Integrated Approach to Paleontological and Geochemical Data of the Upper Campanian–Maastrichtian Harami Formation (Elazığ, Eastern Türkiye)

Abstract

The aim of this study is to determine the age and depositional environment of the Harami Formation through an integrated interpretation of XRF, ICP-MS, SEM-EDS, stable isotope, and radioactive isotope analysis results and findings of larger benthic foraminifera. Findings from the authors’ previous studies regarding the characteristics of larger benthic foraminifera were utilized; geochemical analysis results were obtained for the first time in this study. The mean ⁸⁷Sr/⁸⁶Sr values are generally consistent with the ages determined from the larger benthic foraminifera. The analysis results and Y/Ho values indicate that CaO is the dominant component, while negative Ce and positive Eu anomalies are indicative of a low-grade diagenetic effect. The absence of hydrothermal activity in the depositional environment and the proximity of the values to those of seawater are supported by the Y/Ho and Eu/Sm ratios. Similarly, to the larger benthic foraminifera content that indicates the Harami Formation was deposited in a shallow, oxic environment, the trace element ratios (V/Cr and V/V+Ni) and slightly negative Ce anomalies also support deposition in an oxic environment, while the Sr/Ba and Sr/Rb analysis results indicate a shallow marine environment. The analysis results further indicate that the depositional environment ranged from warm-temperate to arid conditions. A decrease in mean temperature is observed from the late Campanian to the early Maastrichtian, with an increase from the early Maastrichtian to the late Maastrichtian.

1. Introduction

The Harami Formation was first described by Erdoğan [1] in the vicinity of Harami village, north of Gölbaşı district, Adıyaman province. This formation crops out in very limited areas around the Elazığ province in the eastern part of the Taurus Orogenic Belt due to the depositional environmental conditions and the extensive post-late Cretaceous tectonism in the region (Figure 1A,B).

In the study area and its immediate surroundings, units with different lithologies and ages, from the Paleozoic to the Cenozoic, crop out [2,3,4,5,6] (Figure 2A–C). The outcrops of the Harami Formation around Elazığ lie on top of the Elazığ Magmatics, which are a product of an island arc. The formation consists mostly of layered, reddish sandy limestone and limestone in the lower levels and massive limestone in the upper levels; very limited areas of reddish conglomerates and sandstones deposited in a fan delta environment can be observed in very limited areas at its base (Figure 3A–G). While some researchers [4,5], taking into account the limited areas of conglomerate and sandstone, state that the Harami Formation is unconformably overlying the Elazığ Magmatics, there are also studies suggesting that the boundary between the two units is conformable [6].

Although research has been conducted on the larger benthic foraminifera content of the Harami Formation outcrops around Elazığ [5,6,7,8,9,10], no study based on geochemical characteristics has been carried out. The formation’s outcrops around Elazığ have been studied under different names, and their ages have been determined based on their benthic foraminifera contents. Data on this subject is summarized in

Table 1. The ages determined for the Harami Formation in previous studies based on the larger benthic foraminifera (modified from [9]).

Author(s)	Study Area	Formation	Larger Benthic Foraminifera Contents	Age
Özgen et al. [7]	Southwest, northeast, and west of Elazığ	Harabekayış Formation	Orbitoides medius O. apiculatus Omphalocyclus macroporus Lepidorbitoides minor Siderolites calcitrapoides Hellenocyclina beotica Smoutina cruysi Scandonea aff. samnitica Rotalia perovalis Rotalia sp. Anomalina sp. Eponides sp. Mississipina sp. Idalina sinjarica Mississipina binkhorsti Daviesina danieli Cuvillierina sireli Rotalia trocidiformis Kathina selveri Miscellanea miscella Planorbulina aff. antiqua Orbitoclypeus seunesi	late Maastrichtian-Thanetian
İnceöz [5]	Northeast of Elazığ	Harami Formation	Orbitoides medius O. apiculatus Orbitoides spp. Omphalocyclus macroporus Lepidorbitoides spp. Siderolites calcitrapoides Hellenocyclina beotica Sirtina orbitoidiformis Smoutina cruysi Scandonea samnitica Triloculina sp. Quinqueloculina sp. Rotaliidae Textulariidae	late Maastrichtian
Aksoy et al. [6]	Southwest of Elazığ	Harami Formation	Orbitoides cf. medius Orbitoides sp. Lepidorbitoides spp. Pseudosiderolites vidali Pseudosiderolites sp. Praesiderolites dordoniensis	late Campanian–early Maastrichtian
Kaya and İnceöz [8]	The northeast, southwest, and west of Elazığ	Harami Formation	Orbitoides medius O. apiculatus Omphalocyclus macroporus Sirtina orbitoidiformis Hellenocyclina beotica Lepidorbitoides minor L. cf. socialis Lepidorbitoides sp. Pseudorbitoides trechmanni Siderolites calcitrapoides Smoutina sp.	middle–late Maastrichtian
Kayğılı [9], Kayğılı et al. [10]	Northeast and southwest of Elazığ	Harami Formation	Orbitoides medius O. megaloformis O. gruenbachensis O. apiculatus O. gensacicus Lepidorbitoides campaniensis L. bisambergensis L. cf. minor Omphalocyclus macroporus Pseudomphalocyclus blumenthali Siderolites gr. calcitrapoides Pseudosiderolites vidali Hellenocyclina beotica Sirtina cf. orbitoidiformis	late Campanian–Maastrichtian

