This study presents a novel, integrated modeling framework that combines machine learning (ML) techniques with physics-based approaches to forecast both CO2 emissions and global temperature anomalies. Unlike prior research that typically addresses these components in isolation, this work concurrently applies and compares five advanced ML models—Long Short-Term Memory (LSTM), Extreme Gradient Boosting

Large-scale solar promises a low-carbon energy alternative. However, solar production in North America given anticipated climate change has been studied only seasonally in terms of solar irradiance. This work integrates more of the predictive potential of climate-change models by exploring other environmental variables, such as humidity and temperature. Here, a Continental US (CONUS) model is produced

Social media usage surges during natural disasters, offering critical insights into public sentiment and needs. This study leverages artificial intelligence (AI) and advanced natural language processing (NLP) techniques to analyze Twitter (X) data from the 2018 California Camp Fire. By combining sentiment analysis, emotion classification, and humanitarian topic classification, we provide a nuanced

Machine-learning algorithms have been employed in river ice research for flood estimation. This study aimed to introduce a machine learning-based model for predicting ice jam floods. An ice-jam dataset was created using a stochastic modelling approach in which thousands of possible scenarios were simulated. This approach integrated a hydrodynamic model, RIVICE, into a Monte Carlo Analysis (MOCA) framework

Recent studies have shown that LSTM performs well in runoff prediction in large sample regional modeling and can estimate hydrological concepts based on its internal information. However, compared to process-based models, it still produces erroneous predictions that violate the physical laws. To explore the reasons for the above phenomenon, this study analyzes the evolution of LSTM's performance in

Accurately simulating soil organic carbon (SOC) dynamics is essential for carbon-related assessments. Process-oriented SOC models employ temperature (f(T)) and soil moisture (f(W)) response functions derived from specific conditions to simulate SOC responses to climate change, yet are widely applied in regional and global-scale studies. How these functions affect regional SOC simulations remains unclear

Farming is increasingly data-driven, leveraging high-frequency and precision data from IoT devices, sensors, and remote tools. Effective data collection, organization, and management are essential to link datasets with agronomic details, forming the foundation for predictive models. These models, using AI and machine learning, optimize decision-making, forecast crop yields, predict pest outbreaks,

Urban flood forecasting is crucial for timely public warnings and effective flood management. Traditional mechanistic models face challenges such as high computational costs and limited real-time capabilities. Recent advancements in Artificial Intelligence (AI), including machine learning (ML), deep learning (DL), and large language models (LLMs), address these limitations by improving data handling

The rainfall-runoff relationship frequently undergoes changes and exhibits a non-stationary state due to the impacts of climate and human activities. This non-stationarity often results in performance degradation of most existing runoff prediction models, which were designed and applied under the assumption of a stationary rainfall-runoff relationship. This study proposes an adaptive rainfall-runoff

The efficacy of a deep learning error-updating model in predicting the hydrological model-simulated errors influenced by reservoir regulation is assessed. Two daily discharge forecasting model frameworks without (Case I) and with (Case II) error-updating of the discharge forecasts at a downstream location are developed. The best discharge forecasts are forced as inputs to a hydrodynamic model to simulate

Prescribed fires are planned to achieve conservation and fuel reduction objectives while minimizing smoke ground concentration to limit health impacts and road visibility impairment. Prescribed burns cannot indeed be conducted if those hazards are not within predefined limits. This paper proposes a new framework to evaluate road visibility that overcomes the limitation of the state of the art model

A stochastic generator for rainfall is built from a Hawkes process, which is modeling the occurrence and serial correlation of non-zero rainfall values. Hawkes processes are suited to model intermittent signals, which is the case of rainfall at a fine enough observation frequency. This Hawkes process has a two-scale intensity function accounting for two orders of clustering in rainfall time series

Natural environmental systems and human activities are deeply interconnected, especially in agriculture. Despite advancements in agricultural techniques, weather remains a critical factor influencing crop yields and livestock health. Precision agriculture relies on weather predictions to mitigate environmental risks caused by weather. However, numerical weather predictions are generated by global or

Identifying rainfall-runoff events is routinely performed in many hydrologic applications. Absence of a ground-based truth makes rainfall-runoff event identification largely subjective. As a result, current algorithms often disagree on the start and end of events, leading to events within a given set of rainfall and runoff time-series with inconsistent properties – referred to hereafter as ‘uncertainty

The unique freeze‒thaw cycle in cold regions complicates irrigation. Field monitoring and experiments simulated the water cycle during thawing and growing periods, analyzing hydraulic connections. This led to coupling a hydrological balance model, the Environmental Policy Integrated Climate (EPIC) model, and a carbon emission model into a multi-objective optimization framework for rice irrigation,

The number of web portals and online tools to support or inform decision-making on environmental and climate issues has grown steadily in recent decades. This paper explores the benefits and challenges of applying user-centred design (UCD) in environmental tool development, drawing on three case studies at the science-policy interface. We examine the roles and perspectives of scientists, funders, software

Hazard maps for shallow landslides at the basin or regional scale often provide information solely about past events and/or potential source areas. Despite the availability of several propagation assessment software tools, runout maps for potential shallow landslides at the basin scale remain scarce.

