Climate Reanalysis Research

Reanalysis integrates the instrumental record with model forecasts, produces a coherent set of climate variables for monitoring and analysis of climate variability, and supports climate prediction and climate model development. Reanalysis datasets have been identified as an important component of the WMO Global Framework for Climate Services. Major reanalysis datasets have recently been developed in the US (CFSR, MERRA, 20CR) and internationally (e.g., ERA-Interim, JRA-25). Evaluation of these datasets has demonstrated many advances (e.g., improvements in the representation of intraseasonal variability, and the capability to represent climate anomalies back to the early 1900s) but has also highlighted issues that limit the use of these datasets for climate studies. These include spurious trends and jumps associated with the changing observing system, shortcomings in the representation of the hydrological cycle and surface fluxes, and issues with the quality of reanalyses in the stratosphere and in polar regions. (For a technical report on this topic, see https://reanalyses.org/reports/progress-and-prospects). Newly available remotely-sensed data, the development of Earth System models, and the benefits that would derive from more consistent states across analysis components, all point to a considerable interest in integrated Earth System reanalysis systems, including some level of coupling among the various analysis components (e.g., atmosphere and ocean).