Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
37
result(s) for
"Lakshmanan, Valliappa"
Sort by:
The History and Practice of AI in the Environmental Sciences
Artificial intelligence (AI) and machine learning (ML) have become important tools for environmental scientists and engineers, both in research and in applications. Although these methods have become quite popular in recent years, they are not new. The use of AI methods began in the 1950s and environmental scientists were adopting them by the 1980s. Although an “AI winter” temporarily slowed the growth, a more recent resurgence has brought it back with gusto. This paper tells the story of the evolution of AI in the field through the lens of the AMS Committee on Artificial Intelligence Applications to Environmental Science. The environmental sciences possess a host of problems amenable to advancement by intelligent techniques. We review a few of the early applications along with the ML methods of the time and how their progression has impacted these sciences. While AI methods have changed from expert systems in the 1980s to neural networks and other data-driven methods, and more recently deep learning, the environmental problems tackled have remained similar. We discuss the types of applications that have shown some of the biggest advances due to AI usage and how they have evolved over the past decades, including topics in weather forecasting, probabilistic prediction, climate estimation, optimization problems, image processing, and improving forecasting models. We finish with a look at where AI as employed in environmental science appears to be headed and some thoughts on how it might be best blended with physical/dynamical modeling approaches to further advance our science.
Journal Article
Improved Nowcasts by Blending Extrapolation and Model Forecasts
Planning and managing commercial airplane routes to avoid thunderstorms requires very skillful and frequently updated 0–8-h forecasts of convection. The National Oceanic and Atmospheric Administration’s High-Resolution Rapid Refresh (HRRR) model is well suited for this purpose, being initialized hourly and providing explicit forecasts of convection out to 15 h. However, because of difficulties with depicting convection at the time of model initialization and shortly thereafter (i.e., during model spinup), relatively simple extrapolation techniques, on average, perform better than the HRRR at 0–2-h lead times. Thus, recently developed nowcasting techniques blend extrapolation-based forecasts with numerical weather prediction (NWP)-based forecasts, heavily weighting the extrapolation forecasts at 0–2-h lead times and transitioning emphasis to the NWP-based forecasts at the later lead times. In this study, a new approach to applying different weights to blend extrapolation and model forecasts based on intensities and forecast times is applied and tested. An image-processing method of morphing between extrapolation and model forecasts to create nowcasts is described and the skill is compared to extrapolation forecasts and forecasts from the HRRR. The new approach is called salient cross dissolve (Sal CD), which is compared to a commonly used method called linear cross dissolve (Lin CD). Examinations of forecasts and observations of the maximum altitude of echo-top heights ≥18 dBZ and measurement of forecast skill using neighborhood-based methods shows that Sal CD significantly improves upon Lin CD, as well as the HRRR at 2–5-h lead times.
Journal Article
An Automated Technique to Quality Control Radar Reflectivity Data
Echoes in radar reflectivity data do not always correspond to precipitating particles. Echoes on radar may result from biological targets such as insects, birds, or wind-borne particles; from anomalous propagation or ground clutter; or from test and interference patterns that inadvertently seep into the final products. Although weather forecasters can usually identify and account for the presence of such contamination, automated weather-radar algorithms are drastically affected. Several horizontal and vertical features have been proposed to discriminate between precipitation echoes and echoes that do not correspond to precipitation. None of these features by themselves can discriminate between precipitating and nonprecipitating areas. In this paper, a neural network is used to combine the individual features, some of which have already been proposed in the literature and some of which are introduced in this paper, into a single discriminator that can distinguish between “good” and “bad” echoes (i.e., precipitation and nonprecipitation, respectively). The method of computing the horizontal features leads to statistical anomalies in their distributions near the edges of echoes. Also described is how to avoid presenting such range gates to the neural network. The gate-by-gate discrimination provided by the neural network is followed by more holistic postprocessing based on spatial contiguity constraints and object identification to yield quality-controlled radar reflectivity scans that have most of the bad echo removed while leaving most of the good echo untouched. A possible multisensor extension, utilizing satellite data and surface observations, to the radar-only technique is also demonstrated. It is demonstrated that the resulting technique is highly skilled and that its skill exceeds that of the currently operational algorithm.
Journal Article
A Real-Time Weather-Adaptive 3DVAR Analysis System for Severe Weather Detections and Warnings
A real-time, weather-adaptive three-dimensional variational data assimilation (3DVAR) system has been adapted for the NOAA Warn-on-Forecast (WoF) project to incorporate all available radar observations within a moveable analysis domain. The key features of the system include 1) incorporating radar observations from multiple Weather Surveillance Radars-1988 Doppler (WSR-88Ds) with NCEP forecast products as a background state, 2) the ability to automatically detect and analyze severe local hazardous weather events at 1-km horizontal resolution every 5 min in real time based on the current weather situation, and 3) the identification of strong circulation patterns embedded in thunderstorms. Although still in the early development stage, the system performed very well within the NOAA's Hazardous Weather Testbed (HWT) Experimental Warning Program during preliminary testing in spring 2010 when many severe weather events were successfully detected and analyzed. This study represents a first step in the assessment of this type of 3DVAR analysis for use in severe weather warnings. The eventual goal of this real-time 3DVAR system is to help meteorologists better track severe weather events and eventually provide better warning information to the public, ultimately saving lives and reducing property damage.
