Searcharxiv⌕ Search

arXiv subjects

Paula J. Brown

Publications and source records attributed to Paula J. Brown.

2 recordsLinked to original sources

Towards GPROF V08: A Machine-Learning Framework for Simulating Passive Microwave Observations of Precipitation

Spatially resolved precipitation retrievals based on convolutional neural networks (CNNs) can improve precipitation estimates from passive microwave (PMW) observations and will be used to implement the next operational PMW retrieval algorithm for the Global Precipitation Measurement (GPM) mission. However, the conventional radiative transfer simulations used to extend the retrieval algorithm to the various sensors of the GPM constellation do not provide the spatial coverage required to train CNN-based retrievals. To address this, we introduce a machine-learning-based simulation framework that enables the training of the CNN-based GPROF-NN 3D retrieval algorithm for the full GPM constellation. At the core of this framework is Satformer, a new neural-network architecture that can translate PMW observations between any pair of GPM sensors. The resulting synthetic observations exhibit lower errors than conventional radiative transfer simulations and lead to more accurate precipitation retrievals when used to train GPROF-NN precipitation retrievals. We further demonstrate the applicability of the framework to new instruments by developing GPROF-NN retrievals for the Microwave Imager aboard the Weather System Follow-on Microwave satellite. While GPROF-NN retrievals based on the new simulation framework achieve lower instantaneous retrieval errors and higher correlations compared to independent reference data than conventional retrievals, residual simulation errors can introduce systematic biases in the resulting retrievals. The proposed simulation framework provides a key component for extending GPROF-NN to the full GPM constellation as part of the upcoming operational GPROF V08 retrieval algorithm. The results demonstrate the potential of machine-learning-based simulations to support spatially resolved precipitation retrievals across heterogeneous satellite constellations.

physics.ao-ph↗

GPROF V08: Improving Satellite Precipitation Estimates Across a Heterogeneous Constellation of Passive-Microwave Sensors

Precipitation estimates from the passive microwave observations (PMW) of the Global Precipitation Measurements (GPM) mission serve a wide range of scientific, societal, and economic applications. The Goddard Profiling Algorithm (GPROF) is the operational retrieval used to produce the precipitation estimates for all GPM PMW sensors. This study presents the implementation GPROF V08 -- the most recent version of GPROF and a major update aimed at improving the accuracy of precipitation estimates from the GPM constellation. GPROF V08 replaces the conventional Bayesian scheme used by earlier GPROF versions by a convolutional neural network (CNN). The first part of this two-part study described a novel ML framework for generating synthetic observations required for training CNN retrievals for all GPM sensors. This second part describes the final, operational implementation, which adds a second finetuning stage to the training.Retrievals from the new GPROF V08 algorithm for all principal GPM sensors are validated against reference precipitation estimates from various sources. Our results show that, while the magnitude of improvements vary by sensor and surface type, they are robust across sensors and metrics and can be as high as 40 \% for the linear correlation coefficient for precipitation estimates from SSMIS and ATMS over ocean compared to combined radar and radiometer retrievals from GPM. While some challenges remain -- such as an increased sensitivity to sensor degradation and the persistence of climatological biases, which remain of the same order as for GPROF V07 -- GPROF V08 constitutes a substantial advancement in spaceborne precipitation monitoring. Since the improvements are robust across all sensor of the constellation, the update is expected to improve the full record of GPM precipitation estimates and thus benefit a wide range of downstream datasets and applications.

physics.ao-ph↗