Search

Satellite remote sensing plays a pivotal role in characterizing hydrometeorological components including cloud types and their associated precipitation. The Cloud Profiling Radar (CPR) on the Polar Orbiting CloudSat satellite has provided a unique dataset to characterize cloud types. However, data from this nadir-looking radar offers limited capability for estimating precipitation because of the narrow satellite swath coverage and low temporal frequency. We use these high-quality observations to build a Deep Neural Network Cloud-Type Classification (DeepCTC) model to estimate cloud types from multispectral data from the Advanced Baseline Imager (ABI) onboard the GOES-16 platform. The DeepCTC model is trained and tested using coincident data from both CloudSat and ABI over the CONUS region. Evaluations of DeepCTC indicate that the model performs well for a variety of cloud types including Altostratus, Altocumulus, Cumulus, Nimbostratus, Deep Convective and High clouds. However, capturing low-level clouds remains a challenge for the model. Results from simulated GOES-16 ABI imageries of the Hurricane Harvey event show a large-scale perspective of the rapid and consistent cloud-type monitoring is possible using the DeepCTC model. Additionally, assessments using half-hourly Multi-Radar/Multi-Sensor (MRMS) precipitation rate data (for Hurricane Harvey as a case study) show the ability of DeepCTC in identifying rainy clouds, including Deep Convective and Nimbostratus and their precipitation potential. We also use DeepCTC to evaluate the performance of the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS) product over different cloud types with respect to MRMS referenced at a half-hourly time scale for July 2018. Our analysis suggests that DeepCTC provides supplementary insights into the variability of cloud types to diagnose the weakness and strength of near real-time GEO-based precipitation retrievals. With additional training and testing, we believe DeepCTC has the potential to augment the widely used PERSIANN-CCS algorithm for estimating precipitation.

High‐resolution gridded climate data products are crucial to research and practical applications in climatology, hydrology, ecology, agriculture, and public health. Previous works to produce multiple data sets were limited by the availability of input data as well as computational techniques. With advances in machine learning and the availability of several daily satellite data sets providing unprecedented information at 1 km or higher spatial resolutions, it is now possible to improve upon earlier data sets in terms of representing spatial variability. We developed the NEX (NASA Earth Exchange) Gridded Daily Meteorology (NEX‐GDM) model, which can estimate the spatial pattern of regional surface climate variables by aggregating several dozen two‐dimensional data sets and ground weather station data. NEX‐GDM does not require physical assumptions and can easily extend spatially and temporally. NEX‐GDM employs the random forest algorithm for estimation, which allows us to find the best estimate from the spatially continuous data sets. We used the NEX‐GDM model to produce historical 1‐km daily spatial data for the conterminous United States from 1979 to 2017, including precipitation, minimum temperature, maximum temperature, dew point temperature, wind speed, and solar radiation. In this study, NEX‐GDM ingested a total of 30 spatial variables from 13 different data sets, including satellite, reanalysis, radar, and topography data. Generally, the spatial patterns of precipitation and temperature produced were similar to previous data sets with the exception of mountain regions in the western United States. The analyses for each spatially continuous data set show that satellite and reanalysis led to better estimates and that the incorporation of satellite data allowed NEX‐GDM to capture the spatial patterns associated with urban heat island effects. The NEX‐GDM data is available to the community through the NEX data portal. [ABSTRACT FROM AUTHOR]

