Environmental Prediction Research Articles

Large-Scale Surface Air Temperature Bias in Summer over the CONUS and Its Relationship to Tropical Central Pacific Convection in the UFS Prototype 8

Abstract This study aims to understand the source of surface air temperature bias over the contiguous United States (CONUS) during boreal summer (June–September) in the Unified Forecast System (UFS) coupled model prototype 8 (P8), developed by the National Centers for Environmental Prediction (NCEP) and the National Oceanic and Atmospheric Administration (NOAA). The focus is on the subseasonal variability defined as a weekly average in weeks 2–5 of forecast leads (total 224 cases; 4 weeks × 2 initialization dates × 4 months × 7 years). The large-scale surface air temperature bias pattern is extracted using the empirical orthogonal function (EOF) analysis. The associated principal component describes the variability of bias for each reforecasting week throughout the 2011–17 reforecasting period. The leading EOF of surface air temperature bias exhibits an east–west dipole pattern over the CONUS, explaining 31.6% of the total variability of weekly temperature bias. This bias pattern is strongly related to the upper-level Rossby wave induced by a bias in convection over the central tropical Pacific. Furthermore, the mean bias of background flow in the extratropics degrades the representation of teleconnections from the tropics to the midlatitudes. UFS P8 has weaker zonal wind over the North Pacific with stronger vertical wind shear than the ERA5 reanalysis. The weak zonal wind hampers the Rossby wave’s propagation, while strong vertical shear reduces its amplitude.

Incorporating Spatial Information for Regionalization of Environmental Parameters in Machine Learning Models

AbstractMachine learning models have gained popularity for environmental variable predictions due to their capacity to capture complex relationships and automate learning. However, incorporating spatial information as covariates into these models remains a challenge, as they may struggle to recognize spatial structures or autocorrelation without explicit training. In this study, we address this challenge by integrating spatial information into a random forest model, enhancing nitrate concentration predictions in groundwater. Using a dataset from 1,550 well locations in Baden-Wuerttemberg, Germany, spanning 2016 through 2019, we consider various environmental covariates including climate data, topography, land cover, soil properties, and hydrology. To incorporate spatial information, we employ eight techniques leveraging spatial coordinates (geographic coordinates, polynomial geographic coordinates, oblique geographic coordinates) or distances (Wendland transformed coordinates, Euclidean distance fields, Euclidean distance matrix, principal component analysis, eigenvector spatial filtering). Results are compared with a baseline model and a univariate ordinary kriging benchmark, evaluated through leave-one-out cross validation, various error metrics, and Moran’s I of residuals. Our findings highlight that integrating spatial information significantly enhances random forest model accuracy in predicting groundwater nitrate concentrations. Distance-based methods, like the Euclidean distance matrix, outperform coordinate-based approaches, albeit with higher computational requirements. Employing a dimension-reduced matrix strikes a balance between performance and accuracy. This study advances groundwater management and demonstrates the effectiveness of machine learning models in environmental studies.

The role of the Indian Ocean Dipole in modulating the austral spring ENSO teleconnection to the Southern Hemisphere

Abstract. The combined influence of the El Niño–Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD) on the extratropical circulation in the Southern Hemisphere (SH) during austral spring is examined. Reanalyses and the large ensemble of National Centers for Environmental Prediction (NCEP) Climate Forecast System version 2 (CFSv2) model outputs were used to compute composites and linear regressions for relevant variables. The results show that a positive IOD can reinforce the El Niño-induced circulation by merging the Indian Ocean wave train with the Pacific South American (PSA) pattern over the Pacific Ocean. In addition, the results obtained with the CFSv2 model output shows that strong positive IODs can contribute to enhancing the circulation signal of the El Niño anomalies and the Indian Ocean wave train. On the other hand, negative IODs in combination with La Niña do not have that combined circulation response. While there is a moderate intensification of the circulation anomalies associated with La Niña, accompanied by some changes in the location of their main action centers, results vary considerably between linear regression, the observed composites, and model composites. Regarding the influence of the IOD activity (independent of ENSO), reanalysis-based results show that the IOD positive phase has a significant impact over the entire SH, while the negative phase is associated with weaker anomalies and less consistent atmospheric response.

