Pub Date : 2024-07-15DOI: 10.1175/aies-d-23-0099.1
Jorge Baño-Medina, M. Iturbide, Jesús Fernández, José Manuel Gutiérrez
Regional climate models (RCMs) are essential tools for simulating and studying regional climate variability and change. However, their high computational cost limits the production of comprehensive ensembles of regional climate projections covering multiple scenarios and driving Global Climate Models (GCMs) across regions. RCM emulators based on deep learning models have recently been introduced as a cost-effective and promising alternative that requires only short RCM simulations to train the models. Therefore, evaluating their transferability to different periods, scenarios, and GCMs becomes a pivotal and complex task in which the inherent biases of both GCMs and RCMs play a significant role. Here we focus on this problem by considering the two different emulation approaches introduced in the literature as perfect and imperfect, that we here refer to as Perfect Prognosis (PP) and Model Output Statistics (MOS), respectively, following the well-established downscaling terminology. In addition to standard evaluation techniques, we expand the analysis with methods from the field of eXplainable Artificial Intelligence (XAI), to assess the physical consistency of the empirical links learnt by the models. We find that both approaches are able to emulate certain climatological properties of RCMs for different periods and scenarios (soft transferability), but the consistency of the emulation functions differ between approaches. Whereas PP learns robust and physically meaningful patterns, MOS results are GCM-dependent and lack physical consistency in some cases. Both approaches face problems when transferring the emulation function to other GCMs (hard transferability), due to the existence of GCM-dependent biases. This limits their applicability to build RCM ensembles. We conclude by giving prospects for future applications.
{"title":"Transferability and explainability of deep learning emulators for regional climate model projections: Perspectives for future applications","authors":"Jorge Baño-Medina, M. Iturbide, Jesús Fernández, José Manuel Gutiérrez","doi":"10.1175/aies-d-23-0099.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0099.1","url":null,"abstract":"\u0000Regional climate models (RCMs) are essential tools for simulating and studying regional climate variability and change. However, their high computational cost limits the production of comprehensive ensembles of regional climate projections covering multiple scenarios and driving Global Climate Models (GCMs) across regions. RCM emulators based on deep learning models have recently been introduced as a cost-effective and promising alternative that requires only short RCM simulations to train the models. Therefore, evaluating their transferability to different periods, scenarios, and GCMs becomes a pivotal and complex task in which the inherent biases of both GCMs and RCMs play a significant role. Here we focus on this problem by considering the two different emulation approaches introduced in the literature as perfect and imperfect, that we here refer to as Perfect Prognosis (PP) and Model Output Statistics (MOS), respectively, following the well-established downscaling terminology. In addition to standard evaluation techniques, we expand the analysis with methods from the field of eXplainable Artificial Intelligence (XAI), to assess the physical consistency of the empirical links learnt by the models. We find that both approaches are able to emulate certain climatological properties of RCMs for different periods and scenarios (soft transferability), but the consistency of the emulation functions differ between approaches. Whereas PP learns robust and physically meaningful patterns, MOS results are GCM-dependent and lack physical consistency in some cases. Both approaches face problems when transferring the emulation function to other GCMs (hard transferability), due to the existence of GCM-dependent biases. This limits their applicability to build RCM ensembles. We conclude by giving prospects for future applications.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"27 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141647748","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-10DOI: 10.1175/aies-d-23-0091.1
Carl G. Schmitt, E. Järvinen, M. Schnaiter, D. Vas, Lea Hartl, Telayna Wong, M. Stuefer
Machine Learning (ML) has rapidly transitioned from a niche activity to a mainstream tool for environmental research applications including atmospheric science cloud microphysics studies. Two recently developed cloud particle probes measure the light scattered in the near forward direction and save digital images of the scattering light. Scattering pattern images collected by the Particle Phase Discriminator (PPD-2K) and the Small Ice Detector version 3 (SID-3) provide valuable information for particle shape and size characterization. Since different particle shapes have distinctly different light scattering characteristics, the images are ideally suited for ML. Here results of a ML project to characterize ice particle shapes sampled by the PPD-2K in ice fog and diamond dust during a 3-year project in Fairbanks, Alaska. 2.15 million light scattering pattern images were collected during three years of measurements with the PPD-2K. Visual Geometry Group (VGG) Convolutional Neural Network (CNN) was trained to categorize light scattering patterns into 8 categories. Initial training images (120 each category) were selected by human visual examination of data and the training dataset was augmented using an automated iterative method for image identification of further images which were all visually inspected by a human. Results were well correlated to similar categories identified from previously developed classification algorithms. ML identify characteristics not included in automated analysis such as sublimation. Of the 2.15 million images analyzed, 1.3% were categorized as spherical (liquid), 43.5% were categorized as having rough surfaces, 15.3% were pristine, 16.3% were categorized as sublimating and the remaining 23.6% did not fit into any of those categories (irregular or saturated).
