Geochronology and geochemistry data for the Elbrus, Tyrnyauz, and Chegem magmatic centers, Greater Caucasus, Russia

The dataset presented here is associated with the article “Young Silicic Magmatism of the Greater Caucasus, Russia with implication for its delamination origin based on zircon petrochronology and thermomechanical modeling” [1]. We present detailed sample descriptions and source locations for the rocks from the Chegem, Tyrnyauz, and Elbrus volcanic center localities presented in that study. The dataset presents extensive isotope and trace element geochemistry of zircon crystals from these rocks, major phenocrysts, and whole rock O and H isotopic and elemental compositions. Zircon ages, trace element compositions, and Hf and O isotopic compositions were obtained by both laser ablation ICP-MS and secondary ionization mass spectrometry in situ techniques and chemical abrasion isotope dilution-thermal ionization mass spectrometry techniques. We also present whole-rock major element compositions obtained by X-ray fluorescence and trace element compositions obtained by solution inductively-coupled plasma mass spectrometry. We also report δ18O analyses of phenocrysts and groundmass in samples, δ18O-δ13C analyses of limestones and limestone xenoliths in the Chegem ignimbrite, and coupled δ18O-δD-Δ17O analyses of glass and groundmass of rock samples from the Chegem ignimbrites that show abundant evidence of post-emplacement interaction with meteoric waters. To supplement the associated study [1], this article also includes field photographs, cathodoluminescence images of zircons, plots of trace element compositions in zircon, plots of stable isotopic variations in Chegem ignimbrites vs. stratigraphy, and selected trace elemental whole-rock diagrams.

Chegem ignimbrite, and coupled δ 18 O-δD- 17 O analyses of glass and groundmass of rock samples from the Chegem ignimbrites that show abundant evidence of post-emplacement interaction with meteoric waters. To supplement the associated study [1] , this article also includes field photographs, cathodoluminescence images of zircons, plots of trace element compositions in zircon, plots of stable isotopic variations in Chegem ignimbrites vs. stratigraphy, and selected trace elemental whole-rock diagrams.

Value of the Data
• This is a large new chemical and isotopic analytical dataset for young silicic rocks of the Greater Caucasus.
• This includes a large and targeted zircon U-Pb, and O and Hf isotopic zircon dataset for Chegem, Elbrus and Tyrnyauz. • This study adds large new H and O isotopes and water concentrations for ignimbrites of Chegem and C isotopes in carbonate xenoliths, documenting secondary alteration and postemplacement interaction of tuff with xenoliths.

Data Description
The data presented here include whole-rock chemical analyses of 40 samples of young volcanic rocks from the Greater Caucasus for major and trace elements ( Supplementary Table 4 ), oxygen isotopic composition of rocks and minerals ( Supplementary Table 3 ), hydrogen and carbon isotopic composition of rocks and glasses ( Supplementary Table 3 ), zircon geochronology from 9 samples including CA-IDTIMS for 3 samples, SHRIMP for 2 samples, and LA-ICP_MS for 6 samples ( Supplementary Tables 2 and 6 ). It also presents LA-ICPMS analyses of trace element variations in dated zircons, as well as in situ Hf and O isotopes in these zircons ( Supplementary  Tables 2 and 3 ). We also present field observations in the form of measured sections and field photographs presented here and in the associated research article.
Supplementary Table 5 provides a list of samples that were used in this study and lists the methods that were used to study them. It also provides GPS coordinates and field (outcrop) description of the place of their collection.
Supplementary Table 6 provides results of CA_ID-TIMS investigation (precise U-Pb dating) of a subset of three zircon populations from the Chegem Caldera, top and the bottom ignimbrites and a granite porphyry. It also provides in situ analyses of core and rims for U-Pb age and trace elements as obtained by LA-ICP-MS for the same crystals, later dated for CA-IDTIMS.
Supplementary Table 7  Supplementary Table 2 provides results of SHRIMP U-Pb dating, trace elements, and same spot co-registered oxygen SIMS isotope values, and epsilon Hf isotope analyses (by LAICPMS) made in three mounts of zircons extracted from nine samples of Elbrus, Chegem, Zayukovo, and Tyrnyauz areas. Examples of zircons that were analyzed are given in Fig. 4 . Data from these Table are plotted in Figs. 4-9 of the associated research paper [1] .
Supplementary Table 3 provide raw and reduced data for LAICPMS analyses of Hf isotopes of zircons, also listed in the supplementary Table 2 , and studied for U-Pb age using LA-ICP-MS ( Supplementary Table 7 ) performed at the University of Geneva. Data for concurrently run standards is below.
Supplementary Table 4 presents XRF and ICPMS analyses of major and trace elements of studied samples of ignimbrites and lavas, used for dating and other types of analyses, color-coded by expected source/region of origin as is determined by U-Pb dating and O-Hf isotopic fingerprinting (see associated research paper [1] for discussion. Elbrus-area ignimbrites, for example,   thermometry (c,f) . Age uncertainty is within symbols unless shown. Notice stronger heterogeneity of cores vs. rims but an overall lack of trends. Ti-in-zircon temperatures are lower than zircon saturation temperatures (f) and rims show an overall increase in temperature with youth explained by their growth during heating upon reservoir assembly. Data is presented in the Supplementary Table 7 of this paper and further trace elemental plots in zircons are presented in Fig. 9 of the associated research paper.  are color-coded to originate from the Tyrnyauz area. Data is plotted on discrimination diagrams ( Fig. 5 ) and spidergrams ( Fig. 6 ).