Environmental Suitability Predictions for the Distribution and Potential Cultivation of Artemisia afra in South Africa

Cultivation is advocated as a solution for the sustainable exploitation of medicinal plants. Understanding environmental factors influencing plant species distribution will eliminate the indiscriminate introduction of medicinal plants to inappropriate cultivation regions. This study investigated environmental conditions for the distribution of Artemisia afra and mapped out potential areas for its cultivation in South Africa. Soil samples were collected for analysis in the Free State Province in South Africa. To identify suitable environmental conditions for the natural distribution of A. afra, the South African National Botanical Institute database and physically collected Global Positioning System points were used in a maximum entropy model. Monthly long-term average interpolated weather surfaces were used to estimate the effect of climate change on future climate suitability for A. afra distribution. Sixty-one percent of soil samples from different A. afra populations were clay loam soils with a slightly acidic to neutral pH. The carbon source utilization, Shannon Weaver Index, and species richness were positively correlated with one group of fourteen soil samples, and species evenness was positively correlated with the second group, consisting of four samples. Climate change will only affect the distribution of A. afra in the very long term. The current study provides critical information for identifying suitable cultivation areas for A. afra while supporting conservation efforts from an ecological point of view.

Multi-layer perceptron artificial neural network for environmental risks prediction of SW-induced pollution in Dar es Salaam, Tanzania

<p>Many metropolitan areas face significant environmental challenges posed by improper disposal and management of solid waste. As a result, environmental risks have emerged as a pressing concern, prompting dedicated research efforts. This study on environmental risk prediction of Dar es Salaam SW coincides with a mounting governmental effort over rising pollution levels from inadequate SW management. Using the multi-layer perceptron artificial neural network (MLP-ANN) model, it effectively examines the prevailing conditions and forecasts waste generation rates (WGRs) and environmental risk index (ERI) associated with SW pollution. As confirmed with 94.5% prediction accuracy and 86.5% success rate of the MLP-ANN model, WGRs in Dar es Salaam have doubled in less than two decades. Besides, over 40% of the overall generated SW is left unattended. Consequently, the ERI exhibits a consistent upward trajectory throughout the assessment period, with intermittent fluctuations between Level II and III but a persistent overall increase. Projections indicate an escalation of ERI to Level IV by 2025/26 and to a critical threshold (Level V) by 2038. The key indices such as pressure, state, and impact are anticipated to reach critical thresholds ahead of the comprehensive ERI. This underscores the imperative for timely interventions and the urgency of addressing SW management issues to curb the escalating environmental risks in Dar es Salaam and other metropolises with similar challenges.<strong></strong></p>

Contribution of radio occultation data to atmospheric river landfall forecasts in the eastern North Pacific

AbstractThis study assesses the impact of Global Navigation Satellite System radio occultation (RO) data assimilation (DA) on low‐predictability atmospheric river (AR) landfall forecasts. The period of study is October–March 2020–2022, focusing on AR events over the eastern North Pacific. Eighteen AR landfall events with large forecast errors in the Global Forecast System of the National Centers for Environmental Prediction are selected. Two numerical experiments are conducted for each AR event. One assimilates RO data using a non‐local excess phase operator to improve its initial condition (labeled as EPH forecast), while the other does not (reference run; labeled as REF forecast). Relative to REF, the root‐mean‐squared errors of integrated water vapor transport (IVT), moisture, and wind are reduced by EPH in the main AR activity areas throughout the whole forecast period, except for zonal winds. EPH significantly reduces the underestimation of strong IVT occurrence rate by 2.6%–4.8% and the underestimation of IVT intensity by 15.7–35.0 kg·m−1·s−1, compared to REF. EPH also substantially improves the precipitable water distribution and AR‐related precipitation over the ocean. Despite these basin‐wide improvements, AR coverage, location, and precipitation over land are not significantly affected by RO DA. Specifically, the main contribution of RO DA to the root‐mean‐squared error reduction of AR forecast is the wind improvement prior to AR landfall and the mid‐level moisture improvement after AR landfall. In summary, this study highlights the potential positive contributions of RO DA to AR forecasts, and explores the mechanisms and spatial and temporal characteristics of the improvements.

