Linking HEPEX and GEWEX: Synergies in Seasonal Hydrologic Prediction

Linking HEPEX and GEWEX: Synergies in Seasonal Hydrologic Prediction

Contributed by Andy Wood and Eric Wood

The overarching goal of HEPEX is “to demonstrate the added value of hydrological ensemble predictions (HEPS) for emergency management and water resources sectors to make decisions that have important consequences for economy, public health and safety.”  To this end, HEPEX focuses on cross-cutting methodological elements of ensemble forecast production – e.g., meteorological ensemble forecast downscaling and calibration, streamflow forecast post-processing, verification – that are as applicable to localized flash flood forecasts at lead times of minutes as to long range reservoir inflow predictions at lead times beyond 1 year.  The longer  seasonal to interannual (SI) time and space scales have been represented in HEPEX via efforts focusing on hydrologic predictands that are typically monthly or seasonal averages (rather than, say, daily or instantaneous flows, for which long range prediction skill is low).  An important SI flow forecast application in the US, for example, is the prediction of spring snowmelt runoff volumes for use in determining future allocations of water for irrigation, municipal supply, recreation, drought mitigation, and other purposes.

Until the 1990s, the traditional techniques for such predictions were not model or ensemble-based but statistical – basically, the regression of future runoff on in situ measurements of rainfall, streamflow and snowpack as predictors. Within HEPEX, the statistical forecasting tradition is well represented by participants such as Dr. QJ Wang from CSIRO, who has developed powerful predictor identification and combination techniques for implementation by the operational Bureau of Meteorology, in Australia.  During the last decade, hydrological model-based ensemble techniques such as Ensemble Streamflow Prediction (ESP) in the US and Hydrological Ensemble Streamflow Prediction Systems (HEPS) in Europe have moved to the forefront in operational use at seasonal time scales.  HEPEX has facilitated researchers partnering with operational groups in support of this operational infusion.  HEPEX has also promoted a science foundation in this area that shares key objectives with international science initiatives such as the WMO Global Energy and Water Cycle Experiment (GEWEX).  Such SI synergies have been highlighted previously at HEPEX meetings, and are even more strongly motivated today by expanding use of hydrologic models and ensemble techniques for the monitoring and prediction of drought, globally.

HEPEX is now discussing a more formal linkage with the GEWEX Hydroclimatology Panel (GHP) in the interest of supporting organized international experimentation and collaboration in the SI focus area.

RHPsMap
GEWEX Regional Hydroclimate Projects (RHPs)

GHP oversees a number of large-scale projects focusing on region-specific hydroclimatic variability and predictability, and guides these projects toward achieving demonstrable skill in predicting changes in water resources and soil moisture as an integral part of the climate system up to seasonal and annual time scales.

Possible outcomes from the HEPEX-GHP interaction might be the formation of a HEPEX regional, seasonal hydrologic prediction experiment, the development of a summer school program in SI hydroclimate prediction, the promotion of targeted talk sessions in the programs of international conferences such as AGU and EGU, and the development of new datasets or methods to support operational SI prediction.  The shorter-range and cross-cutting elements of HEPEX would continue, while the longer-range SI activities would benefit from the perspectives and energy of GHP.

This topic will be discussed at the GEWEX GDAP/GHP meeting next week.  HEPEX and GHP welcome your thoughts, questions, suggestions, and willingness to participate in the proposed interaction.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.