Spatio-temporal Datasets Research Articles

Evaluation coupling-coordination relationship between car anthropogenic heat release and local temperature at multi-spatiotemporal scales

Car anthropogenic heat (CAH) release is a major factor affecting local temperature. However, the coupling-coordination (CC) relationship between CAH release and local temperature is not fully understood, owing to the absence of a quantitative relationship. Therefore, we selected winter season in 2008 in Beijing to conduct our research. We used trajectory data during 3th Feb 2008 and 7th Feb 2008 to build a high spatiotemporal resolution dataset of CAH release, with spatial resolution of 300 m. The coupling-coordination relationship and threshold were measured. The results showed that four categories CC degree temporal patterns were observed in Beijing winter season: near-straight line pattern, one peak-valley pattern, two peak-valley pattern with high value, and two-peak-valley pattern with relative low value. A U-shaped relationship between local temperature and CC degree was found in Beijing winter season. The U-shaped relationship was obvious on weekdays. The curve of the U shape was 0.3 on weekdays and less than 0.3 on weekends. This U-shaped relationship was also detected in Shenzhen, which validated our conclusions. The findings of this study provide suggestions for the development of strategies to mitigate climate change in Beijing winter seasons.

Spatial and Temporal Patterns of Maize Phenology in China From 2001 to 2020

AbstractClimate change has significantly altered crop phenology, which has further impacted crop growth and yield. Accurate monitoring of crop phenology is essential for managing agricultural production in response. However, regional monitoring requires high spatial resolution distribution data, as medium resolution data suffers from mixed pixel issues. This study based on a long‐term high spatiotemporal resolution fusion data set of Normalized Difference Vegetation Index and an annually updated maize distribution data set, used the relative threshold method to identify the maize phenology in 22 provinces of China from 2001 to 2020. We further analyzed the trend of maize phenology and assessed its responses to climate change. The results reveal large inter‐annual fluctuations and spatial variability in maize phenology from 2001 to 2020. The length of the growth season (LOS) of spring maize has prolonged by 4.28 days in the northern maize zone and has shortened by 4.90 days in the southern maize zone. Additionally, the LOS of summer maize in the Huang‐Huai‐Hai region has shortened by 2.24 days. We also found a positive correlation between the length of the vegetative growth stage and the mean temperature and a negative correlation between the length of the reproductive growth stage and accumulated precipitation. This study utilized large‐scale, high‐resolution maize phenology data to analyze the trend of maize phenology and its response to climate change. These findings are expected to provide valuable support for assessing maize growth status and developing agricultural adaptive practices.

Spatiotemporal distribution of global wind erosion over the past four decades

Abstract Wind erosion is a critical environmental issue that degrades land and air quality, affecting global ecosystems, agriculture, and human health. Yet, on the global scale, the long-term spatial variability and controlling factors of wind erosion remain highly uncertain. Here, we develop a high-resolution spatiotemporal dataset of global wind erosion from 1980 to 2020 using the Revised Wind Erosion Equation model, integrated with comprehensive meteorological, terrestrial ecology, and soil datasets. Our analysis indicates that wind erosion annually impacted 359 ± 25 petagrams of soils worldwide during this period. Approximately 70% of this erosion occurred in just ten countries, predominantly in Africa and the Middle East. Due to higher erosion intensities, pasturelands, accounting for 28% of all non-barren land use types, disproportionately contributed to 70% of the erosion in these areas. Furthermore, our analysis highlights an upward trend in global wind erosion over the past four decades, with affected areas expanding worldwide. Although our study reinforces that changing wind speeds and a drier climate are central factors impacting global wind erosion, we find that increasing erosion intensities in pasturelands may also exacerbate erosion in North Africa, South America, and East Asia. This has broad implications for soil erosion issues that impact food productivity, human health, and ecosystem stability. This research provides insights for developing wind erosion warnings and targeted mitigation strategies, supporting global efforts to combat environmental degradation and promote sustainable development.