Climate and Socio-Sexual Environment Predict Interpopulation Variation in Chemical Signaling Glands in a Widespread Lizard.

Many animal species show considerable intraspecific phenotypic variation. For species with broad distributions, this variation may result from heterogeneity in the strength and agents of selection across environments and could contribute to reproductive isolation among populations. Here, we examined interpopulation variation in a morphological trait related to chemical communication, femoral pore number (FP), using 3437 individuals from 55 Pyrenean populations of the common wall lizard (Podarcis muralis). Specifically, we tested the relative roles of genetic relatedness and gene flow, and adaptation to local conditions in generating this variation, with particular interest in the influence of climate and the socio-sexual environment (i.e., the intensity of sexual selection, estimated using sexual size dimorphism [SSD] and adult sex ratio as proxy measures). We found significant interpopulation variation and sexual dimorphism in FP, as well as high genomic differentiation among populations driven by both geographic and environmental distances. Specifically, FP differences across populations were best predicted by a combination of positive allometry and the local intensity of sexual selection, as determined by SSD, or local climatic conditions. Higher FP in more male-competitive environments, or with higher temperature and vegetation complexity, is consistent with adaptation to maintaining signaling efficacy of territorial scent marks. These results suggest that adaptation to local conditions contributes to interpopulation divergence in FP and thus environmental changes can potentially impact the fine-tuning of chemical communication mediating social and sexual behavior.

Air Temperature Variations Analysis of the Hualian M6.9 Earthquake

This study examines the three-dimensional layered air temperature variations associated with the Hualian M6.9 earthquake, which occurred on 18 September 2022, using data from the National Center for Environmental Prediction (NCEP) and extracting background information for air temperature through tidal forces. Changes in air temperature stratification revealed that near the epicenter, temperature anomalies began on 12 September and peaked on 13 September, with pronounced increases in magnitude and area. These anomalies followed a seismic thermal anomaly pattern, i.e., one with greater amplitude and a wider range near the land surface, decreasing with altitude until they dissipated, while the other exhibited high atmospheric temperature anomalies, i.e., the upper atmospheric warming exceeded that near the land surface, likely due to meteorological factors. Stable weather conditions and a low geomagnetic Kp index and Dst index during the research period supported the reliability of these findings. The deformation field recorded by InSAR reflects crustal deformation directly caused by fault slip and stress accumulation; moreover, the area with significantly higher air temperature coincides with the areas with the most concentrated deformation.

The state of Earth Observing system today: updates compiled from recent EUMETSAT Meteorological Satellite Conferences

Abstract This work presents highlights from the 2021 and 2022 EUMETSAT conferences, drawn from their presentations, posters and panel discussions and placed into the wider context of the global Earth Observing (EO) system. These highlights collectively reveal much about the current state of knowledge about the EO system, its potential future evolutions and how that system may be used to produce useful products and services. European, American and Asian space agencies presented their visions for the next generation of operational satellite programmes and demonstrated how these will continue to improve environmental forecasting and monitoring products. User communities presented updates in the use of satellite data, including climate records, novel precipitation retrievals, drought monitoring, weather forecasting and retrievals of a broadening range of trace gases. On the technology side, discussion on the impact of Artificial Intelligence (AI) for Earth observation was a major theme, particularly for weather forecasting, data assimilation or other environmental predictions. Cloud computing was another topic due to its potential to streamline the workflow of EO scientists, enhancing collaborations and unlocking access to previously-unavailable data or computing resources. Finally, discussion on the miniaturization of observational instruments was another major theme of both conferences, highlighting both the possibility of novel or enhanced observations and the emerging economic case for commercial entities to operate fleets of meteorological satellites.

