An SVM-based machine learning method for the separation of alteration zones in Sungun porphyry copper deposit

Sungun porphyry copper deposit is in East Azarbaijan province, NW of Iran. There exist four hypogene alteration types in Sungun: potassic, propylitic, potassic–phyllic, and phyllic. Copper mineralization is essentially associated more with the potassic and less with the phyllic alterations and their separation is, therefore, quite important. This research has tried to separate these two alteration zones in Sungun porphyry copper deposit using the Support Vector Machine (SVM) method based on the fluid inclusion data, and seven variables including homogenization temperatures, salinity, pressure, depth, density and the Cu grade have been measured and calculated for each separate sample. To apply this method, use is made of the radial basis function (RBF) as the kernel function. The best values for λ and C (the most important SVM parameters) that perform well in the training and test data are 0.0001 and 1, respectively. If these values for λ and C are applied, the phyllic and potassic alteration zones in the training and test data will be separated with an accuracy of about 95% and 100%, respectively. This method can help geochemists in separating the alteration zones because classifying and separating samples microscopically is not only very hard, but also quite time and money consuming.     

Phosphorus export dynamics and hydrobiogeochemical controls across gradients of scale,topography and human impact

Concentration‐discharge (c‐Q) plots are routinely used as an integrated signal of watershed response to infer solute sources and travel pathways. However, the interpretation of c‐Q data can be difficult unless these data are fitted using statistical models. Such models are frequently applied for geogenic solutes, but it is unclear to what extent they might aid in the investigation of nutrient export patterns, particularly for total dissolved phosphorus (TDP) which is a critical driver of downstream eutrophication problems. The goal of the present study was therefore to statistically model c‐Q relations (where c is TDP concentrations) in a set of contrasting watersheds in the Northern Great Plains—ranging in size from 0.2 to 1000+ km²—to assess the controls of landscape properties on TDP transport dynamics. Six statistical models were fitted to c‐Q data, notably (a) one linear model, (b) one model assuming that c‐Q relations are driven by the mixing of end‐member waters from different landscape locations (i.e., hydrograph separation), (c) one model relying on a biogeochemical stationarity hypothesis (i.e., power law), (d) one model hypothesizing that c‐Q relations change as a function of the solute subsurface contact time (i.e., hyperbolic model), and (e) two models assuming that solute fluxes are mostly dependent on reaction rates (i.e., chemical models). Model performance ranged from mediocre (R² < 0.2) to very good (R² > 0.9), but the hydrograph separation model seemed most universal. No watershed was found to exhibit chemostatic behaviour, but many showed signs of dilution or enrichment behaviour. A tendency toward a multi‐model fit and better model performance was observed for watersheds with moderate slope and higher effective drainage area. The relatively poor model performance obtained outside these conditions illustrates the likely importance of controls on TDP concentrations in the region that are independent of flow dynamics.     

Enhancement of hydrothermal alteration zones using the spectral feature fitting method in Rabor area, Kerman, Iran

Rabor exploration area is in southeast of Kerman province, south of Urumieh-Dokhtar volcanic belt. Since there are many evidences of porphyry copper mineralization in this belt, prospecting and exploration of the intact and undiscovered places, especially the southern part of this belt, seems necessary. In this paper, use has been made of the “spectral feature fitting” (SFF) method to process ASTER satellite data (in Rabor exploratory area as a case study) for the identification and enhancement of hydrothermal alteration zones related to probable porphyry copper mineralization. The method is based on the comparison of absorption features in the image and the reference spectra. The distribution map of the indicator clay minerals, such as kaolinite, muscovite, illite, montmorillonite, alunite, pyrophyllite, dickite, chlorite, and epidote in Rabor exploratory area has been prepared with the help of this method. Identification and enhancement of alteration zones and overlaying Pay-Negin ore index on every individual zone reveal the capability and high efficiency of the SFF method in processing ASTER satellite data and preparing the distribution map of alteration minerals. Field and laboratory studies have shown that main alterations in this area are phyllic and argillic. These studies confirm the results obtained from remote sensing in the area.