Journal Article
A Real-Time, Three-Dimensional, Rapidly Updating, Heterogeneous Radar Merger Technique for Reflectivity, Velocity, and Derived Products
With the advent of real-time streaming data from various radar networks, including most Weather Surveillance Radars-1988 Doppler and several Terminal Doppler Weather Radars, it is now possible to combine data in real time to form 3D multiple-radar grids. Herein, a technique for taking the base radar data (reflectivity and radial velocity) and derived products from multiple radars and combining them in real time into a rapidly updating 3D merged grid is described. An estimate of that radar product combined from all the different radars can be extracted from the 3D grid at any time. This is accomplished through a formulation that accounts for the varying radar beam geometry with range, vertical gaps between radar scans, the lack of time synchronization between radars, storm movement, varying beam resolutions between different types of radars, beam blockage due to terrain, differing radar calibration, and inaccurate time stamps on radar data. Techniques for merging scalar products like reflectivity, and innovative, real-time techniques for combining velocity and velocity-derived products are demonstrated. Precomputation techniques that can be utilized to perform the merger in real time and derived products that can be computed from these three-dimensional merger grids are described.
Journal Article
An Objective Method of Evaluating and Devising Storm-Tracking Algorithms
Although storm-tracking algorithms are a key ingredient of nowcasting systems, evaluation of storm-tracking algorithms has been indirect, labor intensive, or nonspecific. A set of easily computable bulk statistics that can be used to directly evaluate the performance of tracking algorithms on specific characteristics is introduced. These statistics are used to evaluate five widely used storm-tracking algorithms on a diverse set of radar reflectivity data cases. Based on this objective evaluation, a storm-tracking algorithm is devised that performs consistently and better than any of the previously suggested techniques.
Journal Article
Examination of a Real-Time 3DVAR Analysis System in the Hazardous Weather Testbed
Forecasters and research meteorologists tested a real-time three-dimensional variational data assimilation (3DVAR) system in the Hazardous Weather Testbed during the springs of 2010–12 to determine its capabilities to assist in the warning process for severe storms. This storm-scale system updates a dynamically consistent three-dimensional wind field every 5 min, with horizontal and average vertical grid spacings of 1 km and 400 m, respectively. The system analyzed the life cycles of 218 supercell thunderstorms on 27 event days during these experiments, producing multiple products such as vertical velocity, vertical vorticity, and updraft helicity. These data are compared to multiradar–multisensor data from the Warning Decision Support System–Integrated Information to document the performance characteristics of the system, such as how vertical vorticity values compare to azimuthal shear fields calculated directly from Doppler radial velocity. Data are stratified by range from the nearest radar, as well as by the number of radars entering into the analysis of a particular storm. The 3DVAR system shows physically realistic trends of updraft speed and vertical vorticity for a majority of cases. Improvements are needed to better estimate the near-surface winds when no radar is nearby and to improve the timeliness of the input data. However, the 3DVAR wind field information provides an integrated look at storm structure that may be of more use to forecasters than traditional radar-based proxies used to infer severe weather potential.
Journal Article
Data Mining Storm Attributes from Spatial Grids
A technique to identify storms and capture scalar features within the geographic and temporal extent of the identified storms is described. The identification technique relies on clustering grid points in an observation field to find self-similar and spatially coherent clusters that meet the traditional understanding of what storms are. From these storms, geometric, spatial, and temporal features can be extracted. These scalar features can then be data mined to answer many types of research questions in an objective, data-driven manner. This is illustrated by using the technique to answer questions of forecaster skill and lightning predictability.
Journal Article
An Efficient, General-Purpose Technique for Identifying Storm Cells in Geospatial Images
Existing techniques for identifying, associating, and tracking storms rely on heuristics and are not transferrable between different types of geospatial images. Yet, with the multitude of remote sensing instruments and the number of channels and data types increasing, it is necessary to develop a principled and generally applicable technique. In this paper, an efficient, sequential, morphological technique called the watershed transform is adapted and extended so that it can be used for identifying storms. The parameters available in the technique and the effects of these parameters are also explained. The method is demonstrated on different types of geospatial radar and satellite images. Pointers are provided on the effective choice of parameters to handle the resolutions, data quality constraints, and dynamic ranges found in observational datasets.
Journal Article