Search Results (2,671)

Accurate prediction, forecasting and interpretability of air pollutant concentrations are important for sustainable environmental management and protecting public health. An integrated artificial intelligence (AI) framework is proposed to predict, forecast and analyse six major air pollutants, such as particulate matter concentrations (PM_2.5 and PM₁₀), ground-level ozone (O₃), carbon monoxide (CO), nitrogen dioxide (NO₂), and sulphur dioxide (SO₂), using a combination of ensemble and deep learning models. Five years of hourly air quality and meteorological data are analysed through correlation and Granger causality tests to uncover pollutant interdependencies and driving factors. The results of the Pearson correlation analysis reveal strong positive associations among primary pollutants (PM_2.5–PM₁₀, CO–nitrogen oxides NO_x and VOCs) and inverse correlations between O₃ and NO_x (NO and NO₂), confirming typical photochemical behaviour. Granger causality analysis further identified NO₂ and NO as key causal drivers influencing other pollutants, particularly O₃ formation. Among the 23 tested AI models for prediction, XGBoost, Random Forest, and Convolutional Neural Networks (CNNs) achieve the best performance for different pollutants. NO₂ prediction using CNNs displays the highest accuracy in testing (R² = 0.999, RMSE = 0.66 µg/m³), followed by PM_2.5 and PM₁₀ with XGBoost (R² = 0.90 and 0.79 during testing, respectively). The Air Quality Index (AQI) analysis shows that SO₂ and PM₁₀ are the dominant contributors to poor air quality episodes, while ozone peaks occur during warm, high-radiation periods. The interpretability analysis based on Shapley Additive exPlanations (SHAP) highlights the key influence of relative humidity, temperature, solar brightness, and NO_x species on pollutant concentrations, confirming their meteorological and chemical relevance. Finally, a deep-NARMAX model was applied to forecast the next horizons for the six air pollutants studied. Six formulas were elaborated using input data at times (t, t − 1, t − 2, …, t − n) to forecast a horizon of (t + 1) hours for single-step forecasting. For multi-step forecasting, the forecast is extended iteratively to (t + 2) hours and beyond. A recursive strategy is adopted for this purpose, whereby the forecast at (t + 1) is fed back as an input to generate the forecasts at (t + 2), and so forth. Overall, this integrated framework combines predictive accuracy with physical interpretability, offering a powerful data-driven tool for air quality assessment and policy support. This approach can be extended to real-time applications for sustainable environmental monitoring and decision-making systems. Full article

Accurate dew intensity prediction is vital in multiple fields, such as agriculture, meteorology, industry, and transportation. This study addresses the cross-disciplinary demands for dew intensity prediction by proposing a hybrid deep learning model based on the improved hippopotamus optimization (IHO). Key influencing factors were selected through multidimensional meteorological data correlation analysis, and a fusion architecture of a Bidirectional Temporal Convolutional Network (BiTCN) and a Support Vector Machine (SVM) was constructed. The IHO algorithm is adopted to optimize model parameters and enhance prediction accuracy adaptively. Experiments were conducted using ten years of meteorological data to verify the prediction of twelve-hour dew intensity in three typical ecosystems in Northeast China: farmland, marsh wetland, and urban areas. The results show that the optimized IHO-BiTCN-SVM model achieved significant improvements in key indicators, including MAE, MAPE, MSE, RMSE, and R². For the farmland ecosystem, MAE was reduced by 72.2% (0.0016572 vs. 0.0059659), MSE decreased from 6.8552 × 10⁻⁵ to 6.7874 × 10⁻⁶, and R² increased by 12.5% (0.98791 vs. 0.87793). The IHO algorithm reduced the MAE of the farmland system by 39.6%, the MAPE by 41.6%, and the MSE by 60.2%, yet the R² increased by 1.8% compared with the benchmark model. This model effectively overcomes the subjectivity of traditional methods through an intelligent parameter optimization mechanism, providing reliable technical support for precise agricultural irrigation decisions, urban dew formation warnings, and wetland ecological protection. Full article

