A5024715871- Precipitation Measurement and Analysis
- Meteorological Phenomena and Simulations
- Air Quality and Health Impacts
- Climate variability and models
- Air Quality Monitoring and Forecasting
- Fire effects on ecosystems
- Hydrological Forecasting Using AI
- Impact of Light on Environment and Health
- Remote Sensing in Agriculture
- Flood Risk Assessment and Management
- Neural Networks and Applications
- Quantum Computing Algorithms and Architecture
- Gaussian Processes and Bayesian Inference
- Anomaly Detection Techniques and Applications
- COVID-19 impact on air quality
- Statistical Mechanics and Entropy
- Coral and Marine Ecosystems Studies
- Tropical and Extratropical Cyclones Research
- Advanced Computational Techniques and Applications
- Fire Detection and Safety Systems
- Underwater Acoustics Research
- Irrigation Practices and Water Management
- Soil Moisture and Remote Sensing
- Hydrology and Drought Analysis
- Geological Modeling and Analysis
Universities Space Research Association
2019-2024
Ames Research Center
2021-2024
Research Institute for Advanced Computer Science
2022-2024
University of California, Irvine
2017-2020
Center for Hydrometeorology and Remote Sensing
2017-2020
Irvine University
2020
Samueli Institute
2017-2020
Abstract Reservoirs are fundamental human‐built infrastructures that collect, store, and deliver fresh surface water in a timely manner for many purposes. Efficient reservoir operation requires policy makers operators to understand how inflows changing under different hydrological climatic conditions enable forecast‐informed operations. Over the last decade, uses of Artificial Intelligence Data Mining [AI & DM] techniques assisting streamflow subseasonal seasonal forecasts have been...
Abstract Short‐term Quantitative Precipitation Forecasting is important for flood forecasting, early warning, and natural hazard management. This study proposes a precipitation forecast model by extrapolating Cloud‐Top Brightness Temperature (CTBT) using advanced Deep Neural Networks, applying the forecasted CTBT into an effective rainfall retrieval algorithm to obtain (0–6 hr). To achieve such tasks, we propose Long Short‐Term Memory (LSTM) Estimation from Remotely Sensed Information...
Abstract Accurate and timely precipitation estimates are critical for monitoring forecasting natural disasters such as floods. Despite having high-resolution satellite information, estimation from remotely sensed data still suffers methodological limitations. State-of-the-art deep learning algorithms, renowned their skill in accurate patterns within large complex datasets, appear well suited to the task of estimation, given ample amount data. In this study, effectiveness applying...
Abstract Adaptation is key to minimizing heatwaves' societal burden; however, our understanding of adaptation capacity across the socioeconomic spectrum incomplete. We demonstrate that observed heatwave trends in past four decades were most pronounced lowest‐quartile income region world resulting >40% higher exposure from 2010 2019 compared highest‐quartile region. Lower‐income regions have reduced adaptative warming, which compounds impacts exposure. also show individual contiguous...
High-resolution real-time satellite-based precipitation estimation datasets can play a more essential role in flood forecasting and risk analysis of infrastructures. This is particularly true for extended deserts or mountainous areas with sparse rain gauges like Iran. However, there are discrepancies between these estimations ground measurements, it necessary to apply adjustment methods reduce systematic bias products. In this study, we quantile mapping method gauge information the error...
Abstract Sea‐level rise (SLR) increasingly threatens coastal communities around the world. However, not all are equally threatened, and realistic estimation of hazard is difficult. Understanding SLR impacts on extreme sea level challenging due to interactions between multiple tidal non‐tidal flood drivers. We here use global hourly data show how why tides surges interact with mean (MSL) fluctuations. At most locations world, amplitude at least one constituent and/or residual have changed in...
The ability of the seven CMIP5 models to simulate extreme precipitation events over Iran was evaluated using Precipitation Estimation from Remotely Sensed Information Artificial Neural Networks–Climate Data Record (PERSIANN‐CDR) data set. criterion used select availability historical daily for retrospective period 1983–2005, as well future projections three representative concentration pathways emission scenarios (RCP2.6, RCP4.5, and RCP8.5) spatial resolution higher than 2 × 2°. This is...
Abstract In the wake of climate change, extreme events such as heatwaves are considered to be key players in terrestrial biosphere. past decades, frequency and severity have risen substantially, they projected continue intensify future. One question is therefore: how do changes affect carbon cycle? Although soil respiration (Rs) second largest contributor cycle, impacts on Rs not been fully understood. Using a unique set continuous high in-situ measurements from our field site, we...
Variability and spatiotemporal changes in precipitation characteristics can have profound socioenvironmental impacts. Several studies shown that the frequency and/or magnitude of events changed over contiguous United States (CONUS) past decades. Most previous used only one dataset investigated mean or extreme precipitation. Here, using 6 gridded daily datasets, we show there are substantial discrepancies both non-extreme from 1983 to 2017. Our results highlight a single record study...