{"title":"Classification of ice particle shapes using machine learning on forward light scattering images","authors":"Carl G. Schmitt, E. Järvinen, M. Schnaiter, D. Vas, Lea Hartl, Telayna Wong, M. Stuefer","doi":"10.1175/aies-d-23-0091.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0091.1","url":null,"abstract":"\u0000Machine Learning (ML) has rapidly transitioned from a niche activity to a mainstream tool for environmental research applications including atmospheric science cloud microphysics studies. Two recently developed cloud particle probes measure the light scattered in the near forward direction and save digital images of the scattering light. Scattering pattern images collected by the Particle Phase Discriminator (PPD-2K) and the Small Ice Detector version 3 (SID-3) provide valuable information for particle shape and size characterization. Since different particle shapes have distinctly different light scattering characteristics, the images are ideally suited for ML. Here results of a ML project to characterize ice particle shapes sampled by the PPD-2K in ice fog and diamond dust during a 3-year project in Fairbanks, Alaska.\u00002.15 million light scattering pattern images were collected during three years of measurements with the PPD-2K. Visual Geometry Group (VGG) Convolutional Neural Network (CNN) was trained to categorize light scattering patterns into 8 categories. Initial training images (120 each category) were selected by human visual examination of data and the training dataset was augmented using an automated iterative method for image identification of further images which were all visually inspected by a human. Results were well correlated to similar categories identified from previously developed classification algorithms. ML identify characteristics not included in automated analysis such as sublimation. Of the 2.15 million images analyzed, 1.3% were categorized as spherical (liquid), 43.5% were categorized as having rough surfaces, 15.3% were pristine, 16.3% were categorized as sublimating and the remaining 23.6% did not fit into any of those categories (irregular or saturated).","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"7 13","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141661642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-09DOI: 10.1175/aies-d-23-0112.1
Robin Marcille, P. Tandeo, Maxime Thiébaut, Pierre Pinson, R. Fablet
The safe and efficient execution of offshore operations requires short-term (1 to 6 hours ahead) high-quality probabilistic forecasts of metocean variables. The development areas for offshore wind projects, potentially in high depths, make it difficult to gather measurement data. This paper explores the use of deep learning for wind speed forecasting at an unobserved offshore location. The proposed convolutional architecture jointly exploits coastal measurements and numerical weather predictions to emulate multivariate probabilistic short-term forecasts. We explore both Gaussian and non-Gaussian neural representations using normalizing flows. We benchmark these approaches with respect to state-of-art data-driven schemes, including analog methods and quantile forecasting. The performance of the models, and resulting forecast quality, are analyzed in terms of probabilistic calibration, probabilistic and deterministic metrics, and as a function of weather situations. We report numerical experiments for a real case-study off the French Mediterranean coast. Our results highlight the role of regional numerical weather prediction and coastal in situ measurement in the performance of the post-processing. For single-valued forecasts, a 40% decrease in RMSE is observed compared to the direct use of numerical weather predictions. Significant skill improvements are also obtained for the probabilistic forecasts, in terms of various scores, as well as an acceptable probabilistic calibration. The proposed architecture can process a large amount of heterogeneous input data, and offers a versatile probabilistic framework for multivariate forecasting.
