Machine learning (ML) models are increasingly used to study complex environmental phenomena with high variability in time and space. In this study, the potential of exploiting three categories of ML ...regression models, including classical regression, shallow learning and deep learning for predicting soil greenhouse gas (GHG) emissions from an agricultural field was explored. Carbon dioxide (CO2) and nitrous oxide (N2O) fluxes, as well as various environmental, agronomic and soil data were measured at the site over a five-year period in Quebec, Canada. The rigorous analysis, which included statistical comparison and cross-validation for the prediction of CO2 and N2O fluxes, confirmed that the LSTM model performed the best among the considered ML models with the highest R coefficient and the lowest root mean squared error (RMSE) values (R = 0.87 and RMSE = 30.3 mg·m−2·hr−1 for CO2 flux prediction and R = 0.86 and RMSE = 0.19 mg·m−2·hr−1 for N2O flux prediction). The predictive performances of LSTM were more accurate than those simulated in a previous study conducted by a biophysical-based Root Zone Water Quality Model (RZWQM2). The classical regression models (namely RF, SVM and LASSO) satisfactorily simulated cyclical and seasonal variations of CO2 fluxes (R = 0.75, 0.71 and 0.68, respectively); however, they failed to reasonably predict the peak values of N2O fluxes (R < 0.25). Shallow ML was found to be less effective in predicting GHG fluxes than other considered ML models (R < 0.7 for CO2 flux and R < 0.3 for estimating N2O fluxes) and was the most sensitive to hyperparameter tuning. Based on this comprehensive comparison study, it was elicited that the LSTM model can be employed successfully in simulating GHG emissions from agricultural soils, providing a new perspective on the application of machine learning modeling for predicting GHG emissions to the environment.
Display omitted
•Machine learning was used for predicting agricultural soil greenhouse gas emissions.•The performance of nine alternative machine learning models was compared.•Deep learning LSTM revealed the best performance in predicting both N2O and CO2 fluxes.•Classical RF found to be a fast and effective data-driven model for predicting CO2 fluxes.
Metallic (1T) molybdenum disulphide (MoS2) is a promising electrode material in the electrochemical energy storage application due to its excellent electrical conductivity and ionic intercalation ...property. Here, we report the metallic MoS2 modified with porous graphitic carbon nitride (g-C3N4) as a novel nanocomposite for efficient supercapattery electrode. The 1T-MoS2 is grown on the surface of porous g-C3N4, and thus it facilitates easier ion intercalation between these two materials. The prepared electrode of g-C3N4/MoS2 composite attained a maximum specific capacity of 515 C g−1 at 1 A g−1 current density and obtained 66% capacitive retention even at a high current density of 20 A g−1 in three electrode cell. A symmetric cell is fabricated using g-C3N4/MoS2 composite as both positive and negative electrodes to study the practical applicability of the electrode material. The cell exhibits an excellent performance with a maximum specific capacity of 198 C g−1 at a current density of 1 A g−1 and specific energy of 147 Wh kg−1 at a power density of 2691 W kg−1. This symmetric cell still delivers an energy density of 46 Wh kg−1 at a power density as high as 26.74 kW kg−1. Moreover, the cell exhibits extended cyclic stability with capacitive retention of 89% after 8000 cycles. This outstanding electrochemical performance suggests the future scope of g-C3N4/MoS2 nanocomposite as an efficient electrode material for energy storage devices.
Most investigators regard CuFeS2 as having the formal oxidation states of Cu+Fe3+(S2−)2. However, the spectroscopic characterisation of chalcopyrite is clearly influenced by the considerable degree ...of covalency between S and both Fe and Cu. The poor cleavage of CuFeS2 results in conchoidal surfaces. Reconstruction of the fractured surfaces to form, from what was previously bulk S2−, a mixture of surface S2−, S22 and Sn2− (or metal deficient sulfide) takes place. Oxidation of chalcopyrite in air (i.e. 0.2atm of O2 equilibrated with atmospheric water vapour) results in a Fe(III)–O–OH surface layer on top of a Cu rich sulfide layer overlying the bulk chalcopyrite with the formation of Cu(II) and Fe(III) sulfate, and Cu(I)–O on prolonged oxidation. Cu2O and Cu2S-like species have also been proposed to form on exposure of chalcopyrite to air.
S22−, Sn2− and S0 form on the chalcopyrite surface upon aqueous leaching. The latter two of these species along with a jarosite-like species are frequently proposed to result in surface leaching passivation. However, some investigators have reported the formation of S0 sufficiently porous to allow ion transportation to and from the chalcopyrite surface. Moreover, under some conditions both Sn2− and S0 were observed to increase in surface concentration for the duration of the leach with no resulting passivation.
The effect of a number of oxidants, e.g. O2, H2O2, Cu2+, Cr6+ and Fe3+, has been examined. However, this is often accompanied by poor control of leach parameters, principally pH and Eh. Nevertheless, there is general agreement in the literature that chalcopyrite leaching is significantly affected by solution redox potential with an optimum Eh range suggesting the participation of leach steps that involve both oxidation and reduction.
