Reliability of ERA5 and ERA5-Land reanalysis data in the Canadian Prairies

Meteorological data are essential in precision agriculture in the Canadian Prairies and are often associated with spatiotemporal discontinuity and scarcity. Reanalysis data products aim to address this challenge and have recently gained popularity. The European Centre for Medium-Range Weather Forecasts' ERA5, and its high-resolution land component, ERA5-Land, are two reanalysis datasets that provide hourly estimates of many climate variables globally. This paper focuses on evaluating the performance of ERA5 and ERA5-Land over the Canadian prairies, utilizing data from 109 weather stations situated in southern Manitoba, Canada. Various variables are investigated at daily, monthly, and annual aggregation periods, including air temperature, ground temperature, soil water content, wind speed, precipitation, and evaporation. The datasets are evaluated regarding seasonal bias and spatial distribution of errors over the study area. Regression parameters are also presented to address the biases. Among the investigated variables, air temperature and wind speed exhibit the lowest errors. The evaluation further reveals an overall tendency to overpredict ground temperatures and precipitation while underpredicting evaporation. Based on these findings, the ERA5 and ERA5-Land datasets hold significant potential in applications such as climate-smart agriculture, energy demand analysis, assessing renewable energy resources, and facilitating sustainable urban development.