A Cross Spatio-Temporal Pathology-based Lung Nodule Dataset

Recently, intelligent analysis of lung nodules with the assistant of computer aided diagnosis (CAD) techniques can improve the accuracy rate of lung cancer diagnosis. However, existing CAD systems and pulmonary datasets mainly focus on Computed Tomography (CT) images from one single period, while ignoring the cross spatio-temporal features associated with the progression of nodules contained in imaging data from various captured periods of lung cancer. If the evolution patterns of nodules across various periods in the patients’ CT sequences can be explored, it will play a crucial role in guiding the precise screening identification of lung cancer. Therefore, a cross spatio-temporal lung nodule dataset with pathological information for nodule identification and diagnosis is constructed, which contains 317 CT sequences and 2,295 annotated nodules from 109 patients. This comprehensive database is intended to drive research in the field of CAD towards more practical and robust methods, and also contribute to the further exploration of precision medicine related field. To ensure patient confidentiality, we have removed sensitive information from the dataset.

Scalable spatiotemporal prediction with Bayesian neural fields

Spatiotemporal datasets, which consist of spatially-referenced time series, are ubiquitous in diverse applications, such as air pollution monitoring, disease tracking, and cloud-demand forecasting. As the scale of modern datasets increases, there is a growing need for statistical methods that are flexible enough to capture complex spatiotemporal dynamics and scalable enough to handle many observations. This article introduces the Bayesian Neural Field (BayesNF), a domain-general statistical model that infers rich spatiotemporal probability distributions for data-analysis tasks including forecasting, interpolation, and variography. BayesNF integrates a deep neural network architecture for high-capacity function estimation with hierarchical Bayesian inference for robust predictive uncertainty quantification. Evaluations against prominent baselines show that BayesNF delivers improvements on prediction problems from climate and public health data containing tens to hundreds of thousands of measurements. Accompanying the paper is an open-source software package (https://github.com/google/bayesnf) that runs on GPU and TPU accelerators through the Jax machine learning platform.

Predicting Prayagraj's Urbanization Trajectory using CA-ANN Modelling: Population Pressures and Land Use Dynamics

Land Use/Land Cover (LULC) dynamics provide a crucial role in the monitoring, planning, and management of resources. They also offer valuable information for developing strategies to balance conservation efforts, resolve conflicts between different land uses, and address pressures from growth. The present study focuses on the assessment of LULC dynamics, their forecasting, and their changes for Prayagraj city (including its surroundings) of India. Using long-term spatiotemporal Landsat datasets (1988–2018), we have explored the interlinkages between the change dynamics and human population pressure to explore the impact of agriculture and urbanization on the city landscape. Future growth prediction is carried out by incorporating Cellular Automaton (CA) and Artificial Neural Network (ANN) models. Six exploratory layers (viz., roads, educational institutes, railway transition, slope, river, and restricted area) are used in the learning process to determine LULC change (1997–2008) simulation. The validation of real and predicted LULC is carried out for 2018, where the correctness percentage and kappa value are found to be 90.29% and 0.87, respectively. Then, the ANN- Multilayer perceptron (MLP) and CA model are applied to predict LULC–2028 using the same trained transition probabilities. Results show that Built-land has grown highly by 10.03%, whereas Agriculture land and Forest land have significantly decreased by 13.43% and 3.03%, respectively, from 1988 to 2018. The predicted LULC of 2028 reveals that Built-land will keep growing by 2.83% during 2018–2028 at the cost of Agriculture land and Forest land, especially in northern, south–western and southern region, including city's inner sphere. United Nations' human population projection reveals that the city is expected to reach a population of 1.625 million by 2028. This indicates that tremendous pressure will be placed on land resources, particularly on agricultural, barren, and forested areas. To address this alarming scenario, it is imperative to delineate future development areas, ensuring better urban planning for the environmental sustainability and economic prosperity of Prayagraj city.

