Representing spatial data — Global (SMHI)


Working with the Swedish Meteorological and Hydrological Institute, this case study explored ways of representing spatial data, using catchment and administrative boundaries.



Swedish Meteorological and Hydrological Institute (SMHI)


World-Wide HYPE Seasonal Forecasts


SMHI provided raw status and outlooks information for >131,000 watersheds for June 2020 to the HydroSOS team. Python code was developed to reprocess the data to more appropriate scales, using both administrative areas and catchments, over different zoom levels. The size of the derived datasets provides a challenge moving forwards; UKCEH currently stores the reprocessed SMHI data and outputs it to the portal using Web Map Services. Currently, only Administrative GADM Level 1 and Catchment HydroSHEDS Basins Level 5 are displayed. This code is now available to reprocess other global products for a seamless service. Methods of blending multiple global services have been evaluated under the HydroSOS initiative.

HydroSOS integration steps

Download raw SMHI polygon data
Develop spatial weighting files for remapping outputs
Calculate climatology statistics from baseline data
Process data to develop spatial categorised data, and point time series
Serve reprocessed product via Web Map Services


Three datatypes are available here; runoff, precipitation and temperature and can be viewed by administrative areas or catchments. River flow from HYPE is accompanied with modelled precipitation and temperature data to better judge the credibility. 50 ECMWF meteorological ensemble members are used. Simulated observations are used for the status assessment prior to the initialisation of the forecast. The forecast runs for seven months, six of which are displayed here. Forecasts are processed into several graphical formats to represent the ensemble results. Five categories are used to define the runoff values in the forecast, reduced to three on the map.