Geography Reference
In-Depth Information
The utility of the individual observations gathered by
means of the flash flood survey may be extended using
hydrological models driven by the space
inferences from rainfall
runoff data), and to calibrate or
validate models developed elsewhere.
-
time estimates of
rainfall obtained from radar reanalysis (when available).
Ruiz-Villanueva et al.(
2011
) integrated the surveying and
modelling phases through a three-step procedure (applied to
a medium-size catchment in south-west Germany), which
reflects the hierarchy of data use considered in this chapter:
(i) A-priori modelling of peak runoff at multiple locations
based only on land use/land cover data, soil properties, soil
thickness and radar rainfall data. (ii) Calibration of the
model using runoff data from a (distant) downstream stream
gauge, which include the whole area impacted by the flood.
(iii) Comparison with the runoff observations collected
from the post-flood survey and identification of the critical
areas/processes responsible for outlying responses. The
methodology based on post-flood survey affords examin-
ation of key hypotheses concerning the hydrology and
hydraulics of catchment response under flash-flood condi-
tions. Examples include (i) role of antecedent soil moisture
conditions in flood magnitude; (ii) role of land use and
catchment properties in runoff generation; and (iii) depend-
ence of flood properties on basin scale by means of space
-
However, data are more than just inputs to a model. Data
have hydrological context and contain hydrological con-
tent. Data will be the ultimate source of the understand-
ing that is embedded in all models, because when
understood properly they reflect the co-evolution that
is common to all catchments. Therefore, there is value
and much to be learned from the combination of runoff,
climate and catchment data, a learning process that we
have called
'
reading the landscape
'
.
The information content of data products required for
accurate PUB, from global data sets to local observa-
tions, is highly scale dependent and increases with
decreasing temporal and spatial scale of prediction. This
is because system heterogeneity is increasingly sub-
sumed at larger spatial and temporal scales, leading to
simpler catchment response to climate forcing. On the
other hand, at small time and space scales, the hetero-
geneities and process complexities are much stronger,
and are not attenuated, and thus need considerably more
data to resolve them.
-
The scale dependency of data requirements therefore
necessitates a hierarchical strategy of data acquisition.
Given the constraints provided by available resources
and time, different data acquisition strategies may be
adopted at various levels. Global and low resolution data
sets, generally based on remote sensing, provide gener-
alised information at low cost. Regional data sources of
varying availability and accuracy provide detailed infor-
mation at higher cost over smaller scales. Finally, with
increasing time and financial resources, organisation and
collection of short-term measurements may provide a
better understanding of the catchment response at local
scale.
time scaling properties of precipitation.
Surveys of flash-flood response may provide valuable
insights; however, generalising the findings beyond the
areas of interest can prove to be difficult. Each storm
episode seems to have particularities that cannot be speci-
fied in full detail. Advancing the understanding in the
context of flash-flood studies, which are by necessity
opportunistic and event-based, requires the development
of a parsimonious avenue to synthesis. This may be based
on classification and similarity concepts, which can be
profitably used when the processes are not fully under-
stood (Blöschl,
2006
). Contrasting different case studies
and learning from the similarities and dissimilarities should
play a central role in PUB studies.
Large or regional data sets, even of low resolution, are
an important basis for performing comparative hydrol-
ogy, to generate a-priori expectations of dominant pro-
cesses, while very detailed data on the local scale help to
confirm and improve process understanding. Extrapo-
lation from gauged to ungauged basins requires that one
finds connections between what happens locally and
elsewhere: this requires a framework to connect.
3.8 Summary of key points
Predictions in ungauged basins (PUB) in one way or
another involves extrapolation from gauged to ungauged
catchments, which needs data of all kinds. Three kinds
of data will be discussed in this topic: runoff data,
climate data and catchment data.
In general two kinds of data can be distinguished: hard
data measured in the field and soft/proxy data that
provide additional information on hydrological systems.
For PUB the combination of available soft and hard data
relating to runoff, climate and catchment, through the
reading of the landscape, plays an important role in
order to exploit the available information and describe
runoff processes in the best possible way.
The need for data can be summarised under three cat-
egories: (i) data needed to read and understand the
landscape in a hydrological context; (ii) data needed to
develop regression relationships that will be used in
statistical models; and (iii) data needed for process-
based models, such as climatic forcing and parameter
values, data to assist with model development (to make
Field reconnaissance and expert judgement play critical
roles in the assessment of local system characteristics
Search WWH ::
Custom Search