GIATAR: a Spatio-temporal Dataset of Global Invasive and Alien Species and their Traits

Monitoring and managing the global spread of invasive and alien species requires accurate spatiotemporal records of species presence and information about the biological characteristics of species of interest including life cycle information, biotic and abiotic constraints and pathways of spread. The Global Invasive and Alien Traits And Records (GIATAR) dataset provides consolidated dated records of invasive and alien presence at the country-scale combined with a suite of biological information about pests of interest in a standardized, machine-readable format. We provide dated presence records for 46,666 alien taxa in 249 countries constituting 827,300 country-taxon pairs in locations where the taxon’s invasive status is either alien, invasive, or unknown, joined with additional biological information for thousands of taxa. GIATAR is designed to be quickly updateable with future data and easy to integrate into ongoing research on global patterns of alien species movement using scripts provided to query and analyze data. GIATAR provides crucial data needed for researchers and policymakers to compare global invasion trends across a wide range of taxa.

A Semantic and Motion-Aware Spatiotemporal Transformer Network for Action Detection.

This paper presents a novel spatiotemporal transformer network that introduces several original components to detect actions in untrimmed videos. First, the multi-feature selective semantic attention model calculates the correlations between spatial and motion features to model spatiotemporal interactions between different action semantics properly. Second, the motion-aware network encodes the locations of action semantics in video frames utilizing the motion-aware 2D positional encoding algorithm. Such a motion-aware mechanism memorizes the dynamic spatiotemporal variations in action frames that current methods cannot exploit. Third, the sequence-based temporal attention model captures the heterogeneous temporal dependencies in action frames. In contrast to standard temporal attention used in natural language processing, primarily aimed at finding similarities between linguistic words, the proposed sequence-based temporal attention is designed to determine both the differences and similarities between video frames that jointly define the meaning of actions. The proposed approach outperforms the state-of-the-art solutions on four spatiotemporal action datasets: AVA 2.2, AVA 2.1, UCF101-24, and EPIC-Kitchens.

Trusted artificial intelligence for environmental assessments: An explainable high-precision model with multi-source big data

Environmental assessments are critical for ensuring the sustainable development of human civilization. The integration of artificial intelligence (AI) in these assessments has shown great promise, yet the "black box" nature of AI models often undermines trust due to the lack of transparency in their decision-making processes, even when these models demonstrate high accuracy. To address this challenge, we evaluated the performance of a transformer model against other AI approaches, utilizing extensive multivariate and spatiotemporal environmental datasets encompassing both natural and anthropogenic indicators. We further explored the application of saliency maps as a novel explainability tool in multi-source AI-driven environmental assessments, enabling the identification of individual indicators' contributions to the model's predictions. We find that the transformer model outperforms others, achieving an accuracy of about 98% and an area under the receiver operating characteristic curve (AUC) of 0.891. Regionally, the environmental assessment values are predominantly classified as level II or III in the central and southwestern study areas, level IV in the northern region, and level V in the western region. Through explainability analysis, we identify that water hardness, total dissolved solids, and arsenic concentrations are the most influential indicators in the model. Our AI-driven environmental assessment model is accurate and explainable, offering actionable insights for targeted environmental management. Furthermore, this study advances the application of AI in environmental science by presenting a robust, explainable model that bridges the gap between machine learning and environmental governance, enhancing both understanding and trust in AI-assisted environmental assessments.

Maternal ozone exposure lowers infant’s birthweight: A nationwide cohort of over 4 million livebirths in Iran