As network science emerges as a transformative tool in the ‘Big Science’ era, this study harnesses this tool to model ground-level ozone distribution dynamics across US states under different regulatory frameworks from 1980 to 2017. The evolution of these regulations provides a unique natural experiment to analyze how network-driven models evolve amidst varied environmental policies. By constructing

Assessing agroecological similarity is crucial for shaping sustainable agricultural practices and resource allocation, especially in regions undergoing rapid environmental changes. Current evaluation methods face challenges such as managing large datasets, adjusting for temporal variations across locations, and the need for accessible, comprehensive analytical tools. Addressing these challenges, this

The prediction of water quality parameters is vital for sustainable aquaculture. Dissolved oxygen (DO), a key factor influencing the health and growth of aquatic organisms, is challenging to predict due to its non-linearity and significant time lag. This study proposed a DO time-series prediction model based on Transformer architecture. A dynamic interpretable time-series decomposition strategy was

Visual coverage estimation of seagrass for ground truth verification is one of the most critical aspects of marine ecosystem monitoring programs worldwide. It has traditionally been an arduous and tedious task. Commonly used tools like a scuba diver and underwater video transects require manual investigation by domain experts to assess seagrass status. Supervised machine learning methods have had a

Multi-objective optimization (MOO) is becoming increasingly important in environmental decision making, but interpreting highly-dimensional Pareto optimal data often constitutes a cognitive overload for both scientists and stakeholders. To address this challenge, we present PyretoClustR, a modular framework for post-processing Pareto optimal solutions. This tool aims to increase accessibility and applicability

This article compares novel and existing uncertainty quantification approaches for semantic segmentation used in remote sensing applications. We compare the probability estimates produced by a neural network with Monte Carlo dropout-based approaches, including predictive entropy and mutual information, and conformal prediction-based approaches, including feature conformal prediction (FCP) and a novel

This work presents a tool that enhances the hydrological flood modeling process at the event scale by integrating geospatial analysis capabilities, hydrological algorithms, and data. The main purpose is to overcome some of the main simplifications made in many modeling flood hydrographs, contributing to better simulate peak flow hydrographs for fixed return period (i.e., design flood). By leveraging

PyDDC is a particle tracking reservoir simulator capable of solving non-linear density driven convection of single phase carbon-dioxide (CO2)–brine fluid mixture in saturated porous media at the continuum scale. In contrast to the sate-of-the-art Eulerian models, PyDDC uses a Lagrangian approach to simulate the Fickian transport of single phase solute mixtures. This introduces additional flexibility

Modelling urban floods is essential for disaster prevention, yet it faces limitations in accuracy due to technical, operational, and functional constraints. The study employs a primary market research analysis to explore the perspectives of both academic and non-academic experts in urban flood modelling (UFM). Identified issues include inadequate spatial and temporal model resolution, high data requirements

The allocation of river assimilative capacity (RAC) remains a complex challenge due to the trade-offs between environmental fairness and efficiency. To address these issues, a novel integrated framework for the optimal allocation of RAC was proposed. The Luan River Basin in Chengde City, China, was selected as a case study. Hydrodynamic and advection-dispersion modules from the MIKE 11 model were implemented

It is important for both urban residents and government agencies to accurately predict the concentration of fine particulate matter (PM2.5) in the atmosphere. In existing research, various traditional and hybrid network models have been applied and developed, all of which have played a positive role in the prediction of PM2.5 concentration. Despite Transformer-based networks demonstrating unique advantages

The Chemistry Evaluation and Diagnostics Package (ChemDyg) is an open-source tool designed for the Energy Exascale Earth System Model (E3SM) developed by the U.S. Department of Energy. ChemDyg facilitates routine evaluation, tailored development, and in-depth analysis of atmospheric chemistry through its modular architecture, allowing users to compare model outputs with observational data. Version

Deep learning (DL) models have been used for rapid assessments of environmental phenomena like mapping compound flood hazards from cyclones. However, predicting compound flood dynamics (e.g., flood extent and inundation depth over time) is often done with physically-based models because they capture physical drivers, nonlinear interactions, and hysteresis in system behavior. Here, we show that a customized