Multi-model fusion method for predicting CO2 concentration in greenhouse tomatoes

With the rapid development of greenhouse agriculture, accurate prediction of environmental parameters such as temperature, humidity, and carbon dioxide concentration is crucial for optimal crop growth. Traditional forecasting models struggle with the nonlinear and complex nature of greenhouse data, leading to challenges in model robustness. This study addresses these issues by proposing a multi-model fusion strategy for predicting CO2 concentration in greenhouse tomatoes. The proposed method integrates wavelet denoising (WT), variational mode decomposition (VMD), and long short-term memory networks (LSTM). This innovative nonlinear ensemble model effectively extracts key time series features and removes noise, while an introduced attention mechanism enhances the model’s focus on essential time steps, improving prediction accuracy. Experimental results demonstrate that the multi-model fusion approach significantly outperforms single models in terms of accuracy and stability, achieving mean absolute error (MAE) and root mean square error (RMSE) of 0.0117 and 0.0194, respectively. The proposed method offers significant advantages for CO2 prediction in greenhouse crops, providing a theoretical basis and technical support for optimizing and controlling greenhouse parameters. This contributes to the advancement of smart agriculture by offering an efficient environmental monitoring and prediction tool. Additionally, the study presents new ideas and technical solutions for addressing similar agricultural environment prediction challenges, optimizing greenhouse environment control strategies, and improving crop production efficiency.

Building high-resolution projections of temperature potential changes using statistical downscaling for the future period 2026-2100 in the Highland region of Yemen – A supportive approach for empowering environmental planning and decision-making

Environmental resources and ecological systems are significantly affected by the steady rise of the global temperature. However, the degree of temperature change at the regional and local levels is uncertain. The uncertainty arises from various factors, but mostly due to the short length of ground data and dependency of local studies on the large-scale and spatially coarse output of Global Climate Models (GCMs). Therefore, the output of GCM cannot be directly used in impact assessment studies at a regional and local level. In this study, the Statistical Down-Scaling Model (SDSM) is employed to investigate the magnitude of temperature changes (Minimum and Maximum Temperature) for the future period 2026-2100. The SDSM builds relationships between large-scale predictors and local climate variables, allowing for finer-resolution projections at a regional level. The study utilized the Climate Hazard Infra-Red Temperature with Station (CHIRTS-daily) to complete daily missing records in more than 90 ground stations. Additionally, predictors of the National Center for Environmental Prediction (NCEP) for the historical period (1961-2010) and the Canadian Earth System Model (CanESM2) for the future period (2026-2100) are employed to calibrate SDSM and to build finer-resolution scenarios under two representative concentration pathways; RCP2.6 and RCP8.5. The methodology additionally involved validating the SDSM performance using observed historical data before applying it to future projections. The findings indicate that both minimum and maximum temperatures (T-min and T-max) will increase, with a more pronounced rise in minimum temperature (T-min). Over the future period (2026-2100), the projected average temperature rise is 1.10°C (T-max) and 1.43°C (T-min) under RCP2.6. For RCP8.5, the projected average increases are 1.56°C and 2.3°C for T-max and T-min, respectively. Overall, the most significant increase is projected to occur in the 2090s (2076-2100) under RCP8.5, particularly in the lowlands and wadis of Al Mahwit and Raymah governorate. In these areas, the minimum temperature (T-min) exhibited an increased absolute value of up to 3.2°C. This high rise in temperatures is expected to result in increased evapotranspiration, prolonged droughts, and possibly breakouts of some plant diseases and pests. This would require effective adaptation measures such as harvesting rainwater and growing short-time and heat-resistance crops. Engaging in field visits and social discussions added depth to the study by introducing various traditional methods and indigenous practices. Valuable resources for future efforts to mitigate the potential impacts of climate change are offered by these insights.

