Performance analysis of different predictive models for crop classification across an aridic to ustic area of Indian states

The purpose of this study is to present comparative performance analysis of different machine learning algorithms for large area crop classification. Ten Indian districts with significant rabi crops viz. wheat, mustard, gram, red lentils (masoor) have been selected for the study. Most popular classical ensemble models – bagging/ARCing, random forest (RF), gradient boosting and Importance Sampled Learning Ensemble (ISLE) with traditional single model (decision tree) have been selected for comparative analysis. To incorporate dependency of large area crop in different variables viz. parent material and soil, phenology, texture, topography, soil moisture, vegetation, climate etc., 35 digital layers are prepared using different satellite data (ALOS DEM, Landsat-8, MODIS NDVI, RISAT-1, Sentinental-1A) and climatic data (precipitation, temperature). In rabi season, field survey about crop type is carried out to prepare training data. Performance is evaluated on the basis of marginal rates, F-measure and Jaccard’s coefficient of community, Classification Success Index and Agreement Coefficients. Score is calculated to rank the algorithm. RF is best performer followed by gradient boosting for crop classification. Other ensemble methods ARCing, bagging and ISLE are in decreasing order of performance. Traditional non-ensemble method decision tree scored higher than ISLE.     

A Hybrid CNN + Random Forest Approach to Delineate Debris Covered Glaciers Using Deep Features

The main aim of this study is to propose a novel hybrid deep learning framework approach for accurate mapping of debris covered glaciers. The framework comprises of integration of several CNNs architecture, in which different combinations of Landsat 8 multispectral bands (including thermal band), topographic and texture parameters are passed as input for feature extraction. The output of an ensemble of these CNNs is hybrid with random forest model for classification. The major pillars of the framework include: (1) technique for implementing topographic and atmospheric corrections (preprocessing), (2) the proposed hybrid of ensemble of CNNs and random forest classifier, and (3) procedures to determine whether a pixel predicted as snow is a cloud edge/shadow (post-processing). The proposed approach was implemented on the multispectral Landsat 8 OLI (operational land imager)/TIRS (thermal infrared sensor) data and Shuttle Radar Topography Mission Digital Elevation Model for the part of the region situated in Alaknanda basin, Uttarakhand, Himalaya. The proposed framework was observed to outperform (accuracy 96.79%) the current state-of-art machine learning algorithms such as artificial neural network, support vector machine, and random forest. Accuracy assessment was performed by means of several statistics measures (precision, accuracy, recall, and specificity).     

Moisture rainout fraction over the Indian Ocean during austral summer based on $$^{18}\hbox {O}/{}^{16}\hbox {O}$$ ratios of surface seawater,rainwater at latitude range of 10°N–60°S

Oxygen isotope ratios (\(^{18}\hbox {O}/^{16}\hbox {O}\)) of surface seawater and rainwater samples from the Indian Ocean region (10°N–60°S) during austral summer collected onboard ORV Sagar Nidhi during 2011–2013 have been measured along with salinity, sea surface temperature and relative humidity. The rainwater is isotopically lighter (by \(4.6\pm 2.7\permille )\) compared to the equilibrium condensation of the vapour arising from the seawater at the ambient condition. The isotopic composition of the vapour at high altitude responsible for the rain formation at the sampling location is estimated from a global atmospheric water isotope model (IsoGSM2). The apparent deficit of \(\sim \)5\(\permille \) can be explained by invoking a high degree of rainout (on average, about 70% of the overhead atmospheric moisture) during transport of the source vapour to the sampling location undergoing a Rayleigh fractionation. The required rainout fraction is higher (\(\sim \)80%) in the latitude belt 40°–60°S compared to the equatorial belt (\(\sim \)60%). The pattern of variation in the rainout fraction with latitude is consistent with the well-known evaporation/precipitation processes in the Indian Ocean.     

Geochemistry of the Palaeo–Mesoproterozoic Tadpatri shales,Cuddapah basin,India: implications on provenance,paleoweathering and paleoredox conditions

The Palaeo–Mesoproterozoic Tadapatri formation of the Cuddapah basin is comprised of clastic sedimentary rocks with minor carbonates and mafic–ultramafic sill bodies. Geochemistry of the shale is used to study the provenance, paleoweathering and paleoredox conditions of this Tadpatri formation in order to better understand the development of the Cuddapah basin during Palaeo–Mesoproterozoic time. The higher CIA (average 74.39), PIA (average 85.94) and CIW (average 87.59) values of the Tadpatri shales suggest intensely weathered sources. Higher Al₂O₃/TiO₂ (average 30.78) and LREE/HREE ratio (average 8.80) with negative europium anomaly indicate derivation of the clastic sediments from a felsic source rock. The geochemical parameters like U, U/Th, Cu/Zn, Ni/Co, V/Cr ratios reveal that the Tadpatri shales are mainly deposited in an oxic condition.     

