Estimating and Removing the Sampling Biases of the AIRS Obs4MIPs V2 Data

The Atmospheric Infrared Sounder (AIRS) Observations for Model Intercomparison Projects (Obs4MIPs) Version 2 (V2) monthly mean tropospheric air temperature, specific humidity and relative humidity data designed for climate model evaluation in the context of the Coupled Model Intercomparison Project (CMIP) have been publicly available at Earth System Grid Federation (ESGF) data centers since 2018. Due to the limitations of the Aqua satellite orbit and the AIRS retrieval algorithm, the AIRS Ob4MIPs V2 data are not sampled on the regular spatial and temporal grids like the climate model outputs. The resulting sampling biases of the AIRS Obs4MIPs V2 data from the sampling difference between the AIRS Ob4MIPs V2 data and climate model outputs can be large for certain cases and have to be considered when the AIRS Obs4MIPs V2 data are compared to the climate model outputs for climate model evaluation. We have estimated the sampling biases of the AIRS Obs4MIPs V2 data based on the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5) and published them at the ESGF data centers. This sampling bias dataset is intended for users who wish to compare the AIRS Obs4MIPs V2 data with the climate model outputs for climate model evaluation to account the sampling difference between the AIRS and climate model data. This paper documents this sampling bias dataset in terms of its motivation, description, processing methodology, validation, and the issues to be considered for climate model evaluation.