Influences on the Transfer of Hydrological Research into Practice

Research funding

Hydrological research can be pure or applied, and it is sometimes difficult to link the development of practical procedures to a specific research endeavour. Although it is not always the case, there is usually a substantial amount of investment behind a good practical hydrological technique. Countries with relatively buoyant economies are often the only ones that can afford the “luxury” of sustaining sizeable long-term investment in research. Research also offers the benefit of training and developing personnel and can encourage scientific liaison between communities and nations.  

Hydrological research programmes operate in public sector organizations, universities and the private sector. International organizations and frameworks have a potential to extend the country base of hydrological innovation. The European Flood Awareness System (see Box A), which is now reaching a degree of maturity in practice, is an example of one such initiative.   

Political drive

Political motivation to prioritize hydrological research spurs the development of practical techniques and methods. Understandably, there is a tendency for such a drive to arise in the aftermath of disasters, particularly when severe flooding or drought conditions affect public water supply, agriculture and industry.  Responses to disasters sometimes favour the implementation of hydrological schemes rather than the development of new or updated methods. For example, the development and implementation of the River Thames flood barrier scheme in London was prompted by the exceptionally severe river and coastal flooding of 1953, which also provided the impetus to improve the methodology of inland flood risk mapping in the United Kingdom. Policy-makers’ recognition of long-term climate consequences, as well as the short-term weather-induced hydrological hazards, provides a basis for acknowledging the need for sustained longer-term research and development.

The ability to properly address the strategic time scale can help to alleviate problems in the shorter operational time frame. In the river flooding context, for example, good long-term flood frequency estimation carried through to scheme design can ease, to a degree, complete dependence on flood warning systems. Strategic initiatives such as the WMO-led Global Framework on Climate Services (GFCS) and the Sendai Framework for Disaster Risk Reduction[1]may boost the development or enhancement of hydrological techniques, should the necessary funding be available to meet aspirational and ambitious aims.

These political drivers are significant aspects of demand-led methodological development, as opposed to ideas-led research. Political motivation can also be a factor in promoting use of a hydrological technique beyond its country of origin, whether for reasons of influence, commercial interest or humanitarian concern.  For example, WMO and aid organizations in the US established the Southern African Region Flash Flood Guidance System in order to reduce hydrometeorological hazards. This initiative was subsequently expanded to encompass regional flood forecasting and warning systems.

European Flood Awareness System (EFAS) fully operational from 2012 Flash flood alerts in October 2015 ©EFAS

European Union (EU) Joint Research Centre and others Purpose and method: Increased readiness for river floods and consistent flood warning information trans-nationally in Europe.  Use of ensemble precipitation forecasting from ECMWF (European Centre for Medium-Range Weather Forecasts) combined with LISFLOOD hybrid conceptual/physical runoff model linked with river channel routing. Development funding:  European Union DGs and Parliament; country-funded experts; testing via National Hydrological Services. Factors which promoted development:  Recognition at EU level that previous flood alerts were not compatible between countries, and were of variable quality, leading to difficulties in planning and in aid organization. Particular prompt from major Elbe and Danube flooding of 2002.  Uptake by EU as a flagship initiative in trans-national disaster mitigation. Factors affecting transferability:  Broad-scale coverage is built into system; not necessarily appropriate at smaller spatial scales.  Principles transferable at broad scale if meteorological model outputs and data transfer mechanisms are supported and hydrological model can be calibrated.

Generality of methods

Research outputs with the potential for more comprehensive application across varied environments stand a greater chance of widespread acceptance, other things being equal.  Highly robust methods with low-uncertainty outputs are clearly advantageous.   Both research and commercial considerations can mean that limitations in application may not always be specified, but the qualification of the expected range of applicability of a method should be presented in the interest of appropriate and responsible practical application. The nature of simplifying assumptions made in theoretical and numerical methods influences how generally applicable methods are in practice.

The development of hydrological methods and models is often intended for widespread use but methods may also include some implicit local, regional or national features.  These may relate directly to the hydrological environment itself or, for example, to the state of hydrological data collection and availability within a country or region. 

Calculation and modelling systems that are nominally general often require the establishment of locally-applicable parameter values. The Flood Estimation Handbook (see Box B) describes a river flood frequency estimation system with parameterized empirical equations that cover the range of British environments. Conceptually it is more widely applicable, but it has undergone little testing or re-parameterization beyond its region of design.

The level of transferability across spatial and temporal scales also affects how widely a technique can be used.  Different levels of detail are invoked at different spatial scales – something for which high computing power is not necessarily a substitute (even where data are not a constraint).  It is especially difficult to provide robust methodologies for extreme events with large recurrence intervals as they are rarely experienced and even more rarely measured.

Critical mass of uptake

A critical mass of users of a technique is a determining factor in its survival over an array of similar ones.  In large public and private sector organizations, techniques considered of reasonable generality and quality are frequently adopted as standards because of their compatibility and perceived efficiency. This adoption in turn enhances, at least for a period of time, the dominance and longevity of that technique.

For example, an Australian government advisory body recommended a method for modelling low river flows and cessations of flow (see Box C), thus these procedures are widely implemented nationally. Similarly in England and Wales, the environmental regulator uses a process-related indicator system method as a standard basis for determining the potential availability of water abstraction in its countrywide water licensing strategy. The latter case is an example of a method addressing a legal requirement, although the legislation did not specify the precise technique. Longevity and prevalence of use is usually achieved in such circumstances.

Hydrological modelling methods that are used as land boundary conditions in atmospheric modelling systems also gain extensive usage.