. According to benthic foraminifera such as Orbitoides medius, O. megaloformis, O. gruenbachensis, O. apiculatus, O. gensacicus, Lepidorbitoides campaniensis, L. bisambergensis, L. cf. minor, Omphalocyclus macroporus, Pseudomphalocyclus blumenthali, Siderolites gr. calcitrapoides, Pseudosiderolites vidali, Hellenocyclina beotica, and Sirtina cf. orbitoidiformis, the age of the Harami Formation is late Campanian–Maastrichtian [9,10] (Figure 4). Late Campanian–Maastrichtian larger benthic foraminifera (LBF) are of great importance from biostratigraphic and evolutionary perspectives on carbonate platforms [11,12,13,14,15,16,17,18,19]. The presence of L. campaniensis indicates a late Campanian age, whereas L. bisambergensis indicates an early Maastrichtian age, and Orbitoides apiculatus, O. gensacicus, Omphalocyclus macroporus, and Pseudomphalocyclus blumenthali indicate a late Maastrichtian age. This study differs from other studies conducted on this unit by presenting both paleontological and geochemical data together, thus offering an integrated perspective on the unit. Previous findings regarding the age and depositional environment of the unit were based on paleontological and sedimentological data. In this study, geochemical analysis results were also utilized to elucidate the formation’s age and the characteristics of the depositional environment in greater detail, and the conclusions drawn from the paleontological data were compared with those derived from the analyses. Thus, the existing gap in the literature has been addressed by obtaining more detailed information regarding the unit’s age and depositional environment. Multidisciplinary studies integrating stratigraphy, biostratigraphy, and geochemistry have been conducted in other regions [20,21,22].

Figure 1. (A) Location of the study area in Türkiye. (B) The locations of the measured sections in the study area (modified from [9]).

Figure 1. (A) Location of the study area in Türkiye. (B) The locations of the measured sections in the study area (modified from [9]).

Figure 2. (A) Geological map of the area containing the measured sections at Silsilekaya Hill, Buzluk, and Ölbe Stream (modified from [5]). (B) Geological map of the area containing the measured sections at Guremağalan, Nohutlupınarı, Tepeköy, and Eskişehir Hill (modified from [4]). (C) Geological map of the area containing the measured section at Ürünveren (modified from [6]).

Figure 2. (A) Geological map of the area containing the measured sections at Silsilekaya Hill, Buzluk, and Ölbe Stream (modified from [5]). (B) Geological map of the area containing the measured sections at Guremağalan, Nohutlupınarı, Tepeköy, and Eskişehir Hill (modified from [4]). (C) Geological map of the area containing the measured section at Ürünveren (modified from [6]).

Figure 3. (A–G) Field view of the locations of the measured sections and the boundary relationship between the Elazığ Magmatics and the Harami Formation. (A) Silsilekaya Hill section. (B) Buzluk section. (C) Ölbe Stream section. (D) Nohutlupınarı and Guremağalan sections. (E) Eskişehir Hill section. (F) Tepeköy section. (G) Ürünveren section. Some of the measured section views (A,C,D–G) are from Kayğılı [9] and Kayğılı et al. [10] [CC by 4].

