ROBIN: Reference Observatory of Basins for INternational hydrological climate change detection.

Global warming, associated with the burning of fossil fuels, is changing the world's climate, and with this, it is altering the water cycle. Future climate projections suggest hydrological extremes (floods and droughts) will become more frequent and severe - further heightening the already substantial impacts they cause to lives and livelihoods, as well as infrastructure and economies.

To adapt to future changes in water availability, we need projections of future flood and drought occurrence. Numerical simulation models are used to provide such scenarios, but they are very complex and highly uncertain. To better understand and constrain these model-based projections, we need to quantify emerging trends in the water cycle. This requires long records of past hydrological observations. River flows (the volume of water flowing in rivers) are especially useful because river flows integrate climate processes over the large areas covered by drainage basins. River flows are also, in practice, one way in which climate change will most impact society and the environment: through devastating floods at one end of the spectrum to droughts at the other, causing water shortages for public supply, industry, irrigation and wildlife.

Across the world, there have been many studies of long-term trends in river flow. Despite this past research, however, our confidence in observed trends remains very low - even in the major state-of-the-art IPCC reports, which have typically been cautious in their reporting of floods and droughts. The key reason is that most rivers are heavily modified by human disturbances (e.g. dams, large removals of water for irrigation, domestic or industrial consumption). These disturbances can obscure the 'signal' of climate change - that is, trends in many rivers may bear no relation to global warming and may in fact be opposing the climate trend, due to human modifications such as dam construction.

To detect climate-driven trends we need to analyse river basins that are relatively undisturbed by such human impacts. Recognizing this, some countries have declared 'Reference Hydrometric Networks' (RHNs) of locations where river flows are measured, and where human impacts are absent or minimal. However, to date there have been no efforts to integrate these globally. This is a problem for global assessments like the IPCC, as countries use different methods to assess trends, which limits comparison. Members of our consortium have previously pioneered a first trans-Atlantic study in this field. With the ROBIN initiative, we are now advancing a truly worldwide effort to bring together a global RHN.

As well as the network of river basins, ROBIN is the network of researchers and institutions sharing expertise. The network includes leading experts from Brazil, Chile, Malawi, South Africa, India, Thailand, New Zealand and Australia, augmenting our existing network across Europe and North America. Crucially, these new countries span a broad range of different climates and the partners also bring specific expertise (for example, unique knowledge of global datasets that can support ROBIN, or specialist analysis of 'ephemeral' rivers that often run dry).

ROBIN will engage other countries to expand over the lifetime of the project and set out a pathway to a sustainable legacy for the network going into the future. In this regard, crucially, ROBIN is supported by international organisations (UNESCO, WMO and the IPCC) who will ensure sustainability following the two-year project. ROBIN will also deliver the first truly global scale analysis of trends in river flows using undisturbed catchments. This will be a novel, high impact analysis in its own right, but will also showcase the potential of the network. Taken together, these activities will help realise the vision of ROBIN that future IPCC assessments will make more confident appraisals of climate change impacts on the water cycle, including floods and droughts.