Geoscience Reference
In-Depth Information
projections within the ensemble sample itself. This difference is important. Indeed, such are the known
scientific limitations of representing precipitation in climate models that these probabilities might have
little or no relevance to the policy response. To ignore those probabilistic estimates and deal with mag-
nitudes of change directly (without the need for climate simulations) precludes a complete risk-based
strategy but does place the focus directly on what is considered to be affordable in being precautionary.
A particular case in point is protection against flooding. If a changing climate is intensifying the
hydrological cycle, we expect the frequency of floods of a given magnitude to change (even if, given the
nature of extremes, this might be difficult to demonstrate from the available observations, e.g. Wolock
and Hornberger, 1991; Robson, 2002; Kundzewicz
et al.
, 2005; Wilby, 2006; Wilby
et al.
, 2008; Fowler
and Wilby, 2010). A number of studies have invoked the change factors produced by climate models to
examine how flow frequencies might change. This is straightforward to do if it can be assumed that the
parameters calibrated to represent catchment response might not change with changing inputs. It is much
more difficult to do if it is thought that the change in inputs or land use and management might require
that parameter sets be changed to represent new sets of conditions.
A number of strategies are possible so as not to exacerbate the flooding problem: avoiding new
developments on flood plains; improving flood defences; flood proofing of existing buildings; breaching
of existing defences to make more storage; building flood detention basins. In most cases, these solutions
are robust in the sense of not precluding future adaptive management strategies but they all have a
greater or lesser cost. So what is the cost-benefit of protecting against different levels of change. How
precautionary are we prepared to pay to be? This is, essentially, a political decision. The science comes
in estimating the costs and benefits of different policy options, which might be considered to be a far
more realistic goal than the accurate prediction of future change (Beven, 2011).
In addition, there are other factors that might affect future hydrological responses (societal change,
urbanisation, agricultural intensification, energy security, deforestation-afforestation, river training and
re-naturalisation, ...) that might be far more important on the decadal time scales over which we might
achieve some adaptive management strategy. There are certainly model-based scenario projections of
the effects of potential changes in different factors, mostly deterministic in nature (e.g Bronstert
et al.
,
2007; Viney
et al.
, 2009), even though we know that process representations of such factors are subject
to considerable uncertainty (and should really be embarrassed about some of the studies that have been
published that purport to assess the impacts of change using only deterministic model predictions). Such
changes could also be evaluated in the form of the precautionary cost-benefit strategy suggested above.
It is clear that more science and understanding is required to reduce the uncertainties in assessing the
impacts of change as an input to an informed and open policy framing debate. Decision makers require
“evidence” of how great the impacts of change might be; but it would be much more intellectually honest
to change the nature of the game into something more overtly political before the “evidence” comes
to be seen as based on insubstantial foundations. Future food and energy security might, for example,
provide far more politically compelling arguments for climate mitigation policies than uncertain climate
projections. We need better ways of deciding how precautionary to be in planning for the future.
8.10 Key Points from Chapter 8
•
There are two uses of rainfall-runoff modelling in flood prediction: one in forecasting discharges in real
time during flood periods, the other in predicting the frequencies of different flood peak magnitudes.
•
Real-time forecasts are very dependent on the accuracy of input data, particularly of spatial patterns
of rainfall intensities. The availability of radar rainfall data has greatly improved the potential for
forecasting flood peaks. Improvements in the lead time of forecasts are dependent on improvements