BackgroundNationwide evidence linking maternal ozone exposure with fetal growth restriction (FGR) was extensively scarce, especially in the Middle East with dry climate and distinct religious culture. MethodsWe carried out a national retrospective birth cohort study using registry-based records from 749 hospitals across 31 provinces in Iran from 2013 to 2018. Monthly concentrations of maximum daily average 8-hour (MDA8) ozone at 0.125° × 0.125° resolution were extracted from well-validated spatiotemporal grid dataset. Linear and logistic regression models were employed to evaluate associations of maternal MDA8 ozone exposure with birthweight outcomes. Assuming causality, the comparative risk assessment framework was utilized to estimate the burden of low birthweight (LBW), small for gestational age (SGA), and birthweight loss per livebirth (BLL) attributable to ambient ozone pollution. ResultsOf 4030383 livebirths included in the study, 264304 (6.6%) were LBW and 484405 (12.0%) were SGA. Each 10-ppb increase in MDA8 ozone exposure was associated with an odds ratio of 1.123 (95% confidence interval [CI]: 1.104 to 1.142) for LBW and 1.210 (95% CI: 1.197 to 1.223) for SGA, and a 30.5-g (95% CI: 29.0 to 32.0) reduction in birthweight. We observed approximately linear exposure-response relationships of maternal MDA8 ozone exposure with LBW (Pnonlinear= 0.786), SGA (Pnonlinear= 0.156), and birthweight reduction (Pnonlinear= 0.104). Under the premise of causal association, we estimated 6.6% (95% CI: 5.7 to 7.5) of LBW, 10.1% (95% CI: 9.6 to 10.6) of SGA, and 18.8 g (95% CI: 17.9 to 19.7) of BLL could be attributable to maternal ozone exposure in Iran. Considerably greater risk and burden of ozone-related FGR were observed among younger, less-educated, and rural-dwelling mothers. ConclusionsOur study provided compelling evidence that maternal ozone exposure was associated with heightened FGR risk and burden, particularly among socioeconomically disadvantaged mothers. These findings underscored the urgent need for government to incorporate socioeconomic factors into future ozone-related health policies, not only to mitigate pollution, but also minimize inequality.

Evaluating MODIS cloud-free snow cover datasets using massive spatial benchmark data in the Tibetan Plateau

Accurate snow cover data is crucial for understanding climate change, managing water resources, and calibrating models. The MODIS (Moderate-resolution Imaging Spectroradiometer) and its cloud-free snow cover datasets are widely used, but they have not been systematically evaluated due to different benchmark data and evaluation parameters. Conventional methods using station observations as a ground truth suffer from underrepresentation and mismatches in temporal and spatial scales. This study established a scale-matched spatial benchmark dataset, compiling from 18,433 Landsat series and 11,172 Sentinel-2 images over two decades, totaling ∼1.86 billion samples and ∼320 million snow samples. We evaluated seven MODIS cloud-free snow cover datasets for seasons, elevation zones, land covers and subregions using this benchmark data. For the clear-sky part, NIEER_MODIS_SCE (MODIS snow cover extent product over China) performs best due to its use of optimal NDSI thresholds suitable for each land use type. This highlights the importance of regional customization in snow mapping algorithms, and it can be further improved in spring, forests and zone 1 by combining it with M*D10A1GL06. For the cloud removed part, one-step integrated spatiotemporal cloud removal datasets perform better than any other approach does. The second-best dataset is obtained from a simple but effective single temporal cloud removal method using nearby time information. For the whole dataset, the best NIEER_MODIS_SCE has an overall accuracy of 0.82 and snow retrieval accuracy of 84.56 %. It performs excellently in most settings but weakest in forests thus requiring more efficient strategies. This research provides new perspectives and methods for objectively assessing MODIS snow cover products and other relevant datasets. These methods can be readily extended to other regions and adapted to future satellite missions. And such findings may guide the selection of more valid snow cover data and the developing of even better snow detecting strategies.

Detection of spatiotemporal changepoints: a generalised additive model approach

The detection of changepoints in spatio-temporal datasets has been receiving increased focus in recent years and is utilised in a wide range of fields. With temporal data observed at different spatial locations, the current approach is typically to use univariate changepoint methods in a marginal sense with the detected changepoint being representative of a single location only. We present a spatio-temporal changepoint method that utilises a generalised additive model (GAM) dependent on the 2D spatial location and the observation time to account for the underlying spatio-temporal process. We use the full likelihood of the GAM in conjunction with the pruned linear exact time (PELT) changepoint search algorithm to detect multiple changepoints across spatial locations in a computationally efficient manner. When compared to a univariate marginal approach our method is shown to perform more efficiently in simulation studies at detecting true changepoints and demonstrates less evidence of overfitting. Furthermore, as the approach explicitly models spatio-temporal dependencies between spatial locations, any changepoints detected are common across the locations. We demonstrate an application of the method to an air quality dataset covering the COVID-19 lockdown in the United Kingdom.