{"title":"Convolutional encoding and normalizing flows: a deep learning approach for offshore wind speed probabilistic forecasting in the Mediterranean Sea","authors":"Robin Marcille, P. Tandeo, Maxime Thiébaut, Pierre Pinson, R. Fablet","doi":"10.1175/aies-d-23-0112.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0112.1","url":null,"abstract":"\u0000The safe and efficient execution of offshore operations requires short-term (1 to 6 hours ahead) high-quality probabilistic forecasts of metocean variables. The development areas for offshore wind projects, potentially in high depths, make it difficult to gather measurement data. This paper explores the use of deep learning for wind speed forecasting at an unobserved offshore location. The proposed convolutional architecture jointly exploits coastal measurements and numerical weather predictions to emulate multivariate probabilistic short-term forecasts. We explore both Gaussian and non-Gaussian neural representations using normalizing flows. We benchmark these approaches with respect to state-of-art data-driven schemes, including analog methods and quantile forecasting. The performance of the models, and resulting forecast quality, are analyzed in terms of probabilistic calibration, probabilistic and deterministic metrics, and as a function of weather situations. We report numerical experiments for a real case-study off the French Mediterranean coast. Our results highlight the role of regional numerical weather prediction and coastal in situ measurement in the performance of the post-processing. For single-valued forecasts, a 40% decrease in RMSE is observed compared to the direct use of numerical weather predictions. Significant skill improvements are also obtained for the probabilistic forecasts, in terms of various scores, as well as an acceptable probabilistic calibration. The proposed architecture can process a large amount of heterogeneous input data, and offers a versatile probabilistic framework for multivariate forecasting.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"120 11","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141665284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-03DOI: 10.1175/aies-d-23-0053.1
Huiying Ren, Jian Lu, Z. J. Hou, Tse-Chun Chen, L. R. Leung, Fukai Liu
Of great relevance to climate engineering is the systematic relationship between the radiative forcing to the climate system and the response of the system, a relationship often represented by the linear response function (LRF) of the system. However, estimating the LRF often becomes an ill-posed inverse problem due to high-dimensionality and non-unique relationships between the forcing and response. Recent advances in machine learning make it possible to address the ill-posed inverse problem through regularization and sparse system fitting. Here we develop a convolutional neural network (CNN) for regularized inversion. The CNN is trained using the surface temperature responses from a set of Green’s function perturbation experiments as imagery input data together with data sample densification. The resulting CNN model can infer the forcing pattern responsible for the temperature response from out-of-sample forcing scenarios. This promising proof-of-concept suggests a possible strategy for estimating the optimal forcing to negate certain undesirable effects of climate change. The limited success of this effort underscores the challenges of solving an inverse problem for a climate system with inherent nonlinearity.
{"title":"Neural networks to find the optimal forcing for offsetting the anthropogenic climate change effects","authors":"Huiying Ren, Jian Lu, Z. J. Hou, Tse-Chun Chen, L. R. Leung, Fukai Liu","doi":"10.1175/aies-d-23-0053.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0053.1","url":null,"abstract":"\u0000Of great relevance to climate engineering is the systematic relationship between the radiative forcing to the climate system and the response of the system, a relationship often represented by the linear response function (LRF) of the system. However, estimating the LRF often becomes an ill-posed inverse problem due to high-dimensionality and non-unique relationships between the forcing and response. Recent advances in machine learning make it possible to address the ill-posed inverse problem through regularization and sparse system fitting. Here we develop a convolutional neural network (CNN) for regularized inversion. The CNN is trained using the surface temperature responses from a set of Green’s function perturbation experiments as imagery input data together with data sample densification. The resulting CNN model can infer the forcing pattern responsible for the temperature response from out-of-sample forcing scenarios. This promising proof-of-concept suggests a possible strategy for estimating the optimal forcing to negate certain undesirable effects of climate change. The limited success of this effort underscores the challenges of solving an inverse problem for a climate system with inherent nonlinearity.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":" 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141680380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-01DOI: 10.1175/aies-d-23-0011.1
