F. Gallart, THE RELATIVE GEOMORPHIC WORK EFFECTED BY 4 PROCESSES IN RAINSTORMS - A CONCEPTUAL-APPROACH TO MAGNITUDE AND FREQUENCY, Catena, 25(1-4), 1995, pp. 353-364
According to Wolman and Miller (1960) the high geomorphic power of ext
reme events is compensated by its low frequency, because the product o
f a power function (cause-effect relationship) by a negative exponenti
al (frequency distribution of the causes) tends to zero for high value
s. This result is adequate for extreme values, but the difference betw
een the most frequent (modal) events and those which produce the maxim
um of relative work depends on: (1) the exponent of the power function
; and (2) the dispersion (standard deviation) of the frequency distrib
ution. On the other hand, other kind of functions describing the cause
-effect relationship, when multiplied by the frequency distribution ca
n yield a maximum for trigger values whose frequency is very low. Most
of the recent data about processes triggered by extreme rains in moun
tains show the chief role of rapid mass movements, the number of which
increases very quickly with increasing rainfall. A general model whic
h describes the increase of landsliding as a function of event rainfal
l depth is here presented. It is able to describe the differences amon
g diverse climatic and geomorphic scenarios, fits very well to two dat
a sets, allows congruent extrapolations for higher rainfall values, an
d overcomes the problem of thresholds. Unfortunately, in its present f
orm, it is unable to handle different rainfall intensities. A four hil
lslope process model built up with the same assumptions of the model f
rom Wolman and Miller shows that, after a maximum of work made by chem
ical weathering, creep and wash, a new and higher maximum can appear f
or infrequent slide-triggering rains if they are not too rare and the
slopes are steep enough. This model, though in a probabilistic-concept
ual frame, is able to simulate most of the main characteristics of the
magnitude-frequency relationships described elsewhere, tolerates a wi
de range of process combinations, and admits the analysis of the conse
quence of a change of the frequency of extreme rainfall events.