To address the challenge of predicting runoff processes in the middle reaches of the Yangtze River under the influence of complex river–lake relationships and human disturbances, this paper proposes a coupled model based on the Sparrow Search Algorithm-optimized Long Short-Term Memory neural network (SSA-LSTM) for daily runoff forecasting at the Jiujiang Hydrological Station. The input data were preprocessed through feature selection and sequence decomposition. Subsequently, the Sparrow Search Algorithm (SSA) was utilized to perform automated of key hyperparameters of the Long Short-Term Memory (LSTM) model, thereby enhancing the model’s adaptability under complex hydrological conditions. Experimental results based on multi-station hydrological and meteorological data of the middle reaches of the Yangtze River from 2009 to 2016 show that the SSA-LSTM achieves a Nash–Sutcliffe Efficiency (NSE) of 0.98 during the testing period (2016). The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) are reduced by 49.3% and 51.3%, respectively, compared to the standard LSTM. A comprehensive evaluation across different flow levels, utilizing Taylor diagrams and error distribution analysis, further confirms the model’s robustness. The model demonstrates robust performance across different flow regimes: compared to the standard LSTM model, SSA-LSTM improves the NSE from 0.45 to 0.88 in high-flow scenarios, exhibiting excellent capabilities in peak flow prediction and flood process characterization. In low-flow scenarios, the NSE is improved from −0.77 to 0.72, indicating more reliable prediction of baseflow mechanisms. The study demonstrates that SSA-LSTM can effectively capture hydrological nonlinear characteristics under strong river–lake backwater and human disturbances, providing a high-precision and high-efficiency data-driven method for runoff prediction in complex basins. Full article

Large-eddy simulation (LES) within a weather research and forecasting (WRF) model coupled with an active scalar transport equation was used to simulate Atmospheric Boundary Layer conditions during the Mosquito fire, the largest wildland fire in California during September 2022. The simulations were conducted with realistic boundary conditions derived from the National Oceanic and Atmospheric Administration (NOAA) High Resolution Rapid Refresh (HRRR) model, with the aim of better understanding the two-way coupling between the ABL and plume dynamics. The terrain was extremely inhomogeneous, and the topography varied significantly within the numerical domain. Initially, LES of the smoke-free ABL was conducted on nested domains, and detailed ABL data were gathered from 8 to 9 September 2022. LES simulations were validated using four Automated Surface Observing System (ASOS) stations and NOAA meteorological (MET) observations, as well as NOAA met Twin Otter measurements, and the desired accuracy was established. The smoke plume was then released into the ABL at noon on 9 September 2022, and the plume simulations were conducted for a period of one hour following the release. During this period, the ABL transitioned from convective to buoyancy-shear-driven regimes. Late-night and early-morning conditions are influenced by the complex topography and low-level jet, whereas buoyancy and shear control the ABL dynamics during the morning and afternoon hours. The plume vertical transport is influenced by the ABL depth and the size of the vertical turbulence structures during that time, whereas the wind conditions and turbulent kinetic energy within the ABL dictate the horizontal transport scales of the plume. In addition, the results demonstrate that the plume modifies the microclimate along its path. Full article

Since land use considerably affects the spatial variation of PM_2.5 levels, it is crucial to predict PM_2.5 concentrations under future land use changes. However, prior research has primarily concentrated on meteorological factors influencing PM_2.5 predictions, while neglecting the effect of land use configurations. Consequently, in our study, a novel Patch-generating Land Use Simulation–Land Use Regression (PLUS-LUR) method was developed by integrating the PLUS model’s dynamic prediction capability with the LUR model’s spatial interpretation strength. The incorporation of landscape indices as key variables was essential for predicting PM_2.5 concentrations. First, the random forest-optimized LUR method was trained with PM_2.5 datasets from the Pearl River Delta (PRD) monitoring stations and multi-source spatial datasets. We assessed the modeling accuracy with and without considering landscape indices using the test dataset. Subsequently, the PLUS approach was applied to forecast land use as well as associated landscape indices in 2028. Based on these projections, grid-scale influencing factors were input into the previously constructed LUR model to forecast future PM_2.5 distributions at a grid scale. The results reveal a spatial pattern with higher PM_2.5 levels in central areas and lower levels in peripheral regions. Furthermore, the PM_2.5 concentrations in the PRD are all below the Grade II threshold of the China Ambient Air Quality Benchmark in 2028. Notably, the predictions incorporating landscape indices demonstrate higher accuracy and reliability compared to those excluding them. These results provide methodological support for future PM_2.5 assessment and land use management. Full article

