Summary
EM-Earth is a newly developed global ensemble meteorological dataset that addresses limitations in existing deterministic gridded products by providing both point estimates and probabilistic uncertainty quantification across four key variables. The dataset merges station observations (via an intermediate Serially Complete Earth product) with ERA5 reanalysis data to generate 25 correlated ensemble members, enabling uncertainty-aware applications in hydrology, climate impact assessment, and agriculture. Validation across global land areas reveals substantial geographic variation in performance, with probabilistic approaches particularly valuable in data-sparse regions with high inherent meteorological uncertainty.
UK applicability
EM-Earth's ensemble approach and uncertainty quantification may benefit UK agricultural and hydrological applications, particularly for risk assessment and water management. As a Europe-well-performing dataset, it is likely suitable for UK conditions, though users should consult specific validation metrics for British Isles coverage.
Key measures
Spatial resolution (0.1°), temporal coverage (1950–2019), ensemble member count (25), variables (precipitation, mean daily temperature, daily temperature range, dewpoint temperature), validation metrics (leave-one-out cross-validation, independent station comparison, inter-dataset comparison)
Outcomes reported
The study reports development and validation of EM-Earth, a gridded meteorological dataset providing both deterministic and probabilistic (25-member ensemble) estimates of precipitation, temperature, and related variables at 0.1° resolution globally from 1950–2019. Validation shows variable performance across regions, with stronger accuracy in Europe, North America and Oceania than in Africa, Asia and South America.
Topic tags
Dig deeper with Pulse AI.
Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.