Figure 3. (A–G) Field view of the locations of the measured sections and the boundary relationship between the Elazığ Magmatics and the Harami Formation. (A) Silsilekaya Hill section. (B) Buzluk section. (C) Ölbe Stream section. (D) Nohutlupınarı and Guremağalan sections. (E) Eskişehir Hill section. (F) Tepeköy section. (G) Ürünveren section. Some of the measured section views (A,C,D–G) are from Kayğılı [9] and Kayğılı et al. [10] [CC by 4].

Figure 4. Larger benthic foraminifera of the Harami Formation. Views of some larger benthic foraminifera are from Kayğılı [9] and Kayğılı et al. [10] [CC by 4].

Figure 4. Larger benthic foraminifera of the Harami Formation. Views of some larger benthic foraminifera are from Kayğılı [9] and Kayğılı et al. [10] [CC by 4].

2. Materials and Methods

Fieldwork in this study involved measuring stratigraphic sections, such as Eskişehir Hill (E), Tepeköy (T), Nohutlupınarı (N), Guremağalan (G), Silsilekaya Hill (S), Ürünveren (Ü), Buzluk (Bu) and Ölbe Stream (Öl) in the upper Campanian–Maastrichtian Harami Formation outcrops located northeast and southwest of Elazığ city, and collecting rock, Orbitoides, and rudist samples along the same sections (

Table 2. Information on the analyzed samples.

Analysis Type	Number of Samples	Sample Number
Stable Isotope Analysis	50	E1-LBF, E4-LBF, E4-Rudist E5-LBF, E6-LBF, E7-LBF, E4-rock sample, T1-LBF, T2-LBF, T3-LBF, T3-Rudist, T4-LBF, T5-LBF, T6-LBF, T7-LBF, N1-LBF, N2-LBF, N3-LBF, N4-LBF, N5-LBF, N6-LBF, N7-LBF, G1-LBF, G2-LBF, G3-LBF, G4-LBF, G5-LBF, G6-LBF G7a-LBF, G8-rock sample, S1-rock sample, S10a-rock sample, S13-rock sample, Ü1-LBF, Ü3-LBF, Ü6-LBF, Ü7-LBF, Ü10-LBF, Ü13-LBF, Ü16-LBF, Ü19-rock sample, Bu1-rock sample, Bu3-rock sample, Bu4-rock sample, Bu5-rock sample, Öl-1-LBF, Öl-2-rock sample, Öl-3-rock sample, Öl-4-rock sample, Öl-5-rock sample
Radiogenic Isotope Analysis	29	E4-rock sample, E4-Rudist (4 samples), E6-rock sample, T1-Rudist (2 samples), T1-rock sample, T3-LBF (2 samples), T3-Rudist (4 samples), T3-rock sample, T7-Rudist (4 samples), T7-rock sample, S1-rock sample, N1-rock sample, N2-LBF, S10-rock sample, S13-rock sample, Öl-1-Rudist (3 samples)
ICP-MS Analysis	10	E4-Rudist, E6-Rudist, T1-Rudist, T3-LBF, T3-Rudist, T7-Rudist (2 samples), N2-LBF, Öl-1-Rudist (2 samples)
XRF Analysis	10	E4-Rudist, E6-Rudist, T1-Rudist, T3-LBF, T3-Rudist, T7-Rudist (2 samples), N2-LBF, Öl-1-Rudist (2 samples)
SEM-EDS Analysis	6	E1, T1, N1, G3, S10, Ü1

). Thin sections were prepared from loose and rock samples were also gathered from the same sites. The genus/species determination of benthic foraminifera in thin sections prepared from rock and loose samples was carried out in previous studies [9,10].

Radiogenic isotope analysis was conducted to determine the age of the Harami Formation; ICP-MS and stable isotope analyses were performed to obtain information regarding the depositional environment and climate, and ICP-MS, XRF, and SEM-EDS analyses were carried out to indirectly determine the mineralogical composition of the unit’s fossil content.

Radiogenic isotope (⁸⁷Sr/⁸⁶Sr), ICP-MS, and XRF analyses require samples of a specific weight (5–10 g). Due to the lack of a sufficient quantity of samples for Orbitoides and rudist, these analyses could only be performed on 10 samples. However, as the sample weight was suitable for stable isotope analyses, they were conducted on 50 samples from along the measured sections.