Three kinetic models have generally been suggested by researchers to be applicable: diffusion, chemical reaction and a mixed model containing diffusion and chemical components which occur at different stages of leaching. Passivation effects, due to surface diffusion rate control, may be affected by leach conditions such as pH or Eh. However, only initial conditions are generally described and these parameters are not controlled in most studies. However, at fixed pH, Eh and temperature, it appears most likely that leaching in sulfuric acid media in the presence of added Fe3+ is surface reaction rate controlled with some initial period, depending on leach conditions, where the leach rate is surface layer diffusion controlled.
Although bioleaching of some copper ores has been adopted by industry, bioleaching has yet to be applied to predominantly chalcopyrite ores due to the slow resulting leach rates. Mixed microbial strains usually yield higher leach rates, as compared to single strains, as different bacterial strains are able to adapt to the changing leach conditions throughout the leach process. As for chemical leaching, passivation is also observed on bioleaching with jarosite being likely to be the main contributor.
In summary, whilst much has been observed at the macro-scale regarding the chalcopyrite leach process it is clear that interpretation of these phenomena is hampered by lack of understanding at the molecular or atomic scale. Three primary questions that require elucidation, before the overall mechanism can be understood are:1.How does the surface of chalcopyrite interact with solution or air borne oxidants?2.How does the nature of these oxidants affect the surface products formed?3.What determines whether the surface formed will be passivating or not?
These can only realistically be tackled by the application of near atomic-scale analytical approaches, which may include quantum chemical modelling, PEEM/SPEM, TEM, AFM etc.
Display omitted
•The PRI550, WI, OSAVI, WI/NDVI were the most sensitive indices for water stress estimation in tomato plants.•The PRI550 is an improvement over the PRI570 in detecting water stress in greenhouse ...grown tomatoes.•The use of reflectance indices provides an automated, non-destructive estimate of plant water status.•Reflectance indices have potentials to improve irrigation water management of tomato plants.
Innovations in irrigation water management are required to optimize agricultural water use in water stressed regions of the world, and physiological response of plants to water stress is an important criterion. Remotely sensed plant stress indicators, based on the visible and near-infrared spectral regions, provide an alternative to traditional field measurements of plant stress parameters, as this provides information about the spatial and temporal variability of crops and soil. The present study is a proof of concept on the feasibility of using narrow-band hyperspectral derived indices for monitoring water stress in tomato plants (Solanum Lycopersicum L.). Spectral reflectance data were acquired from tomato plants, with five different irrigation regimes namely 100, 80, 60, 40, and 20% of plant available water, in a completely randomized design. Also, plant water stress indicators including canopy temperature (Tc) and relative leaf water content (RWC), as well as volumetric soil moisture content (SMC) were concurrently measured with spectral data acquisition. Normalized Difference Vegetation Index (NDVI), Renormalized Difference Vegetation Index (RDVI), Optimized Soil Adjusted Vegetation Index (OSAVI), Photochemical Reflectance Index centered at 570 nm (PRI570), normalized PRI (PRInorm), Water Index (WI), and Normalized Water Index (NWI) were computed from the spectral data. The relationships between canopy reflectance and water stress indicators were analyzed at different water stress levels. The result showed that the PRI centered at 550 nm wavelength (PRI550), WI, OSAVI, and WI/NDVI were the most sensitive indices to distinguish water stress levels in tomato plants. This study provides an insight into the feasibility of using spectral vegetation indices to monitor water stress in tomato crops for precision irrigation water management.
A
bstract
We provide a simple and complete construction of infinite families of consistent, modular-covariant pairs of characters satisfying the basic requirements to describe twocharacter RCFT. ...These correspond to solutions of generic second-order modular linear differential equations. To find these solutions, we first construct “quasi-characters” from the Kaneko-Zagier equation and subsequent works by Kaneko and collaborators, together with coset dual generalisations that we provide in this paper. We relate our construction to the Hecke images recently discussed by Harvey and Wu.
Nanocrystalline tin oxide (SnO2) powders with different grain size were prepared by chemical precipitation method. The reaction was carried out by varying the period of hydrolysis and the as-prepared ...samples were annealed at different temperatures. The samples were characterized using X-ray powder diffractometer and transmission electron microscopy. The microstrain and crystallite size were calculated for all the samples by using Williamson-Hall (W-H) models namely, isotropic strain model (ISM), anisotropic strain model (ASM) and uniform deformation energy density model (UDEDM). The morphology and particle size were determined using TEM micrographs. The directional dependant young’s modulus was modified as an equation relating elastic compliances (sij) and Miller indices of the lattice plane (hkl) for tetragonal crystal system and also the equation for elastic compliance in terms of stiffness constants was derived. The changes in crystallite size and microstrain due to lattice defects were observed while varying the hydrolysis time and the annealing temperature. The dependence of crystallite size on lattice strain was studied. The results were correlated with the available studies on electrical properties using impedance spectroscopy.
