A data acquisition framework for runoff prediction in ungauged basins - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 50

Figure 3.20. (a) Example of a flood mark. The vegetation removed from the rocky bank and the moss drenched with silt on the downstream side
of the tree show the highest level reached by floodwater (red line). (b) The arrow shows the tree with the flood mark and a phase of the
topographic survey of the river section. (c) Surveying the stream bed. Photos: M. Borga.

logistical planning and properly staffed infrastructure is
ensured, post-event surveys may deliver a spatially consist-
ent analysis of the historical flood response. Surveying the
geomorphic response, through mapping of landslide/debris
flow initiation and deposition areas, is important as well.
This may help to properly identify the flow processes that
occurred in the basin and hence to avoid questionable peak
runoff estimates due to incorrect identification and docu-
mentation of debris flows.
An example of a map of unit peak runoff values
obtained during a post-flood survey is shown in Figure 3.21
for an extreme flash flood that occurred in September 2007
in Slovenia (Zanon et al., 2010 ). Examination of the flood
response shows that the extent and the position of the karst
terrain provide major geological controls on the runoff
response in the region during storms. Differences in geol-
ogy, combined with the orographic and climatic influ-
ences, led to pronounced contrasts in flood response
between nearby basins, with the major flooding occurring
in an area outside the region that received the largest
rainfall intensities and accumulations.
Eyewitness accounts and observations represent an inte-
gral part of the flash flood response survey. It should be
noted that these

Figure 3.21. Map of the Sel š ka Sora catchment in Slovenia with
location of runoff estimates, interviews with eyewitnesses and central
values of unit peak runoff. From Zanon et al.( 2010 ).

might be collected as digital
imagery from movies and pictures. Such observations rep-
resent an extremely important information source to refine
the assessment of flow type/depth, the estimates of flow
velocity and runoff, and for the evaluation of flooding
extent. Interviews with eyewitnesses provide information
and anecdotal evidence on the time sequence and dynamics
of the flood, and as such they add a time dimension to the
spatial patterns of a flash flood response. It should be
recognised that accuracy of the witnesses

' observations '

PUB and for flood events characterised by sharp gradients
in runoff response properties (such as flash floods). Indirect
methods for flood peak estimation include the slope

-

area,
contracted opening, flow-over-dam or flow-through-culvert
approaches. However, any survey has to capture not only
the maxima of peak runoff; less intense responses within the
flood-impacted region are important as well. These lesser
events can be contrasted with the corresponding generating
rainfall intensities and depths obtained by weather radar
reanalysis, thus permitting identification of the catchment
properties controlling the rate-limiting processes (Zanon
et al., 2010 ). Clearly, not all sections of a river may be
suitable for indirect peak runoff estimation. However,
Borga et al.( 2008 ) have shown that, provided that a careful

accounts is
limited (up to ±15 min, according to Borga et al., 2008 ).
Consequently, when these observations are used to esti-
mate the timing of the flood peaks, their information con-
tent should be related to the catchment response time, and
therefore to the catchment scale.

'