Masudur R. Siddiquee, Aurelien O Meray, Zexuan Xu, Hansell Gonzalez-Raymat, Thomas Danielson, Himanshu Upadhyay, Leonel E. Lagos, Carol Eddy-Dilek, Haruko Wainwright
Long-term environmental monitoring is critical for managing the soil and groundwater at contaminated sites. Recent improvementsin state-of-the-art sensor technology, communication networks, and artificial intelligence have created opportunities to modernize this monitoring activity for automated, fast, robust, and predictive monitoring. In such modernization, it is required that sensor locations be optimized to capture the spatiotemporal dynamics of all monitoring variables as well as to make it cost-effective. The legacy monitoring datasets of the target area are important to perform this optimization. In this study, we have developed a machine-learning approach to optimize sensor locations for soil and groundwater monitoring based on ensemble supervised learning and majority voting. For spatial optimization, Gaussian Process Regression (GPR) is used for spatial interpolation, while the majority voting is applied to accommodate the multivariate temporal dimension. Results show that the algorithms significantly outperform the random selection of the sensor locations for predictive spatiotemporal interpolation. While the method has been applied to a four-dimensional dataset (with two-dimensional space, time and multiple contaminants), we anticipate that it can be generalizable to higher dimensional datasets for environmental monitoring sensor location optimization.
{"title":"Machine Learning Approach for Spatiotemporal Multivariate Optimization of Environmental Monitoring Sensor Locations","authors":"Masudur R. Siddiquee, Aurelien O Meray, Zexuan Xu, Hansell Gonzalez-Raymat, Thomas Danielson, Himanshu Upadhyay, Leonel E. Lagos, Carol Eddy-Dilek, Haruko Wainwright","doi":"10.1175/aies-d-23-0011.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0011.1","url":null,"abstract":"\u0000Long-term environmental monitoring is critical for managing the soil and groundwater at contaminated sites. Recent improvementsin state-of-the-art sensor technology, communication networks, and artificial intelligence have created opportunities to modernize this monitoring activity for automated, fast, robust, and predictive monitoring. In such modernization, it is required that sensor locations be optimized to capture the spatiotemporal dynamics of all monitoring variables as well as to make it cost-effective. The legacy monitoring datasets of the target area are important to perform this optimization. In this study, we have developed a machine-learning approach to optimize sensor locations for soil and groundwater monitoring based on ensemble supervised learning and majority voting. For spatial optimization, Gaussian Process Regression (GPR) is used for spatial interpolation, while the majority voting is applied to accommodate the multivariate temporal dimension. Results show that the algorithms significantly outperform the random selection of the sensor locations for predictive spatiotemporal interpolation. While the method has been applied to a four-dimensional dataset (with two-dimensional space, time and multiple contaminants), we anticipate that it can be generalizable to higher dimensional datasets for environmental monitoring sensor location optimization.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"8 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141713871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-24DOI: 10.1175/aies-d-23-0002.1
Chandra M. Pasillas, Christian Kummerow, Michael Bell, Steven D. Miller
Meteorological satellite imagery is a critical asset for observing and forecasting weather phenomena. The Joint Polar Satellite System (JPSS) Visible Infrared Imaging Radiometer Suite (VIIRS) Day-Night Band (DNB) sensor collects measurements from moonlight, airglow, and artificial lights. DNB radiances are then manipulated and scaled with a focus on digital display. DNB imagery performance is tied to the lunar cycle, with best performance during the full moon and worst with the new moon. We propose using feed-forward neural networks models to transform brightness temperatures and wavelength differences in the infrared spectrum to a pseudo lunar reflectance value based on lunar reflectance values derived from observed DNB radiances. JPSS NOAA-20 and Suomi National Polar-orbiting Partnership (SNPP) satellite data over the North Pacific Ocean at night for full moon periods from December 2018 - November 2020 were used to design the models. The pseudo lunar reflectance values are quantitatively compared to DNB lunar reflectance, providing the first-ever lunar reflectance baseline metrics. The resulting imagery product, Machine Learning Night-time Visible Imagery (ML-NVI), is qualitatively compared to DNB lunar reflectance and infrared imagery across the lunar cycle. The imagery goal is not only to improve upon the consistency performance of DNB imagery products across the lunar cycle, but ultimately lay the foundation for transitioning the algorithm to geostationary sensors, making global continuous nighttime imagery possible. ML-NVI demonstrates its ability to provide DNB derived imagery with consistent contrast and representation of clouds across the full lunar cycle for night-time cloud detection.