Wildfires are one of the major disasters among many and responsible for more than 6 million acres burned in United States alone every year. Accurate, insightful, timely wildfire detection is needed to help authorities mitigate prevent further destruction. Uncertainty quantification always a crucial part natural disasters, such as wildfires, modeling products can be misinterpreted without proper uncertainty quantification. In this study, we propose supervised deep generative machine-learning...
Spatiotemporal precipitation trend analysis provides valuable information for water management decision-making. Satellite-based products with high spatial and temporal resolution long records, as opposed to temporally spatially sparse rain gauge networks, are a suitable alternative analyze trends over Iran. This study analyzes the in annual, seasonal, monthly along contribution of each season month annual Iran 1983–2018 period. For analyses, Mann–Kendall test is applied Precipitation...
Identifying safety anomalies and vulnerabilities in the aviation domain is a very expensive time-consuming task. Currently, it accomplished via manual forensic reviews by subject matter experts (SMEs). However, with increase amount of data produced airspace operations, relying on such impractical. Automated approaches, as exceedance detection, have been deployed to flag events which surpass pre-defined threshold. These however, completely rely knowledge outcome SMEs’ can only identify purely...
Providing reliable long-term global precipitation records at high spatial and temporal resolutions is crucial for climatological studies. Satellite-based estimations are a promising alternative to rain gauges providing homogeneous information. Most satellite-based products suffer from short-term data records, which make them unsuitable various hydrological applications. However, Precipitation Estimation Remotely Sensed Information using Artificial Neural Networks-Climate Data Record...
Abstract Recent developments in “headline-making” deep neural networks (DNNs), specifically convolutional (CNNs), along with advancements computational power, open great opportunities to integrate massive amounts of real-time observations characterize spatiotemporal structures surface precipitation. This study aims develop a CNN algorithm, named Deep Neural Network High Spatiotemporal Resolution Precipitation Estimation (Deep-STEP), that ingests direct satellite passive microwave (PMW)...
Abstract Deep generative learning cannot only be used for generating new data with statistical characteristics derived from input but also anomaly detection, by separating nominal and anomalous instances based on their reconstruction quality. In this paper, we explore the performance of three unsupervised deep models—variational autoencoders (VAEs) Gaussian, Bernoulli, Boltzmann priors—in detecting anomalies in multivariate time series commercial-flight operations. We created two VAE models...
High-spatial-resolution air quality (AQ) mapping is important for identifying pollution sources to facilitate local action. Some of the most populated cities in world are not equipped with infrastructure required monitor AQ levels on ground and must rely other sources, such as satellite derived estimates, AQ. Current satellite-data-based models provide a kilometer scale at best. In this study, we focus producing hundred-meter-scale maps urban environments developed cities. We examined...
High spatial resolution information on urban air pollution levels is unavailable in many areas globally, partially due to the high input data needs of existing estimation approaches. We introduced a computer vision method estimate annual means for from street-level images. used mean estimates NO2 and PM2.5 concentrations locally calibrated models as labels London, New York, Vancouver allow compilation sufficiently large dataset (~250 k images each city). Our experimental setup designed...
Coral reefs are one of the most biologically complex and diverse ecosystems within shallow marine environment. Unfortunately, these underwater threatened by a number anthropogenic challenges, including ocean acidification warming, overfishing, continued increase debris in oceans. This requires comprehensive assessment world's coastal environments, quantitative analysis on health extent coral other associated species, as vital Earth Science measurement. However, limitations observational...
Urban air pollution is a critical public health challenge in low-and-middle-income countries (LMICs). At the same time, LMICs tend to be data-poor, lacking adequate infrastructure monitor quality (AQ). As undergo rapid urbanization, socio-economic burden of poor AQ will immense. Here we present globally scalable two-step deep learning (DL) based approach for estimation LMIC cities that mitigates need extensive on ground. We train DL model can map satellite imagery high-income (HICs) with...
Forecasting chaotic systems is a notably complex task, which in recent years has been approached with reasonable success using reservoir computing (RC), recurrent network fixed random weights (the reservoir) used to extract the spatio-temporal information of system. This work presents hybrid quantum reservoir-computing (HQRC) framework, replaces RC circuit. The modular structure and measurement feedback circuit are encode system dynamics states, from classical learning performed predict...
Wildfire occurrences have been increasing for the past decade, leaving devastating traces across world. In recent ef-forts, remote sensing and airborne missions utilized to better understand manage wildfires. This has resulted in an exponential increase volume of data, which pushed need intelligent automation data extraction wildfire studies. Machine learning offers accu-rate detecting such natural anomalies en-able decision-makers take actions a timely manner. Re-cent advances machine...
Landform mapping (also referred to as geomorphology or geomorphometry) can be divided into two domains: general and specific (Evans 2012). Whereas landform categorizes all elements of the study area classes, such ridges, valleys, peaks, depressions, landforms requires delineation (even if fuzzy) individual landforms. The former is mainly driven by physical properties elevation, slope, curvature.  The latter, however, must consider cognitive (human) reasoning that discriminates in...