Latest preprints

Predicting microporosity and permeability in clastic reservoirs is a challenge in reservoir quality assessment, especially in formations where direct measurements are difficult or expensive. These reservoir properties are fundamental in determining a reservoir's capacity for fluid storage and transmission, yet conventional methods for evaluating them, such as Mercury Injection Capillary Pressure (MICP) and Scanning Electron Microscopy (SEM), are resource-intensive. The aim of this study is to develop a cost-effective machine learning model to predict complex reservoir properties using readily available field data and basic laboratory analyses. A Random Forest classifier was employed, utilizing key geological parameters such as porosity, grain size distribution, and spectral gamma-ray (SGR) measurements. An uncertainty analysis was applied to account for natural variability, expanding the dataset, and enhancing the model's robustness. The model achieved a high level of accuracy in predicting microporosity (93%) and permeability levels (88%). By using easily obtainable data, this model reduces the reliance on expensive laboratory methods, making it a valuable tool for early-stage exploration, especially in remote or offshore environments. The integration of machine learning with uncertainty analysis provides a reliable and cost-effective approach for evaluating key reservoir properties in siliciclastic formations. This model offers a practical solution to improve reservoir quality assessments, enabling more informed decision-making and optimizing exploration efforts.

There is evidence that an increase in plant-emitted isoprene causes an increase in cloud condensation nuclei, and is one way in which an increase in cloud cover occurs. Similarly, an increase in plant-emitted water vapour from transpiration can cause an increase in wind speed, which can transfer more heat out of the atmosphere into the ocean. Each of these pathways can reduce global atmospheric surface temperature. With this background, we investigated whether there was empirical evidence that isoprene and transpiration were components of the natural atmospheric temperature control system for which we have previously provided empirical evidence[1][2].Using the statistical hypothesis test of Granger causality as well as autoregressive distributed lag (ARDL) modelling we found generally very highly statistically significant evidence that isoprene and transpiration are components of the control system, and that they are placed in the early part of the control system chain immediately after the lead element, the global biota. Prior evidence[1][2] indicates that the information used by the control system to detect a disturbance to the setpoint of its outcome – global surface temperature – is a combination of the level of atmospheric CO2, its integral, and its first and second derivatives. Evidence is presented that this information pattern is created in each plant that participates in the control system and is impressed on isoprene and water vapour before they are emitted from the stomata of leaves to enter the atmosphere.

Effective resource management and environmental planning in regions with high climatic variability, such as Chile, demand advanced predictive tools. The success in these areas heavily relies on accurately interpreting and forecasting climatic patterns. This study addresses these challenges by employing an innovative and computationally efficient hybrid methodology that integrates machine learning (ML) methods for time series forecasting with established statistical techniques. The spatiotemporal data undergo decomposition using time-dependent Empirical Orthogonal Functions (EOFs), denoted as $\phi_k(t)$, and their corresponding spatial coefficients, $\alpha_k(s)$, to reduce dimensionality. Wavelet analysis provides high-resolution time and frequency information from the $\phi_k(t)$ functions, while neural networks forecast these functions within a medium-range horizon $h$. By utilizing various ML models, particularly a Wavelet–ANN hybrid model, we forecast $\phi_k(t+h)$ up to a time horizon $h$, and subsequently reconstruct the spatiotemporal data using these extended EOFs.This methodology is applied to a grid of climate data comprising 6355 points covering the entire territory of Chile. It transitions from a high-dimensional multivariate spatiotemporal data forecasting problem (involving 6355 time series) to a low-dimensional univariate time series forecasting problem (requiring only a few dozen forecasts). Additionally, cluster analysis with Dynamic Time Warping for defining similarities between rainfall time series, along with spatial coherence and predictability assessments, has been instrumental in identifying geographic areas where model performance is enhanced. This approach also elucidates the reasons behind poor forecast performance in regions or clusters with low spatial coherence and predictability. By utilizing cluster medoids, the forecasting process becomes more practical and efficient. This compound approach significantly reduces computational complexity while generating forecasts of reasonable accuracy and utility.Significance Statement. The approach outlined in this study facilitates the transition from a high–dimensional multivariate spatiotemporal data forecasting problem to a low-dimensional univariate time series forecasting problem. This transition substantially reduces computational complexity while yielding reasonably accurate forecasts and enhances our ability to interpret and predict climatic patterns across the entire territory and over medium-term temporal horizons, despite its high climatic variability.

Clouds play a key role in Earth’s radiation balance with complex effects that introduce large uncertainties into climate models. Real-time 3D cloud data is essential for improving climate predictions. This study leverages geostationary imagery from MSG/SEVIRI and radar reflectivity measurements of cloud profiles from CloudSat/CPR to reconstruct 3D cloud structures. We first apply self-supervised learning (SSL) methods—Masked Autoencoders (MAE) and geospatially-aware SatMAE-on unlabelled MSG images, and then fine-tune our models on matched image-profile pairs. Our approach outperforms state-of-the-art methods like U-Nets, and our geospatial encoding further improves prediction results, demonstrating the potential of SSL for cloud reconstruction.

Floods are natural disasters that can cause significant property damage and sometimes result in loss of life. In Malaysia, floods occur every year, particularly on the East Coast of Peninsular Malaysia, due to the Northeast Monsoon and the impacts of climate change, which can lead to heavy rainfall at the end of the year. Temerloh, a district in Pahang, frequently experiences flooding events, especially between November and January. Despite various efforts in flood mitigation and preparation, the damage to both citizens and property each year results in costs amounting to thousands of Ringgits and the time needed to clean up the aftermath of floods. To address this issue, this research examined the hydrological and meteorological factors contributing to the floods in Temerloh and developed a machine-learning model capable of predicting future flood occurrences. The study utilized a dataset from the National Hydrological Network Management System (SPRHiN), which includes hydrological data and meteorological information for the specific location. The correlation analysis revealed a strong relationship between stream flow and water level to floods, with correlation coefficients (r values) of 0.83 and 0.76, respectively. In contrast, temperature exhibited an inverse relationship with floods, showing a correlation value of -0.28; this suggests that lower temperatures are associated with a higher likelihood of rain and subsequent flooding. The results indicated that the model, developed using an artificial neural network (ANN), achieved an impressive accuracy of 0.9909 and demonstrated strong performance, as evidenced by an area under the receiver operating characteristic (ROC) curve (AUC) value of 0.888. The model also exhibited low error rates, with a mean squared error (MSE) of 0.009 and a root mean squared error (RMSE) of 0.096. Additionally, the R² value of 0.768 and the F1 score of 0.875 indicate that the model possesses high precision and recall. Furthermore, a flood monitoring dashboard was created to provide interactive data visualization. This research is essential for understanding the factors contributing to flooding in Pahang and will offer valuable insights for future studies on floods.

This manuscript develops and integrates the previous studies “Analytical Methods for Tracking Bodies Motions on the Lunar Surface in Apollo XVI Footage” https://doi.org/10.32388/IA8MXE, “Ballistic motion of dust particles in the “Collecting the Big Muley lunar rock sequence of the Apollo XVI Footage” https://doi.org/10.32388/COXHKG.2, “Falling Objects and Dust Particles' Motion in the Collecting Lunar Rock on the Buster Crater sequence of the Apollo XVI Footage” https://doi.org/10.32388/RFR2D5, in order to introduce a robust analytical method to trace and analyze the movement of dust shot during the Apollo XVI mission on the lunar surface. By employing both 2D and 3D analysis techniques, we aim to provide a detailed comparison of the observed kinematic events against theoretical models.The paper extends a previous work focused on the kinematics of lunar dust utilizing footage from the Apollo XVI mission "Ballistic motion of dust particles in the Lunar Roving Vehicle dust trails" published in 2012 in the American Journal of Physics by Mihaly Horanyi and Hsiang-Wen Hsu: https://www.researchgate.net/publication/258468670 [Ann. 1 – Ann. 2].In this further analysis, the famous sequence of the LRV “Grand Prix”, a performance test recorded by the Maurer DAC Camera at the end of EVA 1, is tracked and analyzed with the professional tools adopted in the other sections of the study. The behavior of the so-called "Roster Tails", the dust trails raised by the wheels of the LR during its march, is compared with the motion models related to the lunar environment. A jump of the vehicle that keeps it in free fall for about 1 second is analyzed and starting from the fit tests of the detected data, conclusions are drawn about the characteristics of the moving objects and about the environment in which the sequence takes place.

The elementary theory of relaxation of the source “cooling down” after the main shock of an earthquake is presented axiomatically. The names of the objects under study are given and the relationships between them are determined. A new basic concept of “earthquake source deactivation” is introduced and a procedure for calculating the deactivation coefficient from aftershock frequency measurement data is indicated. An important property of the system is that the axioms do not contain statements regarding the actual process of source relaxation. From two simple axioms a number of meaningful statements (theorems) logically follow. A two-stage mode of source relaxation was discovered. The sharp transition between stages has the character of a bifurcation. It is shown that the classical Omori law has limited applicability. It describes the evolution of aftershocks only at the first stage of relaxation. The well-known Hirano-Utsu law is not applicable to describe aftershocks either at the first or second stages of relaxation. The conclusions of the elementary theory allow for interesting generalizations that expand the possibilities of experimental and theoretical study of the source.