapes.

All the  best

Chris

Brunsden and Thornes (1979) introduced ‘transient form ratio’ which is the ratio of relaxation time of a geomorphic event to its recurrence interval. Values equal or greater than one indicate a constantly changing system while values less than one represent some form of dynamic equilibrium. Confining scope of the discussion to fluvial geomorphology, if the ‘forcings’ responsible for geomorphic events are searched within as well as  beyond geomorphology and assumed that the recurrence frequencies of these events are having wide variations then observing plano-temporal changes of channels (by comparing topographic maps of different times as well as satellite imageries) and spatio-temporal changes of stratal geometry (from seismic sections across different transacts of river valleys) can provide quantitative behaviour of transience. We have worked in this regard in the upper reach of the Brahmaputra valley of Assam. The narrow valley squeezed between two prominent thrust belts show highly variable nature of bankline migration of the Brahmaputra River. We have argued that depending upon the proximity of a bankline with a thrust belt, the relative influence of thrust belts (one of the many ‘forcings’) can be differentiated by finding out the frequency distribution of ‘bankline migrations’ during different periods of time. This helps to understand a specific transient landscape. We have not however, carried out similar exercises for tectonically active river valleys in other regions.

(Ref. Lahiri and Sinha, 2015, Application of fast Fourier transform in fluvial dynamics in the upper Brahmaputra valley, Assam, Current Science 108(1), 90-95)

Dear all,

thanks very much to your contributions which I highly appreciated. From the discussion and the papers it seems that transience is given when the time interval between disturbing events is shorter than the relaxation time of the system, or the transient form ratio larger one. This is in agreement with Brunsden and Thornes (1979) and Allen (2005). However, in geology a transient river seems to be one which has not attained equilibrium in response to an tectonic uplift, very much like the work Siddhartha mentions. With respect to Brunsden and Thornes this is an adjusting system (or “buffered” in terms of Allen), maybe with a very long recovery time but this does not need to be transient (this the disturbance interval might take even longer). Does it make sense to differentiate these to cases? As long as the process regime does not change, I guess not! The system tries to achieve equilibrium irrespective whether is will achieve it or not.