Assessment of the Advanced Very High-Resolution Radiometer (AVHRR) for Snowfall Retrieval in High Latitudes Using CloudSat and Machine Learning

Abstract Precipitation retrieval is a challenging topic, especially in high latitudes (HL), and current precipitation products face ample challenges over these regions. This study investigates the potential of Advanced Very High-Resolution Radiometer (AVHRR) for snowfall HL using CloudSat radar information machine learning (ML). With all known limitations, AVHRR observations should be considered because (1) data have been continuously collected about four decades on multiple platforms with global coverage, similar will likely continue future; (2) passive microwave satellite several issues snow ice surfaces; (3) good coincident between are available training ML algorithms. Using ML, rate was retrieved from AVHRR’s brightness temperature cloud probability, as well auxiliary provided by numerical reanalysis. The results indicate that ML-based algorithm capable detection estimation comparable or better statistical scores than those obtained Atmospheric Infrared Sounder (AIRS) two sensors contributing to Global Measurement (GPM) mission constellation. outcomes also suggest AVHRR-based retrievals spatially temporally reasonable can quantitatively useful input merged require frequent sampling long-term records.