Satellite sensors have been widely used for precipitation retrieval, and a number of precipitation retrieval algorithms have been developed using observations from various satellite sensors. The current operational rainfall rate quantitative precipitation estimate (RRQPE) product from the geostationary operational environmental satellite (GOES) offers full disk rainfall rate estimates based on the observations from the advanced baseline imager (ABI) aboard the GOES-R series. However, accurate precipitation retrieval using satellite sensors is still challenging due to the limitations on spatio-temporal sampling of the satellite sensors and/or the uncertainty associated with the applied parametric retrieval algorithms. In this article, we propose a deep learning framework for precipitation retrieval using the combined observations from the ABI and geostationary lightning mapper (GLM) on the GOES-R series to improve the current operational RRQPE product. Particularly, the proposed deep learning framework is composed of two deep convolutional neural networks (CNNs) that are designed for precipitation detection and quantification. The cloud-top brightness temperature from multiple ABI channels and the lightning flash rate from the GLM measurement are used as inputs to the deep learning framework. To train the designed CNNs, the precipitation product multiradar multi-sensor (MRMS) system from the National Oceanic and Atmospheric Administration (NOAA) is used as target labels to optimize the network parameters. The experimental results show that the precipitation retrieval performance of the proposed framework is superior to the currently operational GOES RRQPE product in the selected study domain, and the performance is dramatically enhanced after incorporating the lightning data into the deep learning model. Using the independent MRMS product as a reference, the deep learning model can reduce the retrieval uncertainty in the operational RRQPE product by at least 31% in terms of the mean squared error and normalized mean absolute error, and the improvement is more significant in moderate to heavy rain regions. Therefore, the proposed deep learning framework can potentially serve as an alternative approach for GOES precipitation retrievals.
more »
« less
Precipitation Retrieval Using ABI and GLM Measurements on the Goes-R Series
Accurate precipitation retrieval using satellite sensors is still challenging due to the limitations on spatio-temporal sampling of the applied parametric retrieval algorithms. In this research, we propose a deep learning framework for precipitation retrieval using the observations from Advanced Baseline Imager (ABI), and Geostationary Lightning Mapper (GLM) on GOES-R satellite series. In particular, two deep Convolutional Neural Network (CNN) models are designed to detect and estimate the precipitation using the cloud-top brightness temperature from ABI and lightning flash rate from GLM. The precipitation estimates from the ground-based Multi-Radar/Multi-Sensor (MRMS) system are used as the target labels in the training phase. The experimental results show that in the testing phase, the proposed framework offers more accurate precipitation estimates than the current operational Rainfall Rate Quantitative Precipitation Estimate (RRQPE) product from GOES-R.
more »
« less
- Award ID(s):
- 2239880
- NSF-PAR ID:
- 10515825
- Publisher / Repository:
- IEEE
- Date Published:
- ISBN:
- 979-8-3503-2010-7
- Page Range / eLocation ID:
- 1822 to 1825
- Format(s):
- Medium: X
- Location:
- Pasadena, CA, USA
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract The terrestrial carbon cycle varies dynamically on hourly to weekly scales, making it difficult to observe. Geostationary (“weather”) satellites like the Geostationary Environmental Operational Satellite - R Series (GOES-R) deliver near-hemispheric imagery at a ten-minute cadence. The Advanced Baseline Imager (ABI) aboard GOES-R measures visible and near-infrared spectral bands that can be used to estimate land surface properties and carbon dioxide flux. However, GOES-R data are designed for real-time dissemination and are difficult to link with eddy covariance time series of land-atmosphere carbon dioxide exchange. We compiled three-year time series of GOES-R land surface attributes including visible and near-infrared reflectances, land surface temperature (LST), and downwelling shortwave radiation (DSR) at 314 ABI fixed grid pixels containing eddy covariance towers. We demonstrate how to best combine satellite and
in-situ datasets and show how ABI attributes useful for ecosystem monitoring vary across space and time. By connecting observation networks that infer rapid changes to the carbon cycle, we can gain a richer understanding of the processes that control it. -
more » « less
Abstract Accurate measurements of global lightning are essential for understanding present and future atmospheric electricity, composition, and climate. The latest space‐based lightning detector, the Geostationary Lightning Mapper (GLM), was the first to be placed in geostationary orbit, with a continuous view of most of the American continents. Prior to the GLM, the Lightning Imaging Sensor (LIS) on the Tropical Rainfall Measuring Mission (TRMM) satellite collected lightning measurements from which numerous lightning climatologies have been developed, including those used in global models. However, this study finds that both the GLM and a second, similar LIS placed on the International Space Station (ISS) in 2017 detect lightning at similar rates and are undercounting lightning compared to ground‐based Lightning Mapping Arrays (LMAs). The GLM undercounts lightning by an average factor of 7.0, reaching a maximum over 120 as a function of satellite zenith angle, radar reflectivity at a height where the temperature is −10°C, flash height, and thunderstorm polarity. The LIS is estimated to undercount lightning by an average factor of 5.6, reaching a maximum of 75.0 as a function of radar reflectivity at the −10°C level, flash height, and thunderstorm polarity. Preliminary predictive equations for the GLM and LIS lightning undercount factor, or scaling factor (SF), use ice‐water content, equilibrium level, flash height, and satellite zenith angle, all of which can be derived in models. These equations are developed to encourage updating lightning parameterizations within global models and will likely increase modeled lightning's effects on atmospheric electrical circuits, composition, chemistry, and climate change.