Orbitoides, rudist, and rock samples taken along the measured stratigraphic sections were selected and ground in the laboratory, and stable (δ¹³C and δ¹⁸O), radioactive isotope (⁸⁷Sr/⁸⁶Sr), ICP-MS, XRF, and SEM-EDS analyses were performed on these samples at the METU, WSS, and MTA laboratories. Information regarding the analysis methods applied in these laboratories is provided below.

The determination of δ¹³C and δ¹⁸O isotope ratios in carbonate samples was carried out at the METU laboratory using the Gas Bench-Continuous Flow Isotope Ratio Mass Spectrometry technique (Delta Plus XP Isotope Ratio Mass Spectrometer–Thermo Finnigan, Thermo Fisher Scientific, Waltham, MA, USA). Samples weighing 0.2–0.6 mg were placed in the sample vials. The samples were allowed to react with 99% ortho-phosphoric acid (Merck, Darmstadt, Germany) (approximately 0.1 mL) at a temperature of approximately 70 °C on the auto-sampler tray for about two hours. The CO₂ gas released as a result of this reaction was purified by the Gas Bench interface and then sent to the mass spectrometer for isotope separation. Ion ratios for CO₂ gas were converted into raw isotopic ratios by the ISODAT software (version 3.0). In the analyses, the “NBS19 Limestone (NIST)” standard (δ¹³C: 1.95‰ and δ¹⁸O: −2.20‰) was analyzed alongside the samples for every experimental set and used to convert the raw isotope ratios determined by the instrument into true isotopic values.

In stable isotope analyses conducted at the WSS laboratory, carbon and oxygen isotopes of carbonate rocks were measured using a Gas Bench II coupled with a DELTA™ Q mass spectrometer (Thermo Scientific™, Waltham, MA, USA) via the phosphoric acid digestion method. Approximately 0.2 mg of 200-mesh carbonate powder was placed in a 12 mL headspace vial. After purging air with high-purity helium (He, 99.999%), 0.1 mL of anhydrous phosphoric acid (H₃PO₄, 100% purity) was injected. Following an 8-h reaction at 70 °C, the generated CO₂ was extracted by an auto-sampler (Triplus, Thermo Fisher Scientific, Waltham, MA, USA), separated using a gas chromatography column (GC column at 50 °C), and then analyzed with the DELTA™ Q mass spectrometer to obtain carbon and oxygen isotope ratios relative to the laboratory reference gas. Linear calibration was performed using international and national reference standards including NBS-18, IAEA-603, GBW04405, GBW04406, GBW04416 and GBW04417. The standard deviations of δ¹³C-VPDB and δ¹⁸O-VPDB were better than 0.2‰.

The method specified in [23] was followed for the strontium analyses conducted at METU. For this analysis, approximately 80 milligrams of each sample were weighed and transferred into PFA vials. The samples were first placed in deionized water and, following the dissolution process, dried to near-dryness. Subsequently, they were kept in 4 mL of 14 N HNO₃ for one day and completely dissolved on a hot plate. Following evaporation, the samples were completely dissolved on a hot plate by keeping them in 4 mL of 6 N HCl for one day. The samples were then evaporated to near-dryness on the hot plate, taken up in 1 mL of 2.5 N HCl, and prepared for chromatography. Strontium was separated in Teflon columns using Bio-Rad AG50 W-X8 (Bio-Rad Laboratories, Hercules, CA, USA) (100–200 mesh) resin and 2.5 N HCl eluent (2 mL volume). The strontium sample was loaded onto a single Re filament using a Ta activator and 0.005 N H₃PO₄, and measurements were performed in static mode. ⁸⁷Sr/⁸⁶Sr data were normalized to ⁸⁷Sr/⁸⁶Sr = 0.1194. During measurement, the Sr NBS 987 standard yielded a value of 0.710260 ± 10 (n = 9). No bias correction was applied to the strontium isotope ratio measurement results. Isotope ratio measurements were performed using multi-collection on a Triton Thermal Ionization Mass Spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Analytical uncertainties are at the 2-sigma level.