PMSTD-Net: A Neural Prediction Network for Perceiving Multi-Scale Spatiotemporal Dynamics.

With the continuous advancement of sensing technology, applying large amounts of sensor data to practical prediction processes using artificial intelligence methods has become a developmental direction. In sensing images and remote sensing meteorological data, the dynamic changes in the prediction targets relative to their background information often exhibit more significant dynamic characteristics. Previous prediction methods did not specifically analyze and study the dynamic change information of prediction targets at spatiotemporal multi-scale. Therefore, this paper proposes a neural prediction network based on perceptual multi-scale spatiotemporal dynamic changes (PMSTD-Net). By designing Multi-Scale Space Motion Change Attention Unit (MCAU) to perceive the local situation and spatial displacement dynamic features of prediction targets at different scales, attention is ensured on capturing the dynamic information in their spatial dimensions adequately. On this basis, this paper proposes Multi-Scale Spatiotemporal Evolution Attention (MSEA) unit, which further integrates the spatial change features perceived by MCAU units in higher channel dimensions, and learns the spatiotemporal evolution characteristics at different scales, effectively predicting the dynamic characteristics and regularities of targets in sensor information.Through experiments on spatiotemporal prediction standard datasets such as Moving MNIST, video prediction dataset KTH, and Human3.6m, PMSTD-Net demonstrates prediction performance surpassing previous methods. We construct the GPM satellite remote sensing precipitation dataset, demonstrating the network's advantages in perceiving multi-scale spatiotemporal dynamic changes in remote sensing meteorological data. Finally, through extensive ablation experiments, the performance of each module in PMSTD-Net is thoroughly validated.

Satellite-Based Reconstruction of Atmospheric CO2 Concentration over China Using a Hybrid CNN and Spatiotemporal Kriging Model

Although atmospheric CO2 concentrations collected by satellites play a crucial role in understanding global greenhouse gases, the sparse geographic distribution greatly affects their widespread application. In this paper, a hybrid CNN and spatiotemporal Kriging (CNN-STK) model is proposed to generate a monthly spatiotemporal continuous XCO2 dataset over China at 0.25° grid-scale from 2015 to 2020, utilizing OCO-2 XCO2 and geographic covariates. The validations against observation samples, CAMS XCO2 and TCCON measurements indicate the CNN-STK model is effective, robust, and reliable with high accuracy (validation set metrics: R2 = 0.936, RMSE = 1.3 ppm, MAE = 0.946 ppm; compared with TCCON: R2 = 0.954, RMSE = 0.898 ppm and MAE = 0.741 ppm). The accuracy of CNN-STK XCO2 exhibits spatial inhomogeneity, with higher accuracy in northern China during spring, autumn, and winter and lower accuracy in northeast China during summer. XCO2 in low-value-clustering areas is notably influenced by biological activities. Moreover, relatively high uncertainties are observed in the Qinghai-Tibet Plateau and Sichuan Basin. This study innovatively integrates deep learning with the geostatistical method, providing a stable and cost-effective approach for other countries and regions to obtain regional scales of atmospheric CO2 concentrations, thereby supporting policy formulation and actions to address climate change.