Fieldwork and sampling
We sampled the most prominent ignimbrites on the northern slopes of modern Elbrus edifice, at Tuzluk Mt. (sample ELB-1), and the Lower (ELB-13, −12) and Upper (ELB-9, −11) Stone  [17] . Extra data for Elbrus plotted in (b) is sourced from reference [18] . Supplementary Table 4 .  Table 5 ]. Also at Elbrus, we sampled an additional 10 lavas of varying composition (other samples in . In Tyrnyauz, we sampled both the Eldjurta granite (ELB-20) and a rhyolite (ELB-21) crosscutting it. At Chegem, we sampled transects of intracaldera tuff (samples CHEG-1 and CHEG-9 targeted for zircon), the intracaldera porphyritic intrusion (CG-C-56), and a sample of the outlying Zayukovo Tuff (ELB-23). Our study also utilized samples from the study of [2] , stored at the Caltech sample repository, for our whole-rock oxygen, hydrogen, and carbon isotope analyses (below). See [1] for further details of our sampling strategy. Figs. 1 and 3 in the associated research paper present observations.

Whole rock chemical composition
Whole rock major and trace element analyses were performed at the Irkutsk Institute of Geochemistry by XRF and solution ICP-MS (A.B. Perepelov, lab director), and at Pomona College (J.S. Lackey, lab director). XRF methods included fusion of ∼1-5 g of sample with Li-tetraborate fuse. USGS rock standards were used for cross checking the data. Supplementary Table 4 presents these data.

Stable isotopes in major phenocrysts and groundmass/glasses
Oxygen isotope measurements on phenocrysts and groundmass were performed by laser fluorination and gas-source mass spectrometry at the University of Oregon, following the methods [5] . For laser fluorination of phenocrysts, we use a regular set up while for groundmass or glass we used an airlock sample chamber and samples were introduced one by one, prior to fluorination. We used BrF 5 as a reagent and boiling Hg diffusion pump to get rid of excess of F 2 gas, then converted purified O 2 into CO 2 and run it in a dual inlet mode on MAT253 mass spectrometer, integrated with the vacuum line. Samples yields were measured in a calibrated volume using a Baratron gage. UWG2 garnet ( δ 18 O = 5.80 ‰ , [20] ) and UOG garnet ( δ 18 O = 6.52 ‰ [5] , calibrated relative to UWG2) were used to calibrate the data on VSMOW scale. Errors of individual analyses is better than ±0.1 ‰ for regular set up and better than ±0.1-0.15 ‰ for an airlock. For triple oxygen isotope analyses of two groundmass samples and the SCO standard we used an airlock sample chamber, and generated O 2 gas was put through the 8 ft long gas chromatographic column for its purification from NFx compounds. Then it was run as O 2 gas 5 timesx8 cycles in a dual inlet mode, against a well calibrated oxygen gas standard. In the same lab, we also made additional analyses of C and O isotopes in carbonates using a Thermo Fisher Gasbench system, using acidification of samples with orthophosphoric acid and He flush prior to continuous flow analysis of generated CO 2 gas on MAT253. We used TCEA (high-temperature-conversion, elemental analyzer) coupled with MAT253 isotope ratio mass spectrometer for H isotopes and total H 2 O simultaneously from a multitude of groundmass and glass samples for δ D (see [6] ) in a continuous flow mode. Mica standards USGS57 (biotite) and USGS58 (muscovite) are used for calibration and are analyzed 3-5 times throughout each analytical session, commonly also with an internal laboratory biotite standard, BUD. All isotope compositions are expressed the attached tables in delta notation relative to Vienna Standard Mean Ocean Water (VSMOW for O and H isotopes), and PDB for C isotopes. See [7] for more details about the thermal decomposition and pyrolysis method. Supplementary Table 7 presents these data.

