Search

The need for adequate data on the spatial and temporal variability of freshwater resources is a significant challenge to the water managers of the world in water resource planning and management. The problems will be acute in the coming years because of the increase in frequency and intensity of hydrologic extremes due to climate change. Therefore, streamflow prediction has become an important area of research because of its importance in flood mitigation, reservoir operation, and water resource management. In this paper, we have tested four Machine Learning models (ML models): Support Vector Machines (SVM), Random Forest (RF), Long Short-Term Memory (LSTM), and Multivariate Adaptive Regression Splines (MARS) for streamflow prediction at daily and monthly time scales in three rivers draining in the different climatic and geological settings. The SVM, RF, LSTM, and MARS models have been trained and tested in the Suvarna, Aghanashini, and Kunderu River Basins in peninsular India. Model intercomparison was made to identify the best suitable model for streamflow prediction. The RF outperforms other models for daily streamflow, and MARS outperforms other models for monthly streamflow prediction in the Suvarna river with Nash-Sutcliffe efficiency (NSE) values of 0.676 and 0.924, respectively. SVM (NSE = 0.741) and RF (NSE = 0.826) are found to be the best models for daily and monthly streamflow prediction in the Aghanashini river. MARS outperformed other models in the case of high, severe, and extreme flow simulation with NSE values of 0.481, 0.374, and 0.455, respectively, in the Aghanashini river. Other hydrological variables (groundwater level data, antecedent soil moisture, potential evapotranspiration data) and a better spatial resolution of rainfall data can be used to develop more accurate machine-learning models for streamflow predictions. [ABSTRACT FROM AUTHOR]

Sedimentary basins, integral to Earth's geological history and energy resource exploration, undergo complex changes driven by sedimentation, subsidence and geological processes. Gravity anomaly inversion is a crucial technique offering insights into subsurface structures and density variations. Our study addresses the challenge of complex subsurface structure assessment by leveraging deep neural networks to invert observed gravity anomalies. Optimization approaches traditionally incorporate known density distributions obtained from borehole data or geological logging for inverting basement depth in sedimentary basins using observed gravity anomalies. Our study explores the application of deep neural networks in accurate architectural assessment of sedimentary basins and demonstrates their significance in mineral and hydrocarbon exploration. Recent years have witnessed a surge in the use of machine learning in geophysics, with deep learning models playing a pivotal role. Integrating deep neural networks, such as the feedforward neural networks, has revolutionized subsurface density distribution and basement depth estimation. This study introduces a deep neural network specifically tailored for inverting observed gravity anomalies to estimate two‐dimensional basement relief topographies in sedimentary basins. To enhance computational efficiency, a one‐dimensional discrete cosine transform based discretization approach is employed. Synthetic data, generated using non‐Gaussian fractals, compensates for the scarcity of true datasets for training the deep neural network model. The algorithm's robustness is validated through noise introduction with comparisons against an efficient and traditional global optimization‐based approach. Gravity anomalies of real sedimentary basins further validate the algorithm's efficacy, establishing it as a promising methodology for accurate and efficient subsurface imaging in geological exploration. [ABSTRACT FROM AUTHOR]

Directional derivative-based edge detectors are commonly used to outline the abrupt lateral variations in densities or magnetizations of geological sources from potential-field datasets. Some subtle traces of faults, lithological contacts, and lineaments can be made more visible on anomaly maps using these algorithms. However, these algorithms generally suffer from common limitations such as blurry edges with low resolution or faulty boundaries. To minimize these challenges, we propose a new efficient edge detector that benefits from the hyperbolic tangent function. The performance of this detector was tested on the responses of a synthetic gravity model and Bishop's complex magnetic model. Gaussian noise content was also added to both synthetic cases to test the response of the proposed operator. In a real-data experiment, residual gravity anomalies of the Aegean Graben System from western Türkiye, obtained via finite element method, were used to delineate abrupt lateral variations in mass densities in the region. The traces of basin-bounding normal faults and the boundaries of major grabens were improved in the resultant image map. A modified nonlocal means filter was also applied in both synthetic and real-data cases to attenuate undesired high-frequency effects. The proposed edge detection operator produced satisfactory outputs in each case. Comparative studies performed with some traditional and recently introduced algorithms further showed the superiority of the operator; thus, this detector is useful in revealing geological features that cannot be easily seen in potential-field anomaly maps and it is a strong alternative to currently used algorithms. [ABSTRACT FROM AUTHOR]

Background and Objectives: Loneliness is a major public health concern; however, limited research has examined the mechanisms contributing to racial/ethnic inequities in loneliness. Race/ethnicity has been hypothesized to be a distal factor influencing loneliness, and racial/ethnic inequities in loneliness may be attributable to socioeconomic factors (e.g., income and education). Our study seeks to confirm these hypotheses by examining mechanisms that contribute to racial/ethnic inequities in loneliness. In other words, if racial/ethnic differences in loneliness among older adults are mediated by income and education. Research Design and Methods: Data came from the Health and Retirement Study Leave-Behind Questionnaire, 2014–2016. Loneliness was measured by the UCLA 3-item loneliness scale. Race/ethnicity categories were White, Black, and Hispanic/Latino. The mediator variables were household income and education. Multivariable linear regression models were used to determine differences in loneliness by race/ethnicity. The Karlson–Holm–Breen (KHB) mediation method was used to determine if income and education mediated racial/ethnic differences in loneliness. Results: In models examining income and education together, a complete mediation was found between White and Black older adults, in that income and education completely mediated differences in loneliness between these groups. A partial mediation was found between White and Hispanic, and Black and Hispanic older adults. When examining income and education separately, we found that income solely accounted for racial/ethnic differences in loneliness compared to education. Discussion and Implications: Our study is the first to explicitly determine if socioeconomic factors mediate race/ethnicity differences in loneliness among a national sample of older adults. These findings illustrate that income may have greater proximate effects for loneliness among older adults in comparison to education. Additionally, these findings can inform evidence-based interventions to reduce loneliness among older adults. Interventions that enhance quality of life and provide opportunities for socialization for racialized low-income older adults may help decrease racial/ethnic inequities in loneliness. [ABSTRACT FROM AUTHOR]