{"title":"Turning Night Into Day: The Creation and Validation of Synthetic Night-time Visible Imagery Using the Visible Infrared Imaging Radiometer Suite (VIIRS) Day-Night Band (DNB) and Machine Learning","authors":"Chandra M. Pasillas, Christian Kummerow, Michael Bell, Steven D. Miller","doi":"10.1175/aies-d-23-0002.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0002.1","url":null,"abstract":"\u0000Meteorological satellite imagery is a critical asset for observing and forecasting weather phenomena. The Joint Polar Satellite System (JPSS) Visible Infrared Imaging Radiometer Suite (VIIRS) Day-Night Band (DNB) sensor collects measurements from moonlight, airglow, and artificial lights. DNB radiances are then manipulated and scaled with a focus on digital display. DNB imagery performance is tied to the lunar cycle, with best performance during the full moon and worst with the new moon. We propose using feed-forward neural networks models to transform brightness temperatures and wavelength differences in the infrared spectrum to a pseudo lunar reflectance value based on lunar reflectance values derived from observed DNB radiances. JPSS NOAA-20 and Suomi National Polar-orbiting Partnership (SNPP) satellite data over the North Pacific Ocean at night for full moon periods from December 2018 - November 2020 were used to design the models. The pseudo lunar reflectance values are quantitatively compared to DNB lunar reflectance, providing the first-ever lunar reflectance baseline metrics. The resulting imagery product, Machine Learning Night-time Visible Imagery (ML-NVI), is qualitatively compared to DNB lunar reflectance and infrared imagery across the lunar cycle. The imagery goal is not only to improve upon the consistency performance of DNB imagery products across the lunar cycle, but ultimately lay the foundation for transitioning the algorithm to geostationary sensors, making global continuous nighttime imagery possible. ML-NVI demonstrates its ability to provide DNB derived imagery with consistent contrast and representation of clouds across the full lunar cycle for night-time cloud detection.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"6 18","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141100745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-22DOI: 10.1175/aies-d-23-0098.1
Da Fan, S. Greybush, E. Clothiaux, David John Gagne
Convective initiation (CI) nowcasting remains a challenging problem for both numerical weather prediction models and existing nowcasting algorithms. In this study, an object-based probabilistic deep learning model is developed to predict CI based on multichannel infrared GOES-16 satellite observations. The data come from patches surrounding potential CI events identified in Multi-Radar Multi-Sensor Doppler weather radar products over the Great Plains region from June and July 2020 and June 2021. An objective radar-based approach is used to identify these events. The deep learning model significantly outperforms the classical logistic model at lead times up to 1 hour, especially on the false alarm ratio. Through case studies, the deep learning model exhibits dependence on the characteristics of clouds and moisture at multiple altitudes. Model explanation further reveals that the contribution of features to model predictions is significantly dependent on the baseline, a reference point against which the prediction is compared. Under a moist baseline, moisture gradients in the lower and middle troposphere contribute most to correct CI forecasts. In contrast, under clear-sky baselines, correct CI forecasts are dominated by cloud-top features, including cloud-top glaciation, height, and cloud coverage. Our study demonstrates the advantage of using different baselines in further understanding model behavior and gaining scientific insights.
