Introduction - Runoff Prediction in Ungauged Basins: Synthesis across Processes, Places and Scales - page 6

Figure 1.4. Guiding principles of
the PUB Science Plan, SSG (2003),
p. 47.

Guiding Principles Behind PUB Science and Implementation Plan
A number of key principles naturally arise from the broad community objectives presented before,
and have guided the development of the science and implementation plan of PUB. Special gratitude
to Dunne (1998) for the inspiration it provided.

In view of its societal obligations, PUB will focus on real hydrologic phenomena, such as floods,
droughts, eutrophication of receiving waters, degradation of natural ecosystems, effects of
climatic variability and change and/or land use changes etc. (after Dunne, 1998*);
In being essentially a science initiative, PUB will mainly seek to advance fundamental knowledge
of hydrological processes, even to the extent of going beyond the immediate problem solving
needs or the community interest of the present time (after Dunne, 1998*);
PUB will constantly focus attention on what is not known, indeed it will gain energy from its own
uncertainties, emphasizing the need for empirical exploration and explicit attempts to validate or
falsify new ideas (after Dunne, 1998*);
PUB emphasizes learning from data from selected basins in different biomes or hydro-climatic
regions, demonstrating the value of data and the need for future data requirements, and should not
be seen as an alternative to data collection;
PUB will be obligatorily be self-critical, and will include within it a strong element focused on
continual assessment of its own progress, with the predictive uncertainty being used as the
measure of progress (after Dunne, 1998*);
PUB is necessarily integrative, will avoid and indeed overcome the fragmentation of approaches
that has bedeviled hydrology in the past, and rather will seek convergence of a variety of
approaches towards common objectives, also profiting from lateral perspectives into ancillary
sciences (after Dunne, 1998*);
Develop a hydrological prediction system that is capable of assessing the errors or uncertainty in
model predictions, quantifying the different sources of the uncertainty - parameter estimates,
climatic inputs and model structure - and constraining these uncertainties by making the best
use of the information available from other sites and of measurement programs implemented at the
site of interest.

∗

Dunne, T. (1998). Wolman Lecture: Hydrologic science .... in landscapes ....on a planet ... in the

future. In: Hydrologic Science: Taking Stock and Looking Ahead , National Academy Press,

Washington, D.C., 138p.

Work on this topic also developed as a community
effort and reflects all of the principles that have under-
pinned the PUB initiative ( Figure 1.4 ). The topic itself is
an outcome of a strongly felt need to synthesise the state
of the art of prediction in ungauged basins, and to carry
out a comparative performance assessment of a range of
prediction methods being used for the various runoff
signatures. Since the topic is focused on a synthesis of
current prediction methods, it cannot possibly do justice
to the enormous contributions of the range of activities
that have been carried out under each of the six PUB
themes. However, while the topic is a contribution to
PUB in its own right, its overall organisation has been
inspired by the concepts and clarity of thinking engen-
dered by the PUB initiative. In particular, the six PUB
themes are reflected in the topic in a cross-cutting way,
and the outcomes reflect the progress that has been
achieved over the past 10 years towards improved predic-
tions in ungauged basins.

1.5 What this topic aims to achieve: synthesis
across processes, places and scales
This topic is specifically devoted to predicting runoff in
ungauged basins, i.e., at those locations where no runoff
data are available. It will assess, on a comprehensive,
objective, open and transparent basis, the state of hydro-
logical predictions in the absence of data, and identify what
are the prediction challenges of the future.
It will accomplish this assessment through a synthesis
across processes, places and scales, as a response to the
challenge of fragmentation in catchment hydrology. In this
way it will strive to bring together research on predictions
of runoff in ungauged basins that has so far been disparate.
One of the goals of the proposed synthesis is to bring order
to what otherwise looks like disorder, to identify connec-
tions where none existed, and in this way generate new
ideas and novel approaches to advance the science of
hydrology, and improve the practice of hydrological