Global Genotype by Environment Prediction Competition Reveals That Diverse Modeling Strategies Can Deliver Satisfactory Maize Yield Estimates.

Predicting phenotypes from a combination of genetic and environmental factors is a grand challenge of modern biology. Slight improvements in this area have the potential to save lives, improve food and fuel security, permit better care of the planet, and create other positive outcomes. In 2022 and 2023 the first open-to-the-public Genomes to Fields (G2F) initiative Genotype by Environment (GxE) prediction competition was held using a large dataset including genomic variation, phenotype and weather measurements and field management notes, gathered by the project over nine years. The competition attracted registrants from around the world with representation from academic, government, industry, and non-profit institutions as well as unaffiliated. These participants came from diverse disciplines include plant science, animal science, breeding, statistics, computational biology and others. Some participants had no formal genetics or plant-related training, and some were just beginning their graduate education. The teams applied varied methods and strategies, providing a wealth of modeling knowledge based on a common dataset. The winner's strategy involved two models combining machine learning and traditional breeding tools: one model emphasized environment using features extracted by Random Forest, Ridge Regression and Least-squares, and one focused on genetics. Other high-performing teams' methods included quantitative genetics, machine learning/deep learning, mechanistic models, and model ensembles. The dataset factors used, such as genetics; weather; and management data, were also diverse, demonstrating that no single model or strategy is far superior to all others within the context of this competition.

Optimizing indoor environmental prediction in smart buildings: A comparative analysis of deep learning models

This paper presents a comprehensive investigation into the application of deep learning models for predicting indoor environmental quality in smart buildings. Using data collected from a network of microclimate sensors deployed across a university campus in Sydney, we evaluated the performance of Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and hybrid CNN-LSTM models. Our study encompassed various aspects of model development, including data preparation, architecture design, hyperparameter optimization, and model interpretability. Contrary to common assumptions in time series forecasting, our results demonstrate that CNN models consistently outperformed LSTM and hybrid models in predicting indoor temperature. We found that multivariate input configurations enhanced prediction accuracy across all model types, highlighting the importance of capturing complex interactions between environmental parameters. Through SHapley Additive exPlanations (SHAP) analysis, we identified temperature, humidity, and Heating, Ventilation, and Air Conditioning (HVAC) status as the most influential features for predictions. Our experiments also revealed optimal configurations for historical input length and prediction horizon, providing practical guidelines for model implementation. This research contributes valuable insights for the development of more efficient and accurate smart building management systems, potentially leading to improved energy efficiency and occupant comfort in built environments.

Predicting adaptation and evolution of plasticity from temporal environmental change

Abstract Environmental change can drive evolutionary adaptation and determine geographic patterns of biodiversity. Yet at a time of rapid environmental change, our ability to predict its evolutionary impacts is incomplete. Temporal environmental change, in particular, involves a combination of major components such as abrupt shift, trend, cyclic change, and noise. Theoretical predictions exist for adaptation to isolated components, but knowledge gaps remain regarding their joint impacts. We extend classic evolutionary theory to develop a model for the evolution of environmental tolerance by the evolution of an underlying developmentally plastic trait, in response to major components of temporal change. We retrieve and synthesise earlier predictions of responses to isolated components, and generate new predictions for components changing simultaneously. Notably, we show how different forms of environmental predictability emerging from the interplay of cyclic change, stochastic change (noise) and lag between development and selection shape predictions. We then illustrate the utility of our model for generating testable predictions for the evolution of adaptation and plasticity when parameterised with real time series data. Specifically, we parameterise our model with daily time series of sea‐surface temperature from a global marine hotspot in southern Australia, and use model simulations to predict the evolution of thermal tolerance, and geographic differences in tolerance, in this region. By synthesising theory on evolutionary adaptation to temporal environmental change, and providing new insights into the joint effects of its different components, our framework, embedded in a Shiny app, offer a path to better predictions of biological responses to climate change.