Sr isotope analyses were performed on a Neptune Plus MC-ICP-MS (Thermo Fisher Scientific, Dreieich, Germany) at the WSS. The Neptune Plus, a double-focusing MC-ICP-MS, was equipped with seven fixed electron-multiplier ICs, and nine Faraday cups fitted with 1011 Ω resistors. The faraday collector configuration of the mass system was composed of an array from L4 to H3 to monitor ⁸³Kr+, ¹⁶⁷Er++, ⁸⁴Sr+, ⁸⁵Rb+, ⁸⁶Sr+, ¹⁷³Yb++, ⁸⁷Sr+, ⁸⁸Sr+. The large dry interface pump (120 m³ h⁻¹ pumping speed), the newly designed H skimmer cone and the standard sample cone were used to increase the instrumental sensitivity. A Sr single element solution from Alfa (Alfa Aesar, Karlsruhe, Germany) was used to optimize the instrument’s operating parameters. An aliquot of the international standard solution of 200 μg L−1 NIST SRM 987 was regularly used to evaluate the reproducibility and accuracy of the instrument. Typically, the signal intensities of ⁸⁸Sr in NIST 987 were >~7.0 V. The Sr isotopic data were acquired in the static mode at low resolution. The routine data acquisition consisted of ten blocks of 10 cycles. The total time of one measurement lasted about 7 min.

Trace element and REE analysis of whole rocks was conducted on an Agilent 7700e ICP-MS (Agilent Technologies, Santa Clara, CA, USA) at the WSS. The detailed sample-digesting procedure was as follows: (1) sample powder (200 mesh) was placed in an oven at 105 °C for drying for 12 h; (2) 50 mg of sample powder was accurately weighed and placed in a Teflon bomb; (3) 1 mL HNO₃ and 1 mL HF were slowly added into the Teflon bomb; (4) the Teflon bomb was placed in a stainless steel pressure jacket and heated to 190 °C in an oven for >24 h; (5) after cooling, the Teflon bomb was opened, placed on a hot plate at 140 °C and evaporated to incipient dryness, and then 1 mL of HNO₃ was added and evaporated to dryness again; (6) 1 mL of HNO₃, 1 mL of MQ water and 1 mL of an internal standard solution of 1 ppm was added, and the Teflon bomb was resealed and placed in the oven at 190 °C for >12 h; and (7) the final solution was transferred to a polyethylene bottle and diluted to 100 g by the addition of 2% HNO₃.

In the XRF analysis conducted at the WSS laboratory, sample pretreatment of whole rock for major element analysis was performed via the melting method. The flux was a mixture of lithium tetraborate, lithium metaborate, and lithium fluoride (45:10:5). Ammonium nitrate and lithium bromide were used as oxidants and release agents, respectively. The melting temperature was 1050 °C, and the melting time was 15 min. A Zsx Primus II wavelength dispersive X-ray fluorescence spectrometer (XRF) produced by RIGAKU (Tokyo, Japan), was used to analyze major elements in the whole rock. The X-ray tube had a 4.0 Kw end window Rh target, and the test conditions were a voltage of 50 kV and a current of 60 mA; all major element analysis lines were kα, and the standard curve used the national standard material rock standard sample GBW07101-14, soil standard sample GBW07401-08 and stream sediment standard sample GBW07302-12. The data were corrected using the theoretical α coefficient method. The relative standard deviation (RSD) was less than 2%.

In the SEM-EDS analysis conducted at the MTA laboratory, the samples were first coated with gold. Under high vacuum conditions, secondary electron, backscattered electron, or mixed signal images were taken from the sample surface and photographed. Non-standard semi-quantitative point analysis or line analysis was performed at the desired location. Photographs and EDS analysis results were recorded.

3. Results

3.1. Stable Isotope Analysis Findings

Stable isotope analysis (δ¹³C and δ¹⁸O) was performed on a total of 50 samples taken along eight measured stratigraphic sections at METU and WSS (

Table 3. Stable isotope analysis results of LBF and rock samples. The first letters of the sample numbers in the table represent the names of the measured stratigraphic sections to which they belong. The Eskişehir Hill section is indicated by E, the Tepeköy section by T, the Nohutlupınarı section by N, the Guremağalan section by G, the Silsilekaya Hill section by S, the Ürünveren section by Ü, the Buzluk section by Bu, and the Ölbe Stream section by Öl.

Number	Sample Name	‰ δ13C (VPDