Search for: All records

Harmful algal blooms (HABs) are increasing in frequency in the Chesapeake Bay, posing risks to marine ecosystems, water quality, and public health. Chlorophyll-a (Chl-a) is a widely used indicator of algal biomass, and satellite observations such as Sentinel-3 Ocean and Land Color Instrument (OLCI) enable large-scale monitoring of bloom dynamics. However, cloud cover and atmospheric interference frequently introduce missing pixels in daily satellite products, reducing temporal continuity and limiting monitoring reliability. Satellite-derived chlorophyll-a (Chl-a) data exhibit substantial missingness, with daily pixel gaps ranging from approximately 52.30% to 100% (mean ≈ 88.95%). This study evaluates spatial interpolation, EOF-based, supervised machine-learning, deep-learning, and convolutional autoencoder approaches for reconstructing missing Chl-a values. Sentinel-3 OLCI Chl-a data from 2023–2024 were used for model training, while data from 2025 served as a temporally independent test set to avoid spatiotemporal leakage. To simulate cloud-induced data gaps, artificial missingness scenarios ranging from 50% to 90% were applied for the Inverse Distance Weighting (IDW) and Data Interpolating Empirical Orthogonal Functions (DINEOF) baseline approaches, while machine-learning, deep-learning, and convolutional autoencoder models were evaluated using real satellite-derived missing observations. The evaluated models include IDW, DINEOF, K-Nearest Neighbors (KNN), Random Forest (RF), Extra Trees (ET), XGBoost, a Long Short-Term Memory (LSTM) network, and a Temporal Data Interpolating Convolutional Autoencoder (Temporal DINCAE). Model performance was assessed using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), prediction bias, and the coefficient of determination (R2). Results indicate that tree-based ensemble models outperform spatial interpolation and EOF-based methods, with XGBoost achieving the best overall performance (R2 ≈ 0.86; RMSE ≈ 9.61 mg m−3). The LSTM model achieved lower prediction errors (RMSE ≈ 5.87 mg m−3; MAE ≈ 2.16 mg m−3), highlighting the benefit of incorporating temporal dependencies, although with slightly reduced variance capture. The convolutional autoencoder-based Temporal DINCAE model achieved strong reconstruction performance (R2 ≈ 0.84; RMSE ≈ 11.15 mg m−3). Uncertainty quantification shows that Extra Trees tends to underestimate uncertainty with narrower prediction intervals, whereas XGBoost provides better-calibrated but wider intervals. 

Breathing in fine particulate matter of diameter less than 2.5 µm (PM2.5) greatly increases an individual’s risk of cardiovascular and respiratory diseases. As climate change progresses, extreme weather events, including wildfires, are expected to increase, exacerbating air pollution. However, models often struggle to capture extreme pollution events due to the rarity of high PM2.5 levels in training datasets. To address this, we implemented cluster-based undersampling and trained Transformer models to improve extreme event prediction using various cutoff thresholds (12.1 µg/m3 and 35.5 µg/m3) and partial sampling ratios (10/90, 20/80, 30/70, 40/60, 50/50). Our results demonstrate that the 35.5 µg/m3 threshold, paired with a 20/80 partial sampling ratio, achieved the best performance, with an RMSE of 2.080, MAE of 1.386, and R2 of 0.914, particularly excelling in forecasting high PM2.5 events. Overall, models trained on augmented data significantly outperformed those trained on original data, highlighting the importance of resampling techniques in improving air quality forecasting accuracy, especially for high-pollution scenarios. These findings provide critical insights into optimizing air quality forecasting models, enabling more reliable predictions of extreme pollution events. By advancing the ability to forecast high PM2.5 levels, this study contributes to the development of more informed public health and environmental policies to mitigate the impacts of air pollution, and advanced the technology for building better air quality digital twins. 

Many previous studies have shown that open-source technologies help democratize information and foster collaborations to enable addressing global physical and societal challenges. The outbreak of the novel coronavirus has imposed unprecedented challenges to human society. It affects every aspect of livelihood, including health, environment, transportation, and economy. Open-source technologies provide a new ray of hope to collaboratively tackle the pandemic. The role of open source is not limited to sharing a source code. Rather open-source projects can be adopted as a software development approach to encourage collaboration among researchers. Open collaboration creates a positive impact in society and helps combat the pandemic effectively. Open-source technology integrated with geospatial information allows decision-makers to make strategic and informed decisions. It also assists them in determining the type of intervention needed based on geospatial information. The novelty of this paper is to standardize the open-source workflow for spatiotemporal research. The highlights of the open-source workflow include sharing data, analytical tools, spatiotemporal applications, and results and formalizing open-source software development. The workflow includes (i) developing open-source spatiotemporal applications, (ii) opening and sharing the spatiotemporal resources, and (iii) replicating the research in a plug and play fashion. Open data, open analytical tools and source code, and publicly accessible results form the foundation for this workflow. This paper also presents a case study with the open-source spatiotemporal application development for air quality analysis in California, USA. In addition to the application development, we shared the spatiotemporal data, source code, and research findings through the GitHub repository. 

The outbreak of COVID-19 from late 2019 not only threatens the health and lives of humankind but impacts public policies, economic activities, and human behavior patterns significantly. To understand the impact and better prepare for future outbreaks, socioeconomic factors play significant roles in (1) determinant analysis with health care, environmental exposure and health behavior; (2) human mobility analyses driven by policies; (3) economic pressure and recovery analyses for decision making; and (4) short to long term social impact analysis for equity, justice and diversity. To support these analyses for rapid impact responses, state level socioeconomic factors for the United States of America (USA) are collected and integrated into topic-based indicators, including (1) the daily quantitative policy stringency index; (2) dynamic economic indices with multiple time frequency of GDP, international trade, personal income, employment, the housing market, and others; (3) the socioeconomic determinant baseline of the demographic, housing financial situation and medical resources. This paper introduces the measurements and metadata of relevant socioeconomic data collection, along with the sharing platform, data warehouse framework and quality control strategies. Different from existing COVID-19 related data products, this collection recognized the geospatial and dynamic factor as essential dimensions of epidemiologic research and scaled down the spatial resolution of socioeconomic data collection from country level to state level of the USA with a standard data format and high quality.