Challenges of spatio-temporal trajectory datasets

ABSTRACT The rapid evolution of wireless location technologies has significantly increased the availability of mobility data, particularly spatio-temporal trajectory data, detailing object movements and locations over time. This data holds diverse applications such as predicting travel patterns, optimizing routes, analysing social interactions, and managing urban resources. However, leveraging trajectory datasets presents challenges, including data heterogeneity, volume, velocity requiring real-time processing, privacy concerns with sensitive information, and regulatory compliance. Traditional systematic reviews of literature may not fully capture industry challenges, as professionals often do not publish research. To address this gap, industry expert input is vital. Collaborating with academia, we propose using focus group discussions to explore these challenges comprehensively. Focus groups offer a qualitative platform for professionals and academics to identify practical issues and gaps in the literature. This combined approach will inform technological advancements in creating and utilizing spatio-temporal datasets. Additionally, integrating insights from focus groups with a literature review ensures a thorough analysis of data challenges. This methodology aims to provide a holistic understanding and foster innovation in managing and exploiting trajectory data effectively.

Scaling up time–geographic computation for movement interaction analysis

AbstractUnderstanding interactions through movement provides critical insights into urban dynamic, social networks, and wildlife behaviors. With widespread tracking of humans, vehicles, and animals, there is an abundance of large and high‐resolution movement data sets. However, there is a gap in efficient GIS tools for analyzing and contextualizing movement patterns using large movement datasets. In particular, tracing space–time interactions among a group of moving individuals is a computationally demanding task, which would uncover insights into collective behaviors across systems. This article develops a Spark‐based geo‐computational framework through the integration of Esri's ArcGIS GeoAnalytics Engine and Python to optimize the computation of time geography for scaling up movement interaction analysis. The computational framework is then tested using a case study on migratory turkey vultures with over 2 million GPS tracking points across 20 years. The outcomes indicate a drastic reduction in interaction detection time from 14 days to 6 hours, demonstrating a remarkable increase in computational efficiency. This work contributes to advancing GIS computational capabilities in movement analysis, highlighting the potential of GeoAnalytics Engine in processing large spatiotemporal datasets.

Floods and fire ants, Solenopsis invicta (Hymenoptera: Formicidae): the Australian experience

AbstractThe rafting behaviour of the red imported fire ant, Solenopsis invicta, in response to flooding events is well documented, although studies generally have focussed on the mechanisms of raft assembly and the behaviour of the raft's occupants. Flooding as a means of dispersal of S. invicta is frequently mentioned in the literature, although there are few data on the distances travelled or how effective it is compared to natural flight. In Australia, S. invicta is a priority invasive species with a national eradication program operating for 23 years, focussed on the population in southeast Queensland, which currently encompasses more than 700 000 ha. Flooding presents a risk to the success of the program through extending the infestation area or recolonising successfully treated areas. We used the program's extensive spatiotemporal dataset of known fire ant colony locations to assess the effects of two significant flood events on the dispersal or displacement of S. invicta in Queensland. Results indicated that flooding did not spread S. invicta beyond the known boundaries of infestation but contributed to localised spread, particularly for sites with known polygyne infestations. This situation could change if the ant spreads to new river catchments. A novel method developed to assess the risk of S. invicta dispersal through flooding is presented, alongside program actions that can be applied to mitigate this risk.

Unlocking massively parallel spectral proper orthogonal decompositions in the PySPOD package