对于数值天气预报模式和现有的预报算法来说，对流起始（CI）预报仍然是一个具有挑战性的问题。在本研究中，基于多通道红外 GOES-16 卫星观测数据，开发了一种基于对象的概率深度学习模型来预测对流起始（CI）。数据来自 2020 年 6 月、7 月和 2021 年 6 月大平原地区上空多雷达多传感器多普勒天气雷达产品中确定的潜在 CI 事件周围的斑块。采用基于雷达的客观方法来识别这些事件。深度学习模型在最多 1 小时的准备时间内明显优于经典逻辑模型，尤其是在误报率方面。通过案例研究，深度学习模型表现出与多高度云层和湿度特征的依赖性。对模型的解释进一步表明，特征对模型预测的贡献在很大程度上取决于基线，即与预测进行比较的参考点。在湿润基线下，对流层中下部的湿度梯度对正确的 CI 预测贡献最大。相反，在晴空基线下，正确的 CI 预报主要取决于云顶特征，包括云顶冰蚀、高度和云层覆盖。我们的研究证明了使用不同基线在进一步理解模式行为和获得科学见解方面的优势。
{"title":"Physically Explainable Deep Learning for Convective Initiation Nowcasting Using GOES-16 Satellite Observations","authors":"Da Fan, S. Greybush, E. Clothiaux, David John Gagne","doi":"10.1175/aies-d-23-0098.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0098.1","url":null,"abstract":"\u0000Convective initiation (CI) nowcasting remains a challenging problem for both numerical weather prediction models and existing nowcasting algorithms. In this study, an object-based probabilistic deep learning model is developed to predict CI based on multichannel infrared GOES-16 satellite observations. The data come from patches surrounding potential CI events identified in Multi-Radar Multi-Sensor Doppler weather radar products over the Great Plains region from June and July 2020 and June 2021. An objective radar-based approach is used to identify these events. The deep learning model significantly outperforms the classical logistic model at lead times up to 1 hour, especially on the false alarm ratio. Through case studies, the deep learning model exhibits dependence on the characteristics of clouds and moisture at multiple altitudes. Model explanation further reveals that the contribution of features to model predictions is significantly dependent on the baseline, a reference point against which the prediction is compared. Under a moist baseline, moisture gradients in the lower and middle troposphere contribute most to correct CI forecasts. In contrast, under clear-sky baselines, correct CI forecasts are dominated by cloud-top features, including cloud-top glaciation, height, and cloud coverage. Our study demonstrates the advantage of using different baselines in further understanding model behavior and gaining scientific insights.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"55 17","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141113447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-05-13DOI: 10.1175/aies-d-23-0106.1
Corey K. Potvin, Montgomery Flora, P. Skinner, Anthony E. Reinhart, B. Matilla
Forecasters routinely calibrate their confidence in model forecasts. Ensembles inherently estimate forecast confidence, but are often underdispersive, and ensemble spread does not strongly correlate with ensemble-mean error. The misalignment between ensemble spread and skill motivates new methods for “forecasting forecast skill” so that forecasters can better utilize ensemble guidance. We have trained logistic regression and random forest models to predict the skill of composite reflectivity forecasts from the NSSL Warn-on-Forecast System (WoFS), a 3-km ensemble that generates rapidly updating forecast guidance for 0-6-h lead times. The forecast skill predictions are valid at 1-h, 2-h, or 3-h lead times within localized regions determined by the observed storm locations at analysis time. We use WoFS analysis and forecast output and NSSL Multi-Radar / Multi-Sensor composite reflectivity for 106 cases from the 2017-2021 NOAA Hazardous Weather Testbed Spring Forecasting Experiments. We frame the prediction task as a multi-classification problem, where the forecast skill labels are determined by averaging the extended Fractions Skill Scores (eFSS) for several reflectivity thresholds and verification neighborhoods, then converting to one of three classes based on where the average eFSS ranks within the entire dataset: POOR (bottom 20%), FAIR (middle 60%), or GOOD (top 20%). Initial machine learning (ML) models are trained on 323 predictors; reducing to 10 or 15 predictors in the final models only modestly reduces skill. The final models substantially outperform carefully developed persistence- and spread-based models, and are reasonably explainable. The results suggest that ML can be a valuable tool for guiding user confidence in convection-allowing (and larger-scale) ensemble forecasts.
