Testing the reliability of interpretable neural networks in geoscience using the Madden–Julian oscillation

We test the reliability of two neural network interpretation techniques, backward optimization and layerwise relevance propagation, within geoscientific applications by applying them to a commonly studied geophysical phenomenon, the Madden–Julian oscillation. The Madden–Julian oscillation is a multi-scale pattern within the tropical atmosphere that has been extensively studied over the past decades, which makes it an ideal test case to ensure the interpretability methods can recover the current state of knowledge regarding its spatial structure. The neural networks can, indeed, reproduce the current state of knowledge and can also provide new insights into the seasonality of the Madden–Julian oscillation and its relationships with atmospheric state variables. The neural network identifies the phase of the Madden–Julian oscillation twice as accurately as a linear regression approach, which means that nonlinearities used by the neural network are important to the structure of the Madden–Julian oscillation. Interpretations of the neural network show that it accurately captures the spatial structures of the Madden–Julian oscillation, suggest that the nonlinearities of the Madden–Julian oscillation are manifested through the uniqueness of each event, and offer physically meaningful insights into its relationship with atmospheric state variables. We also use the interpretations to identify the seasonality of the Madden–Julian oscillation and find that the conventionally defined extended seasons should be shifted later by 1 month. More generally, this study suggests that neural networks can be reliably interpreted for geoscientific applications and may thereby serve as a dependable method for testing geoscientific hypotheses.

Structural Study of Cretaceous Formations for Sufaiya Area Northwest Iraq

The seismic surveys by the Iraqi petroleum Company in 1956 for the Sufaiya area showed theexistence of a geophysical phenomenon extending northwest-southeast, followed several seismicsurveys in that area which confirmed the presence of a geological structure in that area. Then,other seismic surveys were carried out to clarify the structural side of the field includes thesurvey of the Bulgarian Techno export Company 1977.The exploration well Sufaiya -1 was drilled in 1973 and the drilling reached the depth 3504meters from Rotary Table Kelly Bushing. The oil appeared in the Gir Bir, Mushorah andShiranish Formations, produced gas and condensate in the Karra Chine formation. In the light ofthis promising information and information obtained from the seismic surveys and drilling wellsin the structure. The drilling wells continued in this structure until to reach the number of wellsin the Sfaiya area 47 wells, including 5 wells penetrating the Gir Bir reservoir completely, whichis Sfaiya - 1, 2, 4, 12 and 29.

Velocity anisotropy and trend in Niger Delta, Nigeria

In geophysical data interpretation, matching the vertical velocity direction from seismic data with borehole-derived velocities is a challenging task because seismic-derived velocities are faster than borehole recorded velocities. This geophysical phenomenon is caused by velocity anisotropy. In this study, we used an empirical approach to estimate the degree of velocity anisotropy in the study area. The results showed that the delta anisotropy in sandstone beds varies from − 2.5% to 7.2% while most of them concentrate between 3.2% and 6.1%. The epsilon ranges between -6.4% and 9.3% while many of them concentrate between 3.2% and 7.2%. The gamma varies from − 6.3% to 7.3% while most of them concentrate between 1.2% and 5%. At shale beds, delta anisotropy varies from − 11.2% to 11.1% but most of them concentrate between 4.3% and 10.5%. The epsilon varies from − 7.2% to 14.5% while most of them concentrate between 4.5% and 10.5%. The gamma varies from 6.4% to 8.2% while majority of them concentrate between 2% and 5.3%. The results indicate that the study area is weakly to moderately anisotropic with shale beds having higher anisotropy values than sandstone beds. This probably results from preferential alignment of clay mineral orientations which also affect in situ velocity propagation. Three distinct velocity gradients (low, moderate and very high) were identified in the study area. These velocities vary erratically but showed northeast–southwest increase in velocities. Thus, the need to derive correction factors for individual wells for improved exploration success.

Testing the Reliability of Interpretable Neural Networks in Geoscience Using the Madden-Julian Oscillation

We test the reliability of two neural network interpretation techniques, backward optimization and layerwise relevance propagation, within geoscientific applications by applying them to a commonly studied geophysical phenomenon, the Madden-Julian Oscillation. The Madden-Julian Oscillation is a multi-scale pattern within the tropical atmosphere that has been extensively studied over the past decades, which makes it an ideal test case to ensure the interpretability methods can recover the current state of knowledge regarding its spatial structure. The neural networks can, indeed, reproduce the current state of knowledge and can also provide new insights into the seasonality of the Madden-Julian Oscillation and its relationships with atmospheric state variables. The neural network identifies the phase of the Madden-Julian Oscillation twice as accurately as linear regression, which means that nonlinearities used by the neural network are important to the structure of the Madden-Julian Oscillation. Interpretations of the neural network show that it accurately captures the spatial structures of the Madden-Julian Oscillation, suggest that the nonlinearities of the Madden-Julian Oscillation are manifested through the uniqueness of each event, and offer physically meaningful insights into its relationship with atmospheric state variables. We also use the interpretations to identify the seasonality of the MJO, and find that the conventionally defined extended seasons should be shifted later by one month. More generally, this study suggests that neural networks can be reliably interpreted for geoscientific applications and may thereby serve as a dependable method for testing geoscientific hypotheses.