Assessment of Impact on Seafloor Features in INDEX Area

Observations on seafloor features from subbottom profiling and seabed photography data show that the undisturbed and natural conditions before the benthic disturbance undergo a substantial change after operation of the disturber. The effects can be seen in the form of deep trenches and grooves, formed by the action of the sleds and the pumping of the sediment along its path, as well as sediment piles on either side of the track. Biological traces such as fecal coils, casts, tubes, burrows, and trails get obliterated in the areas of disturbance and resedimentation. The smooth, uniform nature of the seafloor observed in predisturbance phase shows microtopographic changes in and around the disturbed area as a result of sediment excavation and resettlement. Vertical mixing as well as lateral transport of sediment alter the geological, biological, physical, and chemical conditions on the seafloor, which need to be monitored over time to assess the process and the time taken for restoration of environmental conditions.     

Mineralogical and geochemical variation in hydrothermal sulfides from Vienna Woods field, Manus Basin, Papua New Guinea: constraints on their evolution

Polymetallic sulfides from two hydrothermal chimneys and talus deposit from the Vienna Woods field of Manus Basin were studied for mineralogy, elemental composition, and S-isotope ratio to understand their evolution. The factors including the nature of source fluid, mineral paragenesis, and related geochemical processes have been discussed. Mineralogy and elemental concentration of Cu and Fe-rich large chimney at the central part of this hydrothermal field was completely different from the smaller Zn-rich peripheral chimney and Fe-rich talus deposit, suggesting the variable degree of alterations generate physico-chemically different source fluids responsible for these hydrothermal structures. Similarly, S-isotope ratios also indicate chemically diverse fluids and different modes of precipitation were involved in their evolution. Distinct mineral zonings and associated elemental and isotopic compositions within individual deposit confirm paragenetic shifts were involved during their growth process.     

Effect of nonthermality on the perturbation dynamics of self-gravitating complex fluids

The perturbation dynamics of an unbounded nonthermal self-gravitating inhomogeneous viscoelastic system composed of two-component constitutive fluids is theoretically investigated. The role of fluid turbulence, which is a highly nonlinear hydrodynamic vorticity-driven phenomenology, is included via the Larson logatropic equation of state describing nonlinear fluid pressure effects. The thermodynamics of the variable-temperature bulk fluid is included with the help of a proper heat diffusion equation. The system is coupled by the electro-gravitational Poisson equations in a closed form. A generalized linear dispersion relation (cubic in degree) is procedurally obtained using a standard technique of linear normal mode analysis. The dispersion relation stems from the rudimentary condition of non-vanishing perturbed gravitational potential in a linear order. The propagatory and dispersive features of the composite fluid perturbations are numerically explored with a special attention to the nonthermality effects. Their growth characteristics are analyzed alongside promising indication to applicability in the astro-cosmo-plasmic context.     

Assessing Li and other leachable geochemical proxies for paleo-salinity in lake sediments from the Mono Basin, CA (USA)

Regional climate-driven hydrological changes are accompanied by salinity changes in closed basin lakes. We have investigated acid leachable Li, along with other leachable ions including Mg, Ca and Sr, as geochemical proxies of salinity in lake sediments in the Mono Basin, California. All the elements in the acid leachable suite show a strong correlation with paleo-lake level estimates based on physical and stratigraphic evidence. The CaCO₃ content of lake sediments, which has been shown to be a reliable proxy for lake level changes in the Mono basin and the adjoining Owens Lake basin, corresponds well with our acid-leachable proxy data.     

Diatoms from surface sediments of Enderby Basin of Indian Sector of Southern Ocean

Diatoms from surface sediment samples collected from Enderby Basin of Indian Sector of Southern Ocean were analyzed to determine the relative abundance and distribution of seven key indicator diatom species viz. Sea ice related species Fragilariopsis rhombica, F. separanda, F. curta, F. ritscheri, Thalassiosira tumida and Actinocyclus actinochilus and Open Ocean species F. kerguelensis on the basis of modern physico-chemical parameters. The relative abundances of different species observed viz. F. rhombica — 6.25%; F. separanda — 12.5%; F. curta — 10.53–13.33%; F. ritscheri — 4.55–12.5%; F. kerguelensis — 6.25–63.64%; T. tumida — 3.13% and A. actinochilus — 9.38–13.33%. The increasing abundance of F. kerguelensis consecutively suggests the effect of Antarctic bottom water in the study area which is further substantiated by the presence and increasing abundance of F. ritscheri. The gradual decrease in abundance or absence of sea ice related species from the sampled stations indicates the summer and winter sea ice extent concentration in the study area. The nutrient concentration correlates with the distribution and abundances of diatom species.