-
This study investigates high-frequency mapping of downward shortwave radiation (DSR) at the Earth’s surface using the advanced baseline imager (ABI) instrument mounted on Geo- stationary Operational Environmental Satellite—R Series (GOES- R). The existing GOES-R DSR product (DSRABI) offers hourly temporal resolution and spatial resolution of 0.25°. To enhance these resolutions, we explore machine learning (ML) for DSR estimation at the native temporal resolution of GOES-R Level-2 cloud and moisture imagery product (5 min) and its native spatial resolution of 2 km at nadir. We compared four common ML regres- sion models through the leave-one-out cross-validation algorithm for robust model assessment against ground measurements from AmeriFlux and SURFRAD networks. Results show that gradient boosting regression (GBR) achieves the best performance (R2 = 0.916, RMSE = 88.05 W·m−2) with more efficient computation compared to long short-term memory, which exhibited similar performance. DSR estimates from the GBR model through the ABI live imaging of vegetated ecosystems workflow (DSRALIVE) outperform DSRABI across various temporal resolutions and sky conditions. DSRALIVE agreement with ground measurements at SURFRAD networks exhibits high accuracy at high temporal res- olutions (5-min intervals) with R2 exceeding 0.85 and RMSE = 122 W·m−2 . We conclude that GBR offers a promising approach for high-frequency DSR mapping from GOES-R, enabling improved applications for near-real-time monitoring of terrestrial carbon and water fluxes.more » « less
-
more » « less
Abstract Precipitation plays an important role in cloud and aerosol processes over the Southern Ocean (SO). The main objective of this study is to characterize SO precipitation properties associated with SO stratocumulus clouds. We use data from the Southern Ocean Clouds Radiation Aerosol Transport Experimental Study (SOCRATES), and leverage observations from airborne radar, lidar, and in situ probes. We find that for the cold‐topped clouds (cloud‐top‐temperature <0°C), the phase of precipitation with reflectivity >0 dBZ is predominantly ice, while reflectivity < −10 dBZ is predominantly liquid. Liquid‐phase precipitation properties are retrieved where radar and lidar are zenith‐pointing. Power‐law relationships between reflectivity (Z) and rain rate (R) are developed, and the derived Z–R relationships show vertical dependence and sensitivity to the presence of droplets with diameters between 10 and 40 μm. Using derived Z–R relationships, a reflectivity‐velocity (ZV) retrieval method, and a radar‐lidar retrieval method, we derive rain rate and other precipitation properties. The retrieved rain rate from all three methods shows good agreement with in‐situ aircraft estimates, with rain rates typically being quite light (<0.1 mm hr−1). We examine the vertical distribution of precipitation properties, and find that rain rate, precipitation number concentration, and precipitation liquid water all decrease as one gets closer to the surface, while precipitation size and distribution width increases. We also examine how cloud base rain rate (
R CB ) depends on cloud depth (H) and aerosol concentration (N a ) for particles with a diameter greater than 70 nm, and find thatR CB is proportional to .
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/IGARSS52108.2023.10283405