The entropy weight coefficient method is widely used to aggregate sustainability indicators into a composite index. Although the method itself is conceptually straightforward, its implementation, visualization, and analysis require extensive manual interaction, is highly time-consuming, and is prone to errors. To address these challenges, this study developed an interactive software tool called the

Continuous monitoring data required for performing environmental model simulations using gridded land surface models (LSMs) are often difficult to obtain and manage, making the modeling process challenging and prone to error. In response, this study focuses on automated retrieval and processing of digital elevation models (DEMs from Google Earth Engine (GEE)), meteorologic drivers of hydrology, and

Traditional methods fall short in providing the necessary spatiotemporal resolutions for real-time urban flood forecasting, advancements suggest that surveillance camera image-based techniques are promising alternatives. However, most studies focus on static images, overlooking the impact of moving objects. To address this gap, we introduce the ResNet (Residual Network)-LSTM (Long Short-Term Memory)

Advancements in distributed hydrological modeling require higher temporal and spatial resolutions, increasing the demand for high-performance computing. Runoff-routing models face inefficiencies due to upstream–downstream dependencies. Increasing threads reduce computing time but lower efficiency due to task imbalances. We propose a stepwise spatial–temporal–multimember domain decomposition method

This study presents an approach for deriving stream geometry from watershed morphometric characteristics, addressing data scarcity for hydrologic and hydraulic modeling. The approach was tested in Puerto Rico and involves subdividing the study area into homogeneous regions and developing region-specific stream geometry predictive equations using ordinary least squares regression to correlate morphometric

Contemporary continental and global-scale digital elevation models (DEMs) are not a composite topobathymetric surface, as they tend to lack bathymetry. In this study, we analyzed if continental- and global-scale DEMs can be improved using a global hydrologic model and simple steady-state hydraulic techniques. We used two DEM datasets, globally available streamflow estimates, and simple hydraulic and

This study pioneers the application of a Bayesian-optimized multilayer perceptron (MLP) framework to predict the complete breakthrough curve (BTC) in two conduits under various flow conditions, unlike prior research that predicted only partial BTC. MLP shows significant advances in BTC prediction accuracy compared with Random Forest and Support Vector Regression. The transient storage model then simulates

Non-perennial streams constitute over half the world's stream miles but are not commonly included in streamflow monitoring networks. As a result, Stream Temperature, Intermittency, and Conductivity (STIC) loggers are widely used for characterizing flow presence or absence in non-perennial streams. To facilitate ‘FAIR’ (findable, accessible, interoperable, and reusable) stream intermittency science

This study addresses the lack of standardized evaluation criteria for the DeNitrification–DeComposition (DNDC) model, widely used to assess greenhouse gas emissions in agricultural systems. Based on a comprehensive analysis of literature data, we propose a set of benchmarks to improve the model's reliability, focusing on crop yield, soil organic carbon (SOC), nitrous oxide (N2O), and methane (CH4)

A vine copula is a flexible method for multivariate dependence simulations that assumes stationarity. However, only a few studies have focused on non-stationarity and comprehensively developed nonstationary vine copula functions. In this study, a novel R package, NSVineCopula was developed and presented. Canonical-vine and Drawable-vine structure with 36 bivariate copula functions were considered in

This study conducted a bivariate nonstationary frequency analysis utilizing rainfall events to capture the multidimensional nature of rainfall phenomena and rainfall pattern variability in South Korea. Extreme events were identified by the peaks over threshold (POT) method which enhanced the accuracy of estimation. The nonstationary model, incorporating a nonlinear regression using time as a covariate

River wood (RW) plays a key role in shaping aquatic and riparian habitats while influencing sediment and water dynamics. This study presents the first automated RW detection model using Random Forest classification and near-infrared aerial imagery on the Meurthe River. By progressively incorporating exogenous indices, the model achieved recall, precision, and F1-scores between 12%–39%, 90%–94%, and

Northeastern temperate forests provide essential ecosystem services that are increasingly threatened by climate change and air pollution. To evaluate integrated ecosystem responses to these changes, we applied the PnET-CN-daily model to project carbon, nitrogen, and water cycling dynamics at Harvard Forest (Petersham, MA, USA) throughout the 21st century. The projections were based on future climate

Wastewater treatment plants in combined sewer systems are often required to accommodate the widely fluctuating flow due to the dynamic interactions between multiple water flow sources. A major challenge in wastewater management, and particularly in combined sewer overflow (CSO) mitigation, is decoupling the total sewer flow into its components: dry-weather flow (DWF) and rain-derived inflow and infiltration

