Image enhancement (IE) technology can help enhance the brightness of remote-sensing images to obtain better interpretation and visualization effects. Convolutional neural networks (CNN), such as Low-light CNN (LLCNN) Super-resolution (SRCNN), have achieved great success in image enhancement, super resolution, other image-processing applications. Therefore, we adopt propose a new network architecture with end-to-end strategy for low-light IE, named (RSCNN). In RSCNN, an upsampling operator is...

Accurate prediction of mineral resources is imperative to meet the energy demands modern society. Nonetheless, this task often difficult due estimation bias and limited interpretability conventional statistical techniques machine learning methods. To address these shortcomings, we propose a novel geospatial artificial intelligence approach, denoted as geographically neural network-weighted logistic regression, for prospectivity mapping. This model integrates spatial patterns networks,...

The integration of big data, cloud models, and extensive knowledge to drive new discovery through data is a paradigm for research in the field Earth sciences. Although advancement technologies infrastructures has simplified acquisition, deep-time geoscience still faces challenges such as fragmented difficulties visualization, insufficient computing power. To assist broad community geoscientists, we propose "Deep Platform," one-stop online platform that utilizes advanced technologies....

Super-resolution (SR) is able to improve the spatial resolution of remote sensing images, which critical for many practical applications such as fine urban monitoring. In this paper, a new single-image SR method, deep gradient-aware network with image-specific enhancement (DGANet-ISE) was proposed images. First, DGANet model complex relationship between low- and high-resolution A loss designed in training phase preserve more gradient details super-resolved Then, ISE approach testing further...

Sea surface temperature (SST) is a key factor in the marine environment, and its accurate forecasting important for climatic research, ecological preservation, economic progression. Existing methods mostly rely on convolutional networks, which encounter difficulties encoding irregular data. In this paper, allowing comprehensive of data containing land islands, we construct graph structure to represent SST propose memory neural network (GMNN). The GMNN includes encoder built upon iterative...

Research on the carbon cycle of coastal marine systems has been wide concern recently. Accurate knowledge temporal and spatial distributions sea-surface partial pressure (pCO2) can reflect seasonal heterogeneity CO2 flux is, therefore, essential for quantifying ocean's role in cycling. However, it is difficult to use one model estimate pCO2 determine its controlling variables an entire region due prominent spatiotemporal areas. Cubist a commonly-used zoning; thus, be applied estimation...

Lakes have been identified as an important indicator of climate change and a finer lake area can better reflect the changes. In this paper, we propose effective unsupervised deep gradient network (UDGN) to generate higher resolution from remote sensing images. By exploiting power learning, UDGN models internal recurrence information inside single image its corresponding map images with spatial resolution. The is derived input provide geographical information. Since training samples are only...

Abstract The Pacific decadal oscillation (PDO) is a variability phenomenon occurring in the North Ocean. It has substantial impacts on marine ecosystems and global climate. Due to high complexity unclear evolution mechanism, accurate long‐term prediction of PDO remains challenge. In this paper, deep spatiotemporal embedding network (DSEN) proposed extract features from historical climate data achieve end‐to‐end forecasting index. are recursive continuous index seasonal time scales, thus...

Abstract Variations in the Pacific decadal oscillation (PDO) can influence marine ecosystems and regional climate phenomena. Accurate long‐term forecasts of PDO are therefore crucial for governance. This paper presents a novel seasonal gated recurrent unit (SGRU) model, based on deep learning, forecasting at multiple time scales. The model first decomposes complex nonlinear index series into three separate components, each retaining distinct pattern PDO. Next, three‐pathway GRU is...

As an important parameter to characterize physical and biogeochemical processes, sea surface salinity (SSS) has received extensive attention. Cubist is a data mining model, which can be well-suited estimate analyze SSS in the Gulf of Mexico (GOM) because it reflect internal heterogeneity GOM—overall circular distribution, seasonality related temperature river discharge changes. Using remote sensing reflectance (Rrs) at 412, 443, 488 (490), 555, 667 (670) nm (SST), cubist model was developed...

Traffic guidance, traffic management and emergency vehicle all require keeping abreast of status. Intelligent Transportation Systems (ITS) is highly expected to provide real-time condition information service. To achieve this, the capability handling dynamic data stream collected from multi monitoring sources serving public with timely essential for ITS. With wide spread Internet Things technology, not only amount, but also spatial temporal resolutions have explosive growth, thereby...

The spatial learned index constructs a by learning the distribution, which performs lower cost of storage and query than indices. current update strategies indices can only solve limited updates at performance. We propose novel structure based on Block Range Index (SLBRIN for short). Its core idea is to cooperate history range satisfy fast efficient simultaneously. SLBRIN deconstructs transaction into three parallel operations optimizes them temporal proximity distribution. also provides...

Mesoscale eddies are characterized by swirling currents spanning from tens to hundreds of kilometers in diameter three-dimensional attributes holds paramount significance driving advancements both oceanographic research and engineering applications. Nonetheless, a notable absence models capable adeptly harnessing the scarcity high-quality annotated marine data, efficiently discern morphological mesoscale eddies, is evident. To address this limitation, paper constructs an innovative...

Raster data represent one of the fundamental formats utilized in GIS. As technology used to observe Earth continues evolve, spatial and temporal resolution raster is becoming increasingly refined, while scale expanding. One key issues development GIS determine how make large-scale better provide computation, visualization, analysis services Internet environment. This paper proposes a decentralized COG-pyramid-based map service method (DCPMS). In comparison traditional online technology, such...

Batch paleogeographic point rotation (BPPR) is a PySpark-based extensible batch data method that accelerates during reconstruction. Data an important part of reconstruction and significant tool for exploring the co-evolution Earth life. However, current techniques have challenges with processing speeds when handling extensive data. Therefore, this study introduced parallel-computing framework to construct BPPR. This combines PySpark PyGPlates, which can partition points compute them...

The cross-fertilization of the fast-developing AI technology and spatial indexing has given rise to learned indexes. However, these indexes rely on historical data distributions build models, which limits their ability anticipate that not yet arrived. To address this, we propose a novel Spatio-Temporal Update Method (STUM) enhances conventional by introducing Spatial Delta Area (SDA) for updates without altering hierarchical structure. STUM learns spatio-temporal auto-correlation from...

Abstract. Fault activities modelling holds vital importance for earthquake monitoring, risk management, and early alert. Studies on laboratory earthquakes are instrumental in the of natural fault ruptures enhancing our grasp dynamics. Recently, deep learning methods have been proven effective predicting instantaneous stress settings slow slip events Earth. However, these struggled to conduct steady future prediction lacking grasping complex dynamics highly nonlinear systems. Addressing this,...

The integration of deep-time databases is a key aspect in the creation comprehensive digital model Earth's history, known as "deep-time Earth." This would enable scientists to better comprehend intricate processes that have shaped our planet and its life forms over time. Currently, data are dispersed across various research institutions, making it challenging for fully utilize information contained data.We present virtual databases, where remain sources accessed needed at query We will...

Abstract Young people are the backbone of urban development and an important pillar social stability. The growth young floating population in China has given rise to land expansion. Understanding life pattern for benefits rational effective In this article, we introduce food delivery data into process exploring behavioral patterns youth Hangzhou, Zhejiang Province, China. dynamic time warping (DTW) distance-based k-medoids method is applied explore main activity areas population. results...