We propose a parallel (distributed) version of the spectral proper orthogonal decomposition (SPOD) technique. The parallel SPOD algorithm distributes the spatial dimension of the dataset preserving time. This approach is adopted to preserve the non-distributed fast Fourier transform of the data in time, thereby avoiding the associated bottlenecks. The parallel SPOD algorithm is implemented in the PySPOD library and makes use of the standard message passing interface (MPI) library, implemented in Python via mpi4py. An extensive performance evaluation of the parallel package is provided, including strong and weak scalability analyses. The open-source library allows the analysis of large datasets of interest across the scientific community. Here, we present applications in fluid dynamics and geophysics, that are extremely difficult (if not impossible) to achieve without a parallel algorithm. This work opens the path toward modal analyses of big quasi-stationary data, helping to uncover new unexplored spatio-temporal patterns. Program summaryProgram Title: PySPODCPC Library link to program files:https://doi.org/10.17632/jf5bf26jcj.1Developer's repository link:https://github.com/MathEXLab/PySPODLicensing provisions: MIT LicenseProgramming language: PythonNature of problem: Large spatio-temporal datasets may contain coherent patterns that can be leveraged to better understand, model, and possibly predict the behavior of complex dynamical systems. To this end, modal decomposition methods, such as the proper orthogonal decomposition (POD) and its spectral counterpart (SPOD), constitute powerful tools. The SPOD algorithm allows the systematic identification of space-time coherent patterns. This can be used to understand better the physics of the process of interest, and provide a path for mathematical modeling, including reduced order modeling. The SPOD algorithm has been successfully applied to fluid dynamics, geophysics and other domains. However, the existing open-source implementations are serial, and they prevent running on the increasingly large datasets that are becoming available, especially in computational physics. The inability to analyze via SPOD large dataset in turn prevents unlocking novel mechanisms and dynamical behaviors in complex systems.Solution method: We provide an open-source parallel (MPI distributed) code, namely PySPOD, that is able to run on large datasets (the ones considered in the present paper reach about 200 Terabytes). The code is built on the previous serial open-source code PySPOD that was published in https://joss.theoj.org/papers/10.21105/joss.02862.pdf. The new parallel implementation is able to scale on several nodes (we show both weak and strong scalability) and solve some of the bottlenecks that are commonly found at the I/O stage. The current parallel code allows running on datasets that was not easy or possible to analyze with serial SPOD algorithms, hence providing a path towards unlocking novel findings in computational physics.Additional comments including restrictions and unusual features: The code comes with a set of built-in postprocessing tools, for visualizing the results. It also comes with extensive continuous integration, documentation, and tutorials, as well as a dedicated website in addition to the associated GiHub repository. Within the package we also provide a parallel implementation of the proper orthogonal decomposition (POD), that leverages the I/O parallel capabilities of the SPOD algorithm.

Global area boom for greenhouse cultivation revealed by satellite mapping.

Greenhouse cultivation has been expanding rapidly in recent years, yet little knowledge exists on its global extent and expansion. Using commercial and freely available satellite data combined with artificial intelligence techniques, we present a global assessment of greenhouse cultivation coverage and map 1.3 million hectares of greenhouse infrastructures in 2019, a much larger extent than previously estimated. Our analysis includes both large (61%) and small-scale (39%) greenhouse infrastructures. Examining the temporal development of the 65 largest clusters (>1,500 ha), we show a recent upsurge in greenhouse cultivation in the Global South since the 2000s, including a dramatic increase in China, accounting for 60% of the global coverage. We emphasize the potential of greenhouse infrastructures to enhance food security but raise awareness of the uncertain environmental and social implications that may arise from this expansion. We further highlight the gap in spatio-temporal datasets for supporting future research agendas on this critical topic.

Reconstructing long-term natural flows by ensemble machine learning

Large-scale human activities have caused significant disruption to the natural flow regime of rivers worldwide, leading to substantial deviations from river flows’ natural state. The natural hydrological regime of rivers holds great significance for the conservation and management of riverine ecosystems, which relies on the accurate reconstruction of natural river flows. However, the application of existing high spatiotemporal resolution large-scale natural flow datasets at the regional scale has not been adequately assessed, thereby limiting their regional-scale utility. In this study, utilizing global reanalysis datasets and observation data, in combination with ensemble machine learning models, we constructed long-term natural flow time series for inflows into the Pearl River Delta and evaluated the performance of these reconstructed inflows based on observed flow records during undisturbed periods. The reconstruction shows that the ensemble model, the AutoGluon model, indicates the best performance when compared with other machine learning models and outperforms the existing publicly available large-scale datasets in terms of evaluation metrics for the corresponding natural flows. The results of Indicator of Hydrologic Alteration and Range of Variability Analysis (IHA-RVA) indicate that climate variability, reservoir regulation and human activities have significantly altered the natural flow regime of the Pearl River Delta. Specifically, human activities generally tend to smooth natural flow variability in the Pearl River Delta. This research is expected to provide an efficient and reliable method for reconstructing natural flows and offers guidance for water resource management and ecosystem conservation.