Projected changes in the frequency and severity of droughts as a result of increase in greenhouse gases have a significant impact on the role of vegetation in regulating the global carbon cycle. Drought effect on vegetation Gross Primary Production (GPP) is usually modeled as a function of Vapor Pressure Deficit (VPD) and/or soil moisture. Climate projections suggest a strong likelihood of increasing trend in VPD, while regional changes in precipitation are less certain. This difference in projections between VPD and precipitation can cause considerable discrepancies in the predictions of vegetation behavior depending on how ecosystem models represent the drought effect. In this study, we scrutinized the model responses to drought using the 30-year record of Global Inventory Modeling and Mapping Studies (GIMMS) 3g Normalized Difference Vegetation Index (NDVI) dataset. A diagnostic ecosystem model, Terrestrial Observation and Prediction System (TOPS), was used to estimate global GPP from 1982 to 2009 under nine different experimental simulations. The control run of global GPP increased until 2000, but stayed constant after 2000. Among the simulations with single climate constraint (temperature, VPD, rainfall and solar radiation), only the VPD-driven simulation showed a decrease in 2000s, while the other scenarios simulated an increase in GPP. The diverging responses in 2000s can be attributed to the difference in the representation of the impact of water stress on vegetation in models, i.e., using VPD and/or precipitation. Spatial map of trend in simulated GPP using GIMMS 3g data is consistent with the GPP driven by soil moisture than the GPP driven by VPD, confirming the need for a soil moisture constraint in modeling global GPP. [ABSTRACT FROM AUTHOR]

Satellite remote sensing plays a pivotal role in characterizing hydrometeorological components including cloud types and their associated precipitation. The Cloud Profiling Radar (CPR) on the Polar Orbiting CloudSat satellite has provided a unique dataset to characterize cloud types. However, data from this nadir-looking radar offers limited capability for estimating precipitation because of the narrow satellite swath coverage and low temporal frequency. We use these high-quality observations to build a Deep Neural Network Cloud-Type Classification (DeepCTC) model to estimate cloud types from multispectral data from the Advanced Baseline Imager (ABI) onboard the GOES-16 platform. The DeepCTC model is trained and tested using coincident data from both CloudSat and ABI over the CONUS region. Evaluations of DeepCTC indicate that the model performs well for a variety of cloud types including Altostratus, Altocumulus, Cumulus, Nimbostratus, Deep Convective and High clouds. However, capturing low-level clouds remains a challenge for the model. Results from simulated GOES-16 ABI imageries of the Hurricane Harvey event show a large-scale perspective of the rapid and consistent cloud-type monitoring is possible using the DeepCTC model. Additionally, assessments using half-hourly Multi-Radar/Multi-Sensor (MRMS) precipitation rate data (for Hurricane Harvey as a case study) show the ability of DeepCTC in identifying rainy clouds, including Deep Convective and Nimbostratus and their precipitation potential. We also use DeepCTC to evaluate the performance of the Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS) product over different cloud types with respect to MRMS referenced at a half-hourly time scale for July 2018. Our analysis suggests that DeepCTC provides supplementary insights into the variability of cloud types to diagnose the weakness and strength of near real-time GEO-based precipitation retrievals. With additional training and testing, we believe DeepCTC has the potential to augment the widely used PERSIANN-CCS algorithm for estimating precipitation. [ABSTRACT FROM AUTHOR]

In this paper, we present a state-of-the-art precipitation estimation framework which leverages advances in satellite remote sensing as well as Deep Learning (DL). The framework takes advantage of the improvements in spatial, spectral and temporal resolutions of the Advanced Baseline Imager (ABI) onboard the GOES-16 platform along with elevation information to improve the precipitation estimates. The procedure begins by first deriving a Rain/No Rain (R/NR) binary mask through classification of the pixels and then applying regression to estimate the amount of rainfall for rainy pixels. A Fully Convolutional Network is used as a regressor to predict precipitation estimates. The network is trained using the non-saturating conditional Generative Adversarial Network (cGAN) and Mean Squared Error (MSE) loss terms to generate results that better learn the complex distribution of precipitation in the observed data. Common verification metrics such as Probability Of Detection (POD), False Alarm Ratio (FAR), Critical Success Index (CSI), Bias, Correlation and MSE are used to evaluate the accuracy of both R/NR classification and real-valued precipitation estimates. Statistics and visualizations of the evaluation measures show improvements in the precipitation retrieval accuracy in the proposed framework compared to the baseline models trained using conventional MSE loss terms. This framework is proposed as an augmentation for PERSIANN-CCS (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Network- Cloud Classification System) algorithm for estimating global precipitation. [ABSTRACT FROM AUTHOR]