Determination of Cloud Liquid Water Path over the Oceans from Special Sensor Microwave/Imager (SSM/I) Data Using Neural Networks

Abstract

A neural network (NN) has been developed in order to retrieve the cloud liquid water path (LWP) over the oceans from Special Sensor Microwave/Imager (SSM/I) data. The retrieval with NNs depends crucially on the SSM/I channels used as input and the number of hidden neurons—that is, the NN architecture. Three different combinations of the seven SSM/I channels have been tested. For all three methods an NN with five hidden neurons yields the best results. The NN-based LWP algorithms for SSM/I observations are intercompared with a standard regression algorithm. The calibration and validation of the retrieval algorithms are based on 2060 radiosonde observations over the global ocean. For each radiosonde profile the LWP is parameterized and the brightness temperatures (Tb’s) are simulated using a radiative transfer model. The best LWP algorithm (all SSM/I channels except T85V) shows a theoretical error of 0.009 kg m−2 for LWPs up to 2.8 kg m−2 and theoretical “clear-sky noise” (0.002 kg m−2), which has been reduced relative to the regression algorithm (0.031 kg m−2). Additionally, this new algorithm avoids the estimate of negative LWPs. An indirect validation and intercomparison is presented that is based upon SSM/I measurements (F-10) under clear-sky conditions, classified with independent IR-Meteosat data. The NN-based algorithms outperform the regression algorithm. The best LWP algorithm shows a clear-sky standard deviation of 0.006 kg m−2, a bias of 0.001 kg m−2, nonnegative LWPs, and no correlation with total precipitable water. The estimated accuracy for SSM/I observations and two of the proposed new LWP algorithms is 0.023 kg m−2 for LWP ⩽ 0.5 kg m−2.