Mosquitoes (Diptera: Culicidae) exhibit vertical stratification patterns in forest environments, a fundamental ecological aspect for understanding niche occupation patterns, host-seeking behavior, and consequently arbovirus transmission mechanisms. Despite the relevance of this topic, available studies mostly focus on genera such as Aedes, Haemagogus, and Sabethes which are traditionally associated with arbovirus transmission. There are still important gaps regarding stratification and seasonality in the Mansoniini tribe, whose biology and epidemiological role remain underexplored, especially in highly biodiverse ecosystems such as the Atlantic Forest. This study evaluated the influence of seasonality and vertical stratification on the mosquito community, with a detailed focus on the Mansoniini tribe, in an Atlantic Forest fragment in Brazil, between May 2023 and December 2024. Captures were performed monthly using CDC light traps positioned at 1.5 m and 10 m heights, and specimens were morphologically identified. A total of 880 mosquitoes from nine genera and 24 species were captured, of which 91 (10.3%) belonged to the Mansoniini tribe. The most abundant species were Coquillettidia fasciolata and Mansonia titillans, recorded in both strata. Our results indicate no marked vertical segregation for the studied mosquito community in this specific location, but a strong influence of seasonality, particularly for the Mansoniini tribe, reinforcing the role of meteorological data on the population structure of these species. These site-specific findings offer a foundational ecological portrait and a robust methodological template for a neglected taxon. They generate critical, testable hypotheses about niche partitioning in fragmented forests and underscore the necessity for broader spatial replication to disentangle the relative influence of seasonal versus vertical drivers in similar ecosystems. Full article

Terminal lakes in arid regions are highly vulnerable to climate variability and human water management, yet their long-term hydrological responses under multi-river regulation remain insufficiently quantified. Using Taitema Lake at the terminus of the Tarim Basin as a case study, this research integrates Landsat and Sentinel observations (2005–2025) with meteorological and river-inflow records to examine lake area dynamics and to identify river-specific hydrological controls. The results show pronounced intra- and interannual variability, with the lake expanding to a maximum of 461.52 km² in October 2017 and shrinking to 0.35 km² in October 2008. High-frequency permanent water (~43 km²) is concentrated in the deep central basin and largely influenced by the Qarqan River, whereas seasonal water (~300 km²) is broadly distributed and strongly affected by ecological releases from the Tarim River. Quantified inflow–area relationships indicate that the lake expands by 7–14 km² for each 0.1 × 10⁸ m³ of inflow. Based on frequency-based hydrological analysis, this study develops joint inflow strategies for wet, normal, and dry years, offering a practical hydrological basis for more precise and adaptive water allocation schemes in arid terminal lakes. Full article

The frequency and intensity of tropical cyclones (TCs) in the Arabian Sea have increased in recent decades, heightening concerns regarding regional vulnerability and forecasting difficulties. This study examines the impact of the Madden–Julian Oscillation (MJO) on TCs activity—formation, frequency, and severity—over the Arabian Sea from 1982 to 2021. This study analyzes variations in convection, vertical wind shear (VWS), sea level pressure (SLP), and relative humidity (RH) across different MJO phases utilizing the best-track data from the India Meteorological Department (IMD), the Real-Time Multivariate MJO (RMM) index, and reanalysis datasets from the National Oceanic and Atmospheric Administration (NOAA) and the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR). Results show that more than 80% of TCs form during the convectively active phases of the MJO (P1–P4). These phases have the most noticeable negative outgoing longwave radiation (OLR) anomalies, as well as higher mid-level moisture and low-pressure anomalies, which are good for cyclogenesis. On the other hand, suppressed phases (P6–P8) have positive outgoing longwave radiation, dry air in the middle troposphere, and high-pressure anomalies, which make it harder for TCs to form. While VWS is predominantly favorable during both active and inactive phases, thermodynamic and convective factors principally regulate the modulation of TC activity. The simultaneous presence of active MJO phases with positive Indian Ocean Dipole (pIOD) and neutral or El Niño conditions markedly increases TC frequency, highlighting a combined influence link between interannual–El Niño–Southern Oscillation (ENSO) and IOD– and intraseasonal (MJO) variability. Additionally, the association between MJO and the Indo-Pacific Warm Pool (IPWP) reveals that TC activity peaks during convectively active MJO phases under the second twenty years of this study, emphasizing the influence of large-scale oceanic warming on TC variability. These findings underscore the critical function of the MJO in regulating TC activity variability in the Arabian Sea and stress its significance for enhancing intraseasonal forecasting and disaster preparedness in the area. Full article