Molybdenum disulphide (MoS
2
) is widely explored in the area of electrochemical energy storage devices as an alternative to graphene. Here we introduce a 1T-MoS
2
electrode with defect induced ...active edges for high performance supercapacitors. The metallic phase (1T) facilitates enhanced electrical conductivity and defect rich morphology that results in a high specific capacitance value. The structural characterisation confirms the formation of 1T phase and defect concentration in the material. Transmission electron microscopy imaging further confirms the presence of defects in high defect density MoS
2
(HDD-MoS
2
) and low defect density MoS
2
(LDD-MoS
2
). The electrochemical studies show that the prepared high defect density MoS
2
electrode exhibits a maximum specific capacitance of 379 F g
−1
and 68.9 F g
−1
at 1 A g
−1
current density in three and two electrode systems, respectively. The symmetric HDD-MoS
2
supercapacitor exhibits an outstanding cell performance with an energy density of 21.3 W h kg
−1
, at a power density of 750 W kg
−1
and a capacitance retention of 92% after 3000 cycles. These findings suggest that 1T MoS
2
with high defect density can be a potential candidate for high performance supercapacitor electrode application.
Influence of thiourea on the formation of active-edges and the metallic phase of MoS
2
and investigation of its energy storage properties.
A detailed study on visible light photocatalytic degradation of methylene blue (MB) has been investigated in aqueous heterogeneous media containing hexagonal phase molybdenum oxide (h-MoO3) ...nanocrystals (NCs) which was identified as a new material for visible light driven photocatalysis. A simple and template-free solution based chemical precipitation method was employed to synthesize h-MoO3 NCs by reacting ammonium heptamolybdate tetrahydrate (AHM) with nitric acid. The formation and growth mechanism of h-MoO3 microstructures was explained. In addition, by annealing the h-MoO3 sample, the phase stability of hexagonal was retained up to 410 °C and showed an irreversible phase transition from hexagonal (h-MoO3) to highly stable orthorhombic phase (α-MoO3). Finally, the photocatalytic activities of h-MoO3 and α-MoO3 samples were evaluated using the degradation of MB, representing an organic pollutant of dye wastewater. The effects of various experimental parameters such as catalyst loading, initial dye concentration, light intensity, and operating temperature were analyzed for the degradation of MB. The results demonstrated that the efficiency of visible light assisted MB degradation using h-MoO3 NCs can be effectively enhanced by catalyst loading, light intensity, and operating temperature. However, the efficiency declined with the increase in initial dye concentration. Optimum conditions for higher photocatalytic performance were recognized as a catalyst loading of 100 mg L(-1), a dye concentration of 12 mg L(-1), a light intensity of 350 mW cm(-2), and an operating temperature of 45 °C.
Pyrite is the earth’s most abundant sulfide mineral. Its frequent undesirable association with minerals of economic value such as sphalerite, chalcopyrite and galena, and precious metals such as gold ...necessitates costly separation processes such as leaching and flotation. Additionally pyrite oxidation is a major contributor to the environmental problem of acid rock drainage. The surface oxidation reactions of pyrite are therefore important both economically and environmentally.
Significant variations in electrical properties resulting from lattice substitution of minor and trace elements into the lattice structure exist between pyrite from different geographical locations. Furthermore the presence of low coordination surface sites as a result of conchoidal fracture causes a reduction in the band gap at the surface compared to the bulk thus adding further electrochemical variability. Given the now general acceptance after decades of research that electrochemistry dominates the oxidation process, the geographical location, elemental composition and semi-conductor type (n or p) of pyrite are important considerations.
Aqueous pyrite oxidation results in the production of sulfate and ferrous iron. However other products such as elemental sulfur, polysulfides, hydrogen sulfide, ferric hydroxide, iron oxide and iron(III) oxyhydroxide may also form. Intermediate species such as thiosulfate, sulfite and polythionates are also proposed to occur. Oxidation and leach rates are generally influenced by solution Eh, pH, oxidant type and concentration, hydrodynamics, grain size and surface area in relation to solution volume, temperature and pressure. Of these, solution Eh is most critical as expected for an electrochemically controlled process, and directly correlates with surface area normalised rates. Studies using mixed mineral systems further indicate the importance of electrochemical processes during the oxidation process.
Spatially resolved surface characterisation of fresh and reacted pyrite surfaces is needed to identify site specific chemical processes. Scanning photoelectron microscopy (SPEM) and photoemission electron microscopy (PEEM) are two synchrotron based surface spectromicroscopic and microspectroscopic techniques that use XPS- and XANES-imaging to correlate chemistry with topography at a submicron scale. Recent data collected with these two techniques suggests that species are heterogeneously distributed on the surface and oxidation to be highly site specific.
PDF
