Improving the Quality of the NYS Mesonet Microwave Radiometer Data with a Novel Bias Correction Scheme

Temperature retrievals from Microwave radiometers (MWR) suffer from systematic bias due to instrumental calibration issues and errors in the retrieval algorithm. In this study, we developed a novel bias correction (BC) scheme that can effectively correct the systematic cold bias in the temperature retrievals from the MWR; deployed as a part of the New York State Mesonet (NYSM) Profiler Network. This BC procedure uses the observed brightness temperature from the MWR as the main predictor to model the systematic bias in the temperature retrieval using multiple-linear regression. Because the brightness temperature channels contain information on the atmospheric conditions at different height levels, the BC scheme can adjust the predicted bias at each level based on the concurrent weather conditions.

Verification based on independent radiosonde soundings during the 2024 eclipse event at Belleville, New York shows that the BC scheme not only can effectively remove the cold biases present in the NYSM MWR temperature retrievals throughout the lower- and mid-troposphere, but also reduces the standard deviation of the temperature error by about 10-15%. Compared to uncorrected data, the bias-corrected temperature profile shows a clearer collapse of the planetary boundary layer during the eclipse totality and is more consistent with the lidar observations. The near-surface temperature data also better matches the surface station measurements. These results demonstrate the effectiveness of the bias correction scheme in correcting the systematic bias in the NYSM MWR temperature retrievals and improving the data quality.