We conducted a Bayesian calibration of the IPCC tier-2 Steady-State (IPCCT2) model using long-term experimental (LTE) data from Canadian croplands. A global sensitivity analysis identified key parameters influencing the prediction of soil organic carbon (SOC) stocks, including those governing the temperature response curve, optimal decay rate in the passive pool, and stabilization efficiencies for

River ice plays a critical role in controlling streamflow in cold regions. The U.S. Geological Survey (USGS) qualifies affected water-level measurements and inferred streamflow by ice conditions at a date later than the day of the actual measurements. This study introduces a novel computer vision-based framework, River Ice-Network (RIce-Net), that uses the USGS nationwide network of ground-based cameras

Good practices in model software development are essential for boosting uptake. While user-centric approaches are much advocated, challenges remain in including users in development due to diverse definitions for ‘users’, their perceived credibility as an information source, and the influence of market-based innovation choices and the anticipation of (future) demands. While enhancing user feedback

Ozone (O3), a short-lived climate pollutant, continues to increase despite policies aimed at suppressing its precursors in South Korea. The government operates approximately 500 observatories to monitor O3 and trace gases. Researchers use these data to address the ongoing issue of increasing O3 levels. However, challenges in data retrieval from observatories may introduce biases in O3 studies. In this

Accurate streamflow modeling is crucial for water resource management in dry and semi-arid regions. This study proposes a novel approach combining machine learning (ML) with conceptual and physically-based models to address of traditional model limitations in Iran's semi-arid Jazmourian River Basin. The HBV and SWAT hydrological models are used for conceptual and physically-based simulations, respectively

While the National Water Model (NWM) provides high-resolution, large-scale streamflow data across the United States, its effectiveness as a key water resources management tool in the drought-prone Western US needs further investigation. Previous studies revealed that the NWM has limitations in controlled basins, impacted by reservoir operations and diversions not explicitly included within the model

Reliable anomaly detection is crucial for water resources management, but the complexity of environmental sensor data presents challenges, especially with limited labeled data in water quality analysis. Functional data has experienced significant growth in anomaly detection, but most applications focus on unlabeled datasets. This study assesses the performance of multivariate functional data analysis

The reduction of the Arctic sea ice opens new shipping routes, necessitating advanced planning for safe navigation due to unpredictable ice conditions and severe weather. However, current Arctic route planning systems lack real-time adjustments and comprehensive consideration of the dynamic navigation environment. To address these limitations, we developed RouteView 2.0, an improved intelligent system

To address the limitations of traditional ozone (O3) forecasting models, this study established a novel Transformer-based model integrating attention scores and the peak perception. Attention scores dynamically quantify nonlinear relationships between O3 and influencing factors, while peak perception method penalizes peak O3 errors, ensuring accurate predictions during O3 exceedance events. Our results

The paper presents the open-source software climdex-kit which includes modules to compute, analyze and visualize climate indices based on the input data, target domain and temporal extent defined by the user. It is intended to ease the retrieval and interpretation of meaningful information for climate change studies and support the development of climate services for sectoral applications with flexible

This study introduces the SWAGU model, which overcomes limitations of existing approaches by combining SWMM's robust pipe network modeling capabilities with ANUGA's advanced unstructured mesh-based surface flow simulation, enabling more accurate prediction of flood dynamics in complex urban environments. The model's outputs are integrated into an enhanced cloud model framework. This framework improves

Hydrologic modelling plays a vital role in water resources management but often falls short of achieving the positive change modelers envision. In this position paper we argue that a key contributing factor is the lack of trust and shared understanding among modelers, decision-makers, and the public. Models need to be trusted first—a social challenge as well as a technical one. Through three case studies—involving

Timely precipitation information is essential for water resources management and hazard monitoring. In regions with limited ground-based measurements, satellite precipitation products (SPPs) provide a valuable alternative, though data latency often creates an information gap for real-time applications. This study addresses the latency gap of IMERG-ER using a U-Net-based Convolutional Neural Network

SENTINEL (SEnsor NeTwork INtelligent Emissions Locator) is an application developed in R Shiny to support emerging user groups of lower cost fenceline sensors, such as those monitoring volatile organic compound or methane concentrations inside and near industrial facilities or for emergency response applications. During deployment, sensors collect a large quantity of high-frequency pollutant concentration

Predicting landslides across large regions using physically-based models requires balancing computational accuracy and efficiency. Current methods often use limited resolutions, underutilizing available data. We present a prototype mesh generator that manages multiple resolutions in grid-based modeling frameworks, focusing on identifying likely landslide initiation points and conditionally stable pixels