Accompanying material

Supplementary guidance material – whether produced by the originators of the method, by users or by organizations – encourages the adoption of a particular method.  The provision of training material plays a significant role in the dissemination and practical uptake of new methods and also in their adoption into hydrological educational programmes. Increasing access to web-based material has significantly expanded the geographical reach of information sources and of distance-learning programmes.     

WMO plays a key role in international dissemination and capacity building activities in hydrology. Key examples of hydrological training and reference material from WMO, which are extensively used, include the Guide to Hydrological Practices, the Manual on Flood Forecasting and Warning and the Manual on Low-flow Estimation and Prediction. The emphasis is, appropriately, more on the types of approach to hydrological problems rather than on specific methods and products.

Flood Estimation Handbook (FEH) 1999 and supplements Wallingford 2003 ©RAF Benson

Example application: Upstream input to river hydraulic modelling for country-wide flood risk assessment.  Institute of Hydrology, Wallingford, UK Purpose and method: Estimation of rainfalls and river flood peak discharges for range of recurrence intervals at locations with little or no site-specific data.  Statistical and runoff hydrograph methods based on pooled data. Development funding: England & Wales government Ministry of Agriculture, Fisheries and Food (subsequently Department for Environment, Food and Rural Affairs); supplemented by UK Natural Environment Research Council. Factors which promoted development:  Recognized need at government department level; strong user community; uptake by environmental regulators (Environment Agency, Scottish Environment Protection Agency); evolving recognition that this is arguably the pre-eminent British tool for flood frequency estimation for data-sparse locations.  Designed as follow-on to successful Flood Studies Report of 1975, itself spurred by severe 1953 flooding. Factors affecting transferability:  In principle transferable to other humid temperate regimes for catchments between about 0.5 and 1 000 km2; relationships between hydrological variables and catchment properties are not widely tested beyond UK and require data-rich sources.

Software

Hydrological methods presented as software products offer further advantages for promoting widespread use, especially for developers who produce user-friendly systems and make their software freely or very inexpensively available.  Another considerable advantage is gained by also making source codes available, which enables linkages with other techniques and allows users to make modifications. 

The U.S. Geological Survey’s ‘Modflow’ groundwater modelling system is an example of software that has been adopted by large communities of practitioners, not least because its source code is freely available. 

It is critical to good practice that assumptions are made fully explicit in guidance material and training. Some methods do not make key assumptions apparent when results are viewed, hence the realm of applicability may not be obvious and may be inappropriately exceeded.

Review and feedback

Evaluation of the use of a hydrological technique provides valuable knowledge for future applications.  Full assessment includes information on limitations as well as successes although the former is not always a feature that developers and/or users readily share for commercial or career reasons.  Furthermore, researchers and practitioners do not always liaise fully. Many developers are not themselves practical users nor do they prioritize the dissemination of their work beyond their peer community. Equally, many practitioners are not involved with the minutiae of theoretical and methodological detail. 

Impartial evaluations of performance and practical potential of various methods are available. Some are user- or client-funded to meet a particular need while others, including those by WMO, are undertaken on a more extensive basis.  

Conclusion and recommendation

Among the many hydrological methods developed, widespread adoption and operational use of one method over another is favoured by:

• A sound technical basis,
• Robustness across environmental conditions, in space and time,
• A drive to move the technique beyond the research arena,
• The presence of promoters for its uptake,
• A certain ease-of-use,
• Good guidance and appropriate updating.

Some of these factors follow from conscious decisions and others are rather more nuanced and subject to a degree of chance choices and opportunities.  The process is not necessarily a straightforward deterministic one, nor is it based entirely on technical merit.

There is arguably a wider range of acceptable methodological approaches employed in hydrology compared with meteorology. There is general agreement in meteorology on the preferred type of atmospheric formulations that form the basis of many modelling and model-based weather and climate products.  This difference arises partly because of the great variation in material properties in the physical domains covered by hydrology, and partly from the range of hydrological issues addressed, including the variety of man-made interventions in, and modifications of, the hydrological cycle.  This can enhance the degree of competition between alternative methods in many hydrological applications. This contributes to debate as to whether a convergence to a small core of “standard” hydrological techniques should be encouraged or a proliferation of a wide range of alternative procedures is preferred.

Regardless of the technique used, it is highly recommended that the extent of applicability of the hydrological method developed and/or used is specified, and  that this assessment is updated as experience accrues through practical use. Most applied research projects and practical hydrological applications should comment on this in their reporting for the benefit of the wider hydrological community to improve future delivery of effective hydrology.

Australian Low Flow Estimation 2012

D. Barma and I. Varley, in consultation with others, for the National Water Commission Purpose and method: Collating best modelling practice nationally for the estimation of low river flows and cessation of flow for both regulated and unregulated rivers. Hydrological modelling, with empirical transposition to data-sparse regions. Development funding: National Water Commission, a then government advisory body. Example application: Recalibration of low flows, eg. for Daly catchment, Northern Territory with use of integrated surface and groundwater model. Factors which promoted development:  Government recognition of need for compatible prediction at national scale;  recognition of need for consistent  calibration procedures for modelling especially of lower flow regimes. Factors affecting transferability: Designed for wide application across range of Australian environments;   drawbacks can arise from data insufficiency, particularly with respect to rainfall distribution and man-made interventions e.g. irrigation offtakes.

Acknowledgements

Thanks are due to Bruce Stewart (formerly WMO Director of Climate and Water) for early discussion and to the editorial team at WMO.

Authors

Ann Calver is a research and consultant hydrologist and has held a number of WMO roles. anncalveracw-associates [dot] co [dot] uk (anncalver[at]acw-associates[dot]co[dot]uk)