Zircon geochemistry and U-Pb dating
Zircon grains were imaged via cathodoluminescence and scanning electron microscope ( Fig. 4 ) prior to isotopic and chemical analysis. Details of analytical methods are provided in [1] . Three epoxy mounts were prepared. Mount #1, which had zircons from the Chegem center, was analyzed for δ 18  Mount #2 contained more zircons from the same Chegem samples, first annealed at 900 °C for 48 h prior to mountain and polishing. Zircon were measured for trace elements and U/Pb ages at their cores and rims using the ASI Resonetics 155 laser ablation system coupled to a Perkin Elmer NexION 20 0 0 quadrupole ICP-MS at ETH Zürich. A subset of the grains from Mount #2 that had overlapping core and rim ages were plucked from the mount and chemically abraded [8] for 13 h at 180 °C. Zircons were then spiked with Earthtime 202 Pb-205 Pb-233 U-235 U tracer solution [9] and processed for ID-TIMS analyses using the protocols described in [10] , ( Supplementary Table 6 ).
Mount #3 was made with other Caucasus zircons from the Elbrus, Tyrnyauz, and Zayukovo localities [1] . U/Pb ages and trace element compositions were obtained by sector field LA-ICP-MS at ETH Zürich ( Supplementary Table 1 ). Trace element analyses of zircons at adjacent spots was performed employing an ASI Resonetics 155 laser ablation system coupled to a Thermo Element XR sector field ICP-MS at ETH Zürich. Primary reference materials were the same as for the Q-ICP-MS analyses described above. The majority of the ICP-MS dated grains, including Elbrus and Tyrnyauz samples, were also analyzed for Hf isotopes employing a Photon laser ablation system coupled to a Thermo Neptune Plus MC-ICP-MS at the University of Geneva ( Supplementary Table 3 ). Reproducibility of standard zircons over the course of the analyses was precise at ±0.75 epsilon units based on standard reproducibility. Data were reduced off-line using an excel spreadsheet and consisted of blank subtractions, removing the isobaric interference of 176 Lu and 176 Yb on mass 176 [11] and correcting the resulting 176 Hf/ 177 Hf ratio for mass bias using an exponential law [12] . βHf and βYb mass bias coefficients were calculated from the measured 179 Hf/ 177 Hf and 173 Yb/ 171 Yb with the reference values of [13] ( 179 Hf/ 177 Hf = 0.7325) and [14] ( 173 Yb/ 171 Yb = 1.1234) respectively. Isobaric interferences of 176 Yb and 176 Lu with 176 Hf were corrected using 176 Yb/ 173 Yb = 0.786954 and 176 Lu/ 175 Lu = 0.02645 respectively [14] ). Time resolved 176 Hf/ 177 Hf plots were used to detect any heterogeneities like inclusions or different domains (e.g., inherited core) with different 176 Hf/ 177 Hf ratios. Only non-perturbed spectra were retained. Initial 176 Hf/ 177 Hf ratios and initial εHf were calculated using the 206 Pb/ 238 U date of the respective crystal, the CHUR parameters of [15]

Declaration of Competing Interest
Authors have no competing interests.

Supplementary Materials
Supplementary material associated with this article can be found in the online version at doi: 10.1016/j.dib.2021.106896 .