Near-surface velocity estimation using shear-waves and deep-learning with a U-net trained on synthetic data

Estimation of good velocity models under complex near-surface conditions remains a topic of ongoing research.We propose to predict near-surface velocity profiles from surface-waves transformed to phase velocity-frequency panels in a data-driven manner using deep neural networks.This is a different approach from many recent works that attempt to estimate velocity from directly reflected body waves or guided waves.A secondary objective is to analyze the influence on the prediction accuracy of various commonly employed deep learning practices,such as transfer learning and data augmentations.Through numerical experiments on synthetic data as well as a real geophysical example,we demonstrate that transfer learning as well as data augmentations are helpful when using deep learning for velocity estimation.A third and final objective is to study lack of generalization of deep learning models for out-of-distribution(OOD)data in the context of our problem,and present a novel approach to tackle it.We propose a domain adaptation network for training deep learning models that uses a priori knowledge on the range of velocity values in order to constrain mapping of the output.The final comparison on field data,which was not part of the training data,show the deep neural network predictions compare favorably with a conventional velocity model estimation obtained with a dispersion curve inversion workflow.

Response of Photonic Hydrogels of Homogeneous Particles to Uranyl Ions in Aqueous Solutions

We study here the response of photonic hydrogels(PHs),made of photonic crystals of homogeneous silica particles in polyacrylamide hydrogels(SPHs),to the uranyl ions 22 UO_(2)^(2+) in aqueous solutions.It is found that the reflection spectra of the SPH show a peak due to the Bragg diffraction,which exhibits a blue shift in the presence of 22 UO_(2)^(2+).Upon exposure to the SPH,22 UO_(2)^(2+)gets adsorbed on the SPH and forms complex coordinate bonds with multiple ligands on the SPH,which causes shrinking of hydrogel and leads to the blue shift in the diffraction peak.The amount of the blue shift in the diffraction peak increases monotonically up to 22 UO_(2)^(2+)concentrations as high as 2300μM.The equilibration time for the shift in the Bragg peak upon exposure to 22 UO_(2)^(2+)is found to be~30 min.These results are in contrast to the earlier reports on photonic hydrogels of inhomogeneous microgel particles hydrogel(MPH),which shows the threshold 22 UO_(2)^(2+)concentration of~600μM,below which the diffraction peak exhibits a blue shift and a change to a red shift above it.The equilibration time for MPH is~300 min.The observed monotonic blue shift and the faster time response of the SPH to 22 UO_(2)^(2+)as compared to the MPH are explained in terms of homogeneous nature of silica particles in the SPH,against the porous and polymeric nature of microgels in the MPH.We also study the extraction of 22 UO_(2)^(2+)from aqueous solutions using the SPH.The extraction capacity estimated by the arsenazo-III analysis is found to be 112 mM/kg.