{"title":"Using machine learning to predict convection-allowing ensemble forecast skill: Evaluation with the NSSL Warn-on-Forecast System","authors":"Corey K. Potvin, Montgomery Flora, P. Skinner, Anthony E. Reinhart, B. Matilla","doi":"10.1175/aies-d-23-0106.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0106.1","url":null,"abstract":"\u0000Forecasters routinely calibrate their confidence in model forecasts. Ensembles inherently estimate forecast confidence, but are often underdispersive, and ensemble spread does not strongly correlate with ensemble-mean error. The misalignment between ensemble spread and skill motivates new methods for “forecasting forecast skill” so that forecasters can better utilize ensemble guidance. We have trained logistic regression and random forest models to predict the skill of composite reflectivity forecasts from the NSSL Warn-on-Forecast System (WoFS), a 3-km ensemble that generates rapidly updating forecast guidance for 0-6-h lead times. The forecast skill predictions are valid at 1-h, 2-h, or 3-h lead times within localized regions determined by the observed storm locations at analysis time. We use WoFS analysis and forecast output and NSSL Multi-Radar / Multi-Sensor composite reflectivity for 106 cases from the 2017-2021 NOAA Hazardous Weather Testbed Spring Forecasting Experiments. We frame the prediction task as a multi-classification problem, where the forecast skill labels are determined by averaging the extended Fractions Skill Scores (eFSS) for several reflectivity thresholds and verification neighborhoods, then converting to one of three classes based on where the average eFSS ranks within the entire dataset: POOR (bottom 20%), FAIR (middle 60%), or GOOD (top 20%). Initial machine learning (ML) models are trained on 323 predictors; reducing to 10 or 15 predictors in the final models only modestly reduces skill. The final models substantially outperform carefully developed persistence- and spread-based models, and are reasonably explainable. The results suggest that ML can be a valuable tool for guiding user confidence in convection-allowing (and larger-scale) ensemble forecasts.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"45 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140984841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-26DOI: 10.1175/aies-d-22-0046.1
Daniel Galea, Kevin Hodges, Bryan N. Lawrence
Tropical cyclones (TCs) are important phenomena; understanding their behaviour requires being able to detect their presence in simulations. Detection algorithms vary; here we compare a novel deep-learning-based detection algorithm, TCDetect, with a state-of-the-art tracking system (TRACK) and an observational dataset (IBTrACS) to provide context for potential use in climate simulations. Previous work has shown TCDetect has good recall, particularly for hurricane-strength events. The primary question addressed here is how much the structure of the systems plays a part in detection. To compare with observations of TCs, it is necessary to apply detection techniques to re-analysis. For this purpose, we use ERA-Interim, and a key part of the comparison is the recognition that ERA-Interim itself does not fully reflect the observations. Despite that limitation, both TCDetect and TRACK applied to ERA-Interim mostly agree with each other. Also, when considering only hurricane-strength TCs, TCDetect and TRACK correspond well with the TC observations from IBTrACS. Like TRACK, TCDetect has good recall for strong systems; however, it finds a significant number of false positives associated with weaker TCs (that is, events detected as having hurricane strength, but being weaker in reality) and extra-tropical storms. As TCDetect was not trained to locate TCs, a post-hoc method to perform comparisons was used. While this method was not always successful, some success in matching tracks and events in physical space was also achieved. The analysis of matches suggested the best results were found in the northern hemisphere and that in most regions the detections followed the same patterns in time no matter which detection method was used.
