Igneous rocks form when molten magma is cooled and solidified, either within the Earth's crust (plutonic rocks), or from lava extruded onto the Earth's surface in the atmosphere or underwater (volcanic rocks). The classification of igneous rocks can be done using data from different instrumental techniques. However, these approaches tend to be expensive and time-consuming. In this research work, several models for the classification of granitoids, which are the most abundant plutonic rocks in the Earth's crust, were created with a convolutional neural network developed with TensorFlow. Specifically, several combinations of gabbro, diorite, tonalite, granodiorite, monzodiorite, and granite image samples were used in the experiments. The best result was obtained in the model that classifies images of gabbro, diorite, granodiorite, and granite with an accuracy value of 95%, an average precision value of 96%, an average recall value of 95%, and an average F1 score value of 95%.
Magmatic activity in continental arcs is known to vary in a non-steady-state manner, with the mechanisms driving magmatic activity being a matter of ongoing discussion. Of particular importance is the question of what extent episodic magmatism in continental arcs is governed by external factors (e.g., plate motions) and internal factors (e.g., feedback processes in the upper plate). To test existing models for magmatic episodicity, which are mostly based on temporally and spatially limited records, this study uses large data sets of geochronological, geochemical, and plate kinematic data to document the Paleozoic to Mesozoic development of the North and South American Cordilleras in eight transects from British Columbia to Patagonia. The temporal distribution of U/Pb bedrock and detrital zircon ages, used as a proxy for timing of magmatic accretion, shows that some minima and maxima of zircon abundance are nearly synchronous for thousands of kilometers along the arc. Some age patterns are characterized by a periodicity of 50–80 Ma, suggesting a cyclic controlling mechanism. Other magmatic lulls or flare-ups find no equivalents in adjacent sectors, indicating that either discrete events or variable lag times may also be important in governing magmatic activity in continental arcs. Magma composition in Mexico, the Peninsular Ranges, and the Sierra Nevada varies episodically and proportionally with the temporal record of arc activity. During flare-up events, there is an increase in Sm/Yb, indicating deeper melting, and a decrease in eNdb suggesting a higher degree of crustal assimilation. Geochemical scatter also increases during the initiation of flare-up events. Plate kinematic data provide a means of evaluating mantle heat input. The correlation between plate convergence rate and magmatic accretion varies for each sector, suggesting that different flare-ups or lulls likely reflect variable combinations of processes.
We studied the formation of low-initial-Sr felsic plutons by using data from the Early Cretaceous western Peninsular Ranges batholith near Escondido, California. The systematically sampled Escondido plutons have a uniformly low initial 87Sr/86Sr isotope ratio of Sri < 0.704, but a wide range of SiO2 compositions, from 46 to 78 wt%, which fall in three distinct groups: 20% gabbros, 35% tonalites, and 45% granodiorites. These low-Sri plutons are unique in having undergone one cycle of mantle melting to give basalt composition rocks, and a second cycle of arc basalt melting to give a range...
Podoconiosis is a disease whose etiology remains murky. Currently, the disease is attributed to particles that are believed to move through the skin and into the lymphatic system causing swelling of the lower legs. Identity of these particles or their composition remains unclear, though the presence of silicon and/or aluminum is often noted and frequently cited as causal agents. We applied multivariate analyses to the bedrock compositions of a large set of cases from an online database in an effort to identify underlying patterns or combinations of relative element abundances associated with podoconiosis-endemic regions. Using a combination of principal component analysis, discriminant function analysis, and ANOVA, we analyzed ten oxides from five regions on the African continent known to be associated with podoconiosis. The Hawaiian Islands were included as a control group since they are not known to have cases of podoconiosis despite similarity in geology and agricultural practices. Our analyses suggest that a unique alkaline- and silicon-rich geochemistry underlies regions associated with podoconiosis. Our results also imply that minerals enriched in incompatible elements, such as Ca, K, Mg, and Na, may be stronger predictors of the presence of the disease than either silicon or aluminum.