Lightning-Fast Convective Outlooks: Predicting Severe Convective Environments With Global AI-Based Weather Models.

Severe convective storms are among the most dangerous weather phenomena and accurate forecasts mitigate their impacts. The recently released suite of AI-based weather models produces medium-range forecasts within seconds, with a skill similar to state-of-the-art operational forecasts for variables on single levels. However, predicting severe thunderstorm environments requires accurate combinations of dynamic and thermodynamic variables and the vertical structure of the atmosphere. Advancing the assessment of AI-models toward process-based evaluations lays the foundation for hazard-driven applications. We assess the forecast skill of the top-performing AI-models GraphCast, Pangu-Weather and FourCastNet for convective parameters at lead-times up to 10days against reanalysis and ECMWF's operational numerical weather prediction model IFS. In a case study and seasonal analyses, we see the best performance by GraphCast and Pangu-Weather: these models match or even exceed the performance of IFS for instability and shear. This opens opportunities for fast and inexpensive predictions of severe weather environments.

Machine Learning-Based Morel Mushroom Environmental Monitoring System

As living standards improve, the demand for edible fungi increases, making their efficient cultivation crucial for food supply and agricultural productivity. Traditional manual monitoring has limited accuracy, but the introduction of IoT and embedded technology enhances the intelligence and precision of agricultural monitoring, saving both labor and resources. This study designs a machine learning-based environmental monitoring and prediction system for edible fungi, utilizing low-power ZigBee networks and Kalman filtering to improve data accuracy. The system ensures real-time data transmission via the MQTT protocol and enhances security and scalability through cloud storage. By comparing Transformer, LSTM, and LSTM-Attention models, it was found that the Transformer model performed best in predicting environmental parameters, enabling proactive regulation, boosting yield and quality, and promoting the development of intelligent agriculture.

Improvement in the Low Temperature Prediction Skill During Cold Winters Over the Mid–High Latitudes of Eurasia in CFSv2

ABSTRACTRegional cold winters have occurred frequently in Eurasia since the beginning of the 21st century, increasing the interannual variability in winter temperatures and increasing the difficulty of prediction. In this study, we evaluate the performance of Climate Forecast version 2 (CFSv2) of the National Centers for Environmental Prediction (NCEP) in predicting winter temperature anomalies over the Northern Hemisphere and find that CFSv2 has significantly lower temperature prediction ability for cold winters in the mid–high latitudes of Eurasia since the 21st century. This is mainly due to the stronger response to global warming and the weaker response to sea ice anomalies in the preceding autumn in CFSv2 than the in reanalysis. Accordingly, two targeted correction methods have been developed to improve the prediction ability, with the first method removing the linear temperature trend of CFSv2 predictions and the second method considering the effects of autumn Arctic Sea ice anomalies via a dynamical–statistical correction approach (DSCA). Both methods can effectively improve the prediction ability of winter temperature anomalies in the mid–high latitudes of Eurasia, especially in cold winters. The anomaly correlation coefficient (ACC) increased from −0.03 to 0.13 before and after the modification by the DSCA, and from −0.12 to 0.25 for cold winters. The DSCA significantly reduced the root mean square error (RMSE) of the CFSv2 predictions by approximately 10%.