Farmland N₂O emissions exhibit significant fluctuations in subtropical regions due to notable seasonal rainfall and temperature variations. The dominant factors influencing N₂O emissions in red-soil sloping farmland, which is widely distributed and actively cultivated in the region, remain uncertain. To investigate N₂O emission characteristics of red-soil sloping farmland and responses to meteorological and soil environmental variables and tillage practices, a typical planting system (summer peanut-winter rapeseed rotation system) in southern China was selected. Two common soil micro-environments (conventional tillage, CT, n = 6; and conventional tillage with straw mulching, MT, n = 4) were established within this system, and in situ N₂O emissions were monitored over two consecutive years using the static chamber–gas chromatography method. The N₂O emission peaks across various growing seasons occurred primarily within 1 to 16 days after fertilization. The N₂O emission hotspot periods were observed during the first month following fertilization, accounting for 74.13–91.01% of the total emissions during each growing season. Significant interannual variations in seasonal N₂O cumulative emissions were observed, whereas no significant difference in cumulative N₂O emissions was observed between MT and CT. Changes in weather and soil environment jointly drive the dynamics of N₂O emissions from red soil sloping farmland. Rapeseed-season N₂O emissions were driven mainly by rainfall and air temperature, whereas peanut-season N₂O emissions were also influenced by soil temperature and NO₃⁻-N content at 0–10 cm depths. These findings provide a sound basis for developing eco-agricultural mitigation pathways in subtropical red-soil hilly regions. Full article

Driven by factors such as meteorology, topography, and industrial structure, the concentrations of volatile organic compounds (VOCs) exhibit significant spatial heterogeneity. Investigating the characteristics and sources of VOCs in different regions is therefore crucial for formulating targeted strategies to mitigate their contributions to fine particulate matter (PM_2.5) and ozone (O₃) pollution. This study comprehensively investigated—for the first time—the concentration characteristics, sources, and contributions to secondary organic aerosol (SOA) and O₃ formation of VOCs at an urban background site in Ganzhou, a southern Chinese city, based on hourly observations of VOCs during 2023. Analyses included ozone formation potential (OFP), secondary organic aerosol formation potential (SOAFP), and positive matrix factorization (PMF) source apportionment. The influence of photochemical loss was assessed using a photochemical age parameterization method. The results showed an annual average total VOC concentration of 22.6 ± 13.17 ppbv, with higher levels in winter and lower in summer. Alkanes were the dominant species (45.76%). After correcting for photochemical loss, the initial concentration of VOCs (IC-VOCs) was approximately 60% higher than the observed concentration of VOCs (OC-VOCs), with alkenes becoming the dominant group in IC-VOCs (≈72%). OFP analysis indicated that the OFP calculated using initial VOC concentrations (IC-OFP) was substantially higher (by 320 μg/m³) than the values calculated using observed VOC concentrations (OC-OFP), primarily due to the increased contribution of alkenes. SOAFP was higher in spring and winter, and lower in summer and autumn, with aromatic hydrocarbons being the dominant contributors (>85%). PMF results based on month-case studies identified combustion and industrial process sources as the major contributors (>20%) in August, while combustion and vehicle exhaust dominated in January. Photochemical loss significantly influenced source apportionment, particularly leading to an underestimation of biogenic emissions during a warm month (August). These findings underscore the necessity of accounting for photochemical aging and offer a scientific basis for refining targeted VOC control measures in Ganzhou and similar regions. Full article

Progressive climate change and building morphology influence the specific microclimate of built-up areas. This has a fundamental role in research on energy use and thermal comfort inside buildings. Most studies using data for dynamic energy simulation are based on information collected at meteorological stations in rural areas. This can lead to erroneous predictions. The main goal of the study was to combine two simulation tools—ENVI-met for microclimate predictions around historical building layouts, and DesignBuilder for assessing indoor comfort. Illustrating the impact of input data on simulation results was conducted using three types of weather data: (1) from a field campaign, (2) from a suburban station, and (3) from the typical meteorological year. The obtained results confirm that the highest precision was achieved in analyses where information obtained at a real scale in the city centre was used as boundary conditions (field measurements: MAPE = 0.6 °C, RMSE = 0.7 °C). The next step was to estimate the thermal sensations inside the living room of the existing residential building. Thermal comfort was determined using the operative temperature as an indicator. Incorporating realistic urban weather inputs enhanced the reliability of indoor comfort modelling and provided a more accurate basis for planning thermal resilience in historic residential buildings. Full article

Validation of ICON model configurations optimized over a limited domain is essential before accepting new semi-empirical parameters that influence the behavior of subgrid-scale schemes. Because such parameters can modify the dynamics of a numerical weather prediction (NWP) model in highly nonlinear ways, we analyze one season of forecasts (December 2022, January and February 2023) generated with the NWP ICON-LAM through the lens of machine learning–based diagnostics as a complement to traditional evaluation metrics. The goal is to extract physically interpretable information on the model behavior induced by the optimized parameters. This work represents the first part of a wider study exploring machine learning tools for model validation, focusing on two specific approaches: Empirical Orthogonal Functions (EOFs), which are widely used in meteorology and climate science, and autoencoders, which are increasingly adopted for their nonlinear feature extraction capability. In this first part, EOF analysis is used as the primary tool to decompose weather fields from observed reanalysis and forecast datasets. Hourly 2-m temperature forecasts for winter 2022–2023 from multiple regional ICON configurations are compared against downscaled ERA5 data and in situ observations from ground station. EOF analyses revealed that the optimized configurations demonstrate a high skill in predicting surface temperature. From the signal error decomposition, the fourth EOF mode is effective particularly during night-time hours, and contributes to enhancing the performance of ICON. Analyses based on autoencoders will be presented in a companion paper (Part II). Full article

