The Zhunuo porphyry Cu deposit (2.9 Mt Cu @ 0.48%) in the Gangdese belt, southern Xizang, represents a key Miocene post-collisional system. This study integrates textural, major-, and trace-element analyses of pyrite from distinct alteration zones to unravel its hydrothermal evolution and metal precipitation mechanisms. Our study identifies four distinct pyrite types (Py1-Py4) that record sequential hydrothermal stages: main-stage Py2-Py3 formed at 354 ± 48 to 372 ± 43 °C (based on Se thermometry), corresponding to A and B vein formation, respectively, and late-stage Py4 crystallized at 231 ± 30 °C, coinciding with D-vein development. LA-ICP-MS data revealed pyrite contains diverse trace elements with concentrations mostly below 1000 ppm, showing distinct distribution patterns among different pyrite types (Py1-Py4). Elemental correlations revealed coupled behaviors (e.g., Au-As, Zn-Cd positive correlations; Mo-Sc negative correlation). Tellurium variability (7–82 ppm) records dynamic fO₂ fluctuations during system cooling. A comparative analysis of pyrite from the regional deposits (Xiongcun, Tiegelongnan, Bada, and Xiquheqiao) highlighted discriminative geochemical signatures: Zhunuo pyrite was enriched in Co-Bi-Ag-Pb (galena inclusions); Tiegelongnan exhibited the highest Cu but low Au-As; Xiquheqiao had the highest Au-As coupling; and Bada showed epithermal-type As enrichment. Partial Least Squares Discriminant Analysis (PLS-DA) identified Cu, As, and Bi as key discriminators for deposit types (VIP > 0.8), with post-collisional systems (Zhunuo and Xiquheqiao) showing intermediate Cu-Bi and elevated As versus arc-related deposits. This study establishes pyrite trace-element proxies (e.g., Se/Te, Co/Ni, and As-Bi-Pb) for reconstructing hydrothermal fluid evolution and proposes mineral-chemical indicators (Cu-As-Bi) to distinguish porphyry-epithermal systems in the Qinghai-Tibet Plateau. The results underscore pyrite’s utility in decoding metallogenic processes and exploration targeting in collisional settings.

Lithium is a rare metal widely used in the renewable energy industry. The Altyn region in Xinjiang, China, contains abundant granitic pegmatite-type lithium resources; however, the deeply incised and complex terrain limits the accuracy of conventional two-dimensional remote sensing approaches for dike identification and segmentation. To address this limitation, a remote sensing segmentation method incorporating terrain information was proposed. A digital elevation model (DEM) derived from LiDAR data, together with its associated topographic factors, was integrated into the Spatial–Spectral Mamba framework to enable the joint utilization of spectral and terrain features. Rather than performing explicit three-dimensional geometric modeling, the proposed approach enhances a two-dimensional segmentation framework by introducing elevation-derived information, allowing the model to capture terrain-related spatial variations of pegmatite dikes. This design enables improved representation of both the planar distribution and terrain-influenced morphological characteristics of dikes under deeply incised conditions. The Xichanggou lithium deposit in the Altyn region is a large-scale, economically valuable pegmatite-type lithium deposit, and was therefore selected as the study area for pegmatite dike segmentation. The results demonstrated that, compared with conventional two-dimensional approaches and representative machine learning methods, the proposed method achieved higher segmentation accuracy in complex terrain. Improvements were also observed in the continuity and spatial consistency of the extracted dike patterns. Field verification indicated that the major pegmatite dikes delineated by the model were highly consistent with their actual surface exposures. Sampling analyses further confirmed the validity and reliability of the identification results. Overall, the terrain-integrated remote sensing segmentation approach exhibited good applicability and robustness under deeply incised and complex geomorphological conditions.

Mineral maps can overstate identification when detectable absorptions remain compatible with several minerals. We present a framework that tests candidate labels for a sensor and scene before assigning them pixel by pixel, retaining family evidence where subclasses remain unresolved. We applied it to ZY1-02E AHSI observations of southern Libya, using background-adjusted matched-subspace, wavelet and shape measures for SWIR absorptions and the complete VNIR feature for ferric iron. Thresholds derived from scene backgrounds were evaluated on held-out blocks using numerical mixtures and independent mineral confusers. The broad 2.30 µm response was readily recovered but accepted 36.2% of tested Al-OH mixtures and was withheld. Conditional calcite and Fe/Mg-OH branches had cross-assignment of at most 1.33%. Among accepted Al-OH pixels, 31.2% remained at family level. Release stability varied: Al-OH passed in four of 13 background-sampling designs, whereas the broad 2.30 µm group failed throughout. ASTER and mapped geology were used to contextualise the evidence and identify eight potential follow-up zones covering 34.5 km², with differing persistence under selection and registration changes. The framework links controlled spectral separability to the level of detail retained in mineral maps. Independent observations are still required to confirm mineral identity and geological origin.