{"title":"Investigating differences between Tropical Cyclone detection systems","authors":"Daniel Galea, Kevin Hodges, Bryan N. Lawrence","doi":"10.1175/aies-d-22-0046.1","DOIUrl":"https://doi.org/10.1175/aies-d-22-0046.1","url":null,"abstract":"\u0000Tropical cyclones (TCs) are important phenomena; understanding their behaviour requires being able to detect their presence in simulations. Detection algorithms vary; here we compare a novel deep-learning-based detection algorithm, TCDetect, with a state-of-the-art tracking system (TRACK) and an observational dataset (IBTrACS) to provide context for potential use in climate simulations. Previous work has shown TCDetect has good recall, particularly for hurricane-strength events. The primary question addressed here is how much the structure of the systems plays a part in detection. To compare with observations of TCs, it is necessary to apply detection techniques to re-analysis. For this purpose, we use ERA-Interim, and a key part of the comparison is the recognition that ERA-Interim itself does not fully reflect the observations. Despite that limitation, both TCDetect and TRACK applied to ERA-Interim mostly agree with each other. Also, when considering only hurricane-strength TCs, TCDetect and TRACK correspond well with the TC observations from IBTrACS. Like TRACK, TCDetect has good recall for strong systems; however, it finds a significant number of false positives associated with weaker TCs (that is, events detected as having hurricane strength, but being weaker in reality) and extra-tropical storms. As TCDetect was not trained to locate TCs, a post-hoc method to perform comparisons was used. While this method was not always successful, some success in matching tracks and events in physical space was also achieved. The analysis of matches suggested the best results were found in the northern hemisphere and that in most regions the detections followed the same patterns in time no matter which detection method was used.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"109 33","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140380763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-14DOI: 10.1175/aies-d-23-0073.1
C. E. Graafland, Swen Brands, José Manuel Gutiérrez
The different phases of the Coupled Model Intercomparison Project (CMIP) provide ensembles of past, present, and future climate simulations crucial for climate change impact and adaptation activities. These ensembles are produced using multiple Global Climate Models (GCMs) from different modeling centres with some shared building blocks and inter-dependencies. Applications typically follow the ‘model democracy’ approach which might have significant implications in the resulting products (e.g. large bias and low spread). Thus, quantifying model similarity within ensembles is crucial for interpreting model agreement and multi-model uncertainty in climate change studies. The classical methods used for assessing GCM similarity can be classified into two groups. The a priori approach relies on expert knowledge about the components of these models, while the a posteriori approach seeks similarity in the GCMs’ output variables and is thus data-driven. In this study we apply Probabilistic Network Models (PNMs), a well established machine learning technique, as a new a posteriori method to measure inter-model similarities. The proposed methodology is applied to surface temperature fields of the historical experiments from the CMIP5 multi-model ensemble and different reanalysis gridded datasets. PNMs are capable to learn the complex spatial dependency structures present in climate data, including teleconnections operating on multiple spatial scales, characteristic of the underlying GCM. A distance metric building on the resulting PNMs is applied to characterize GCM model dependencies. The results of this approach are in line with those obtained with more traditional methods, but have further explanatory potential building on probabilistic model querying.
{"title":"A data-driven probabilistic network approach to assess model similarity in CMIP ensembles","authors":"C. E. Graafland, Swen Brands, José Manuel Gutiérrez","doi":"10.1175/aies-d-23-0073.1","DOIUrl":"https://doi.org/10.1175/aies-d-23-0073.1","url":null,"abstract":"\u0000The different phases of the Coupled Model Intercomparison Project (CMIP) provide ensembles of past, present, and future climate simulations crucial for climate change impact and adaptation activities. These ensembles are produced using multiple Global Climate Models (GCMs) from different modeling centres with some shared building blocks and inter-dependencies. Applications typically follow the ‘model democracy’ approach which might have significant implications in the resulting products (e.g. large bias and low spread). Thus, quantifying model similarity within ensembles is crucial for interpreting model agreement and multi-model uncertainty in climate change studies. The classical methods used for assessing GCM similarity can be classified into two groups. The a priori approach relies on expert knowledge about the components of these models, while the a posteriori approach seeks similarity in the GCMs’ output variables and is thus data-driven. In this study we apply Probabilistic Network Models (PNMs), a well established machine learning technique, as a new a posteriori method to measure inter-model similarities. The proposed methodology is applied to surface temperature fields of the historical experiments from the CMIP5 multi-model ensemble and different reanalysis gridded datasets. PNMs are capable to learn the complex spatial dependency structures present in climate data, including teleconnections operating on multiple spatial scales, characteristic of the underlying GCM. A distance metric building on the resulting PNMs is applied to characterize GCM model dependencies. The results of this approach are in line with those obtained with more traditional methods, but have further explanatory potential building on probabilistic model querying.","PeriodicalId":94369,"journal":{"name":"Artificial intelligence for the earth systems","volume":"37 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140244797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：