Advanced Trans-BiGRU-QA Fusion Model for Atmospheric Mercury Prediction

With the deepening of the Industrial Revolution and the rapid development of the chemical industry, the large-scale emissions of corrosive dust and gases from numerous factories have become a significant source of air pollution. Mercury in the atmosphere, identified by the United Nations Environment Programme (UNEP) as one of the globally concerning air pollutants, has been proven to pose a threat to the human environment with potential carcinogenic risks. Therefore, accurately predicting atmospheric mercury concentration is of critical importance. This study proposes a novel advanced model—the Trans-BiGRU-QA hybrid—designed to predict the atmospheric mercury concentration accurately. Methodology includes feature engineering techniques to extract relevant features and applies a sliding window technique for time series data preprocessing. Furthermore, the proposed Trans-BiGRU-QA model is compared to other deep learning models, such as GRU, LSTM, RNN, Transformer, BiGRU, and Trans-BiGRU. This study utilizes air quality data from Vietnam to train and test the models, evaluating their performance in predicting atmospheric mercury concentration. The results show that the Trans-BiGRU-QA model performed exceptionally well in terms of Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared (R2), demonstrating high accuracy and robustness. Compared to other deep learning models, the Trans-BiGRU-QA model exhibited significant advantages, indicating its broad potential for application in environmental pollution prediction.

THE PERCEPTION OF THE INSTITUTIONAL ENVIRONMENT IN THE ROMANIAN AGRICULTURAL SECTOR: CONNECTION BETWEEN BEHAVIOUR AND FINANCIAL INDICATORS

This study analyzes the perception of Romanian farmers regarding the predictability of the institutional environment and its connection with the perception of the importance of subsidies and the financial performance they obtain. Excessive bureaucracy, frequent changes in legislation and misalignment of the objectives of various stakeholders are limiting factors in financing programs, farmers being discouraged from applying for them. The study uses the statistical method of unidirectional analysis of variances (ANOVA) to observe significant differences in the means for the perception of the institutional environment and performance indicators, and the tests are carried out using the SPSS software, on a sample of 201 companies operating in the vegetable agricultural sector. The results of the study show that farmers' perception of the influence of subsidies seems to be subjective, based on the effort made to obtain them and less on the actual result of their use. Farmers who achieve good results in terms of return on assets and have a good liquidity position, consider the institutional environment to be predictable one, which supports them in their development and therefore, can make plans based on existing policies. The lack of confidence can be justified by poor financial results in general, the farmers’ expectations being to receive aid if the agricultural year is not a very good one.

Constructing a 22-year internal wave dataset for the northern South China Sea: spatiotemporal analysis using MODIS imagery and deep learning

Abstract. Internal waves (IWs) are an important ocean phenomenon facilitating energy transfer between multiscale ocean processes. Understanding such processes necessitates the collection and analysis of extensive observational data. IWs predominantly occur in marginal seas, with the South China Sea (SCS) being one of the most active regions, characterized by frequent and large-amplitude IW activities. In this study, we present a comprehensive IW dataset for the northern SCS (https://doi.org/10.12157/IOCAS.20240409.001, Zhang and Li, 2024), covering the area from 112.40 to 121.32° E and from 18.32 to 23.19° N, spanning the period from 2000 to 2022 with a 250 m spatial resolution. During the 22 years, a total of 15 830 MODIS images were downloaded for further processing. Out of these, 3085 high-resolution MODIS true-color images were identified to contain IW information and were included in the dataset with precise IW positions extracted using advanced deep learning techniques. IWs in the northern SCS are categorized into four regions based on extracted IW spatial distributions. This classification enables detailed analyses of IW characteristics, including their spatial and temporal distributions across the entire northern SCS and its specific sub-regions. Interestingly, our temporal analysis reveals characteristic “double-peak” patterns aligned with the lunar day, highlighting the strong connection between IWs and tidal cycles. Furthermore, our spatial analysis identifies two IW quiescent zones within the IW clusters influenced by underwater topography, highlighting regional variations in IW characteristics and suggesting underlying mechanisms which merit further investigation. There are also three gap regions between distinct IW clusters, which may indicate different IW sources. The constructed dataset holds significant potential for studying IW–environment interactions, developing monitoring and prediction models, validating numerical simulations, and serving as an educational resource to promote awareness and interest in IW research.