A comparative study of auroral morphology distribution between the Northern and Southern Hemisphere based on automatic classification

Aurora is a very important geophysical phenomenon in the high latitudes of Arctic and Antarctic regions, and it is important to make a comparative study of the auroral morphology between the two hemispheres. Based on the morphological characteristics of the four labeled dayside discrete auroral types (auroral arc, drapery corona, radial corona and hot-spot aurora) on the 8001 dayside auroral images at the Chinese Arctic Yellow River Station in 2003, and by extracting the local binary pattern (LBP) features and using a k-nearest classifier, this paper performs an automatic classification of the 65 361 auroral images of the Chinese Arctic Yellow River Station during 2004–2009 and the 39 335 auroral images of the South Pole Station between 2003 and 2005. Finally, it obtains the occurrence distribution of the dayside auroral morphology in the Northern and Southern Hemisphere. The statistical results indicate that the four dayside discrete auroral types present a similar occurrence distribution between the two stations. To the best of our knowledge, we are the first to report statistical comparative results of dayside auroral morphology distribution between the Northern and Southern Hemisphere.

A comparative study of auroral morphology distribution between northern and southern hemispheres based on automatic classification

Aurora is a very important geophysical phenomenon in the high latitude of Arctic and Antarctic regions, and it is significant to make a comparative study of the auroral morphology between the two hemispheres. Based on the morphological characteristics of the four labeled dayside auroral types (include auroral arc, drapery corona, radial corona and hot-spot aurora) on the 8001 dayside auroral images at Chinese Yellow River Station in 2003, and by extracting the local binary pattern (LBP) features and using k-nearest classifier, this paper makes an automatic classification to the 65361 auroral images of the Chinese Yellow River Station during 2004–2009 and the 39335 auroral images of the South Pole Station between 2003–2005, and finally obtains the occurrence distribution of the dayside auroral morphology in northern and southern hemispheres. The statistical results indicate that the four auroral types present similar occurrence distribution between the two stations. To the best of our knowledge, we are the first to report the statistical comparative results of dayside auroral morphology distribution between northern and southern hemispheres.

The Upstream Spreading of Bottom-Trapped Plumes

It is well known that numerical simulations of freshwater discharges produce plumes that spread in the direction opposite to that of the propagation of coastally trapped waves (the upstream direction). The lack of a theory explaining these motions in unforced environments deemed the numerical results suspect. Thus, it became a common practice in numerical studies to add a downstream mean flow to arrest the development of the upstream perturbation. This approach is generally unjustified, and it remains a matter of interest to determine if the upstream displacement produced by models is a geophysical phenomenon or a consequence of erroneous assumptions in the model setup. In this article, the results of highly idealized numerical experiments are used to investigate these matters. It is shown that this phenomenon is associated with the geostrophic adjustment of the discharge and that upstream motion is endemic to the baroclinic structure of bottom-trapped plumes. It is also shown that downstream displacements are generated by the cross-shelf barotropic pressure gradient generated by the propagation of coastally trapped waves. Sensitivity experiments indicate that the speed of upstream propagation and the density structure of the plume are affected by bottom friction, the slope of the bottom, and the magnitude of the density anomaly. Bottom friction in particular slows down the progression of the plume and changes its density structure, producing a more homogeneous downstream region and a more stratified upstream region.

The composition of Europa's near-surface atmosphere

The presence of an undersurface ocean renders Europa as one of the few planetary bodies in our Solar System that has been conjectured to have possibly harbored life. Some of the organic and inorganic species present in the ocean underneath are expected to transport upwards through the relatively thin ice crust and manifest themselves as impurities of the water ice surface. For this reason, together with its unique dynamic atmosphere and geological features, Europa has attracted strong scientific interests in past decades.Europa is imbedded inside the Jovian magnetosphere, and, therefore, is constantly subjected to the immerse surrounding radiations, similar to the other three Galilean satellites. The magnetosphere-atmosphere-surface interactions form a complex system that provides a multitude of interesting geophysical phenomenon that is unique in the Solar System. The atmosphere of Europa is thought to have created by, mostly, charged particles sputtering of surface materials. Consequently, the study of Europa's atmosphere can be used as a tool to infer the surface composition. In this paper, we will discuss our recent model studies of Europa's near-surface atmosphere. In particular, the abundances and distributions of the dominant O2 and H2O species, and of other organic and inorganic minor species will be addressed.

Ownership of Evolved Gas in Split Title Situations

This paper analyzes the ownership of evolved gas where ownership of the petroleum rights has been severed from the mineral title. How 'split titles' arise and the geophysical phenomenon of evolved gas are discussed briefly. The main focus of the paper is the statutory and common law considerations concerning the ownership of evolved gas. The effects of errors in payment of royalties are also examined.