Climate change and altered hydrological regimes are restructuring wetland habitats globally, triggering cascading effects on colonial waterbirds. This study investigates how environmental drivers, including thermal anomalies, water-level fluctuations, and aqueous surface extent, influence the distribution and size of waterbird colonies (Ardeidae, Threskiornithidae, and Phalacrocoracidae) in the Danube Delta Biosphere Reserve. We integrated colony census data (2016–2023) with remote-sensing-derived habitat metrics, in situ meteorological and hydrological measurements to model colony abundance dynamics. Our results indicate that elevated early spring temperatures and water level variability are the primary determinants of numerical population dynamics. Spatial analysis revealed a heterogeneous response to hydrological stress: while the westernmost colony exhibited high site fidelity due to its proximity to persistent aquatic surfaces, the central colonies suffered severe declines or local extirpation during extreme drought periods (2020–2022). A discernible eastward shift in bird assemblages was observed toward zones with superior hydrological connectivity and proximity to anthropogenic hubs, suggesting an adaptive spatial response that was consistent with behavioral flexibility. We propose an adaptive management framework prioritizing sustainable solutions for maintaining minimum lacustrine water levels to preserve critical foraging zones. This integrative framework highlights the pivotal role of remote sensing in transitioning from reactive monitoring to predictive conservation of deltaic ecosystems. Full article

This study investigates the vegetation dynamics and phenotypic differentiation of five triticale (×Triticosecale Wittm.) varieties under the region-specific agroecological conditions of Southern Dobruja, Bulgaria, across two growing seasons (2024–2025), with the aim of evaluating how local climatic variability shapes vegetation index patterns. UAV-based multispectral imaging was employed throughout key phenological stages to obtain reflectance indices, including NDVI, SAVI, EVI2, and NIRI, which served as indicators of canopy development and physiological status. NDVI was used as the primary reference index, and a baseline value (NDVIbase), defined as the mean NDVI across all varieties on a given date, was applied to evaluate relative varietal deviations over time. Multiple linear regression analyses were performed to assess the relationship between NDVI and baseline biometric parameters for each variety, revealing that varieties 22/78 and 20/52 exhibited reflectance dynamics most closely aligned with expected developmental trends in 2025. In addition, the relationship between NDVI and meteorological variables was examined for the variety Kolorit, demonstrating that relative humidity exerted a pronounced influence on index variability. The findings highlight the sensitivity of triticale vegetation indices to both varietal characteristics and short-term climatic fluctuations. Overall, the study provides a methodological framework for integrating UAV-based multispectral data with meteorological information, emphasizing the importance of region-specific, time-resolved monitoring for improving precision agriculture practices, optimizing crop management, and supporting informed variety selection. Full article

Aerosol microphysical and optical properties play a crucial role in cloud microphysics, precipitation physics, and flood formation over areas characterized by complex monsoon regimes. This research presents a multi-source data integration approach to analyzing the spatio-temporal interaction between precipitation, aerosols, and flooding in the state of Kerala, incorporating an air mass trajectory analysis to examine its potential contribution to flooding. The results show that the Aerosol Optical Depth (AOD) values were high in the coastal districts (>0.8) in the La Niña year (2021) but low in the El Niño year (2015). On the precipitation side, 2018 and 2021 were both years with a high degree of anomalies, resulting in heavy rainfall that led to widespread flooding in the Thrissur district, among others. The trajectory analysis revealed that the Indian Ocean controls the precipitation during the southwest monsoon and the pre-monsoon. The post-monsoon precipitation is mainly sourced from the Arabian Peninsula and Arabian Sea, transferring marine aerosols along with desert aerosols. The overall study shows that the variability in aerosols and precipitation is more subject to change by the meteorological dynamics, as well as influenced by the regional changes in land use and land cover, causing fluxes in the land–atmosphere interactions. In conclusion, the present study highlights the possible interactive functions of atmospheric dynamics and anthropogenic land use modifications in generating a flood hazard. It provides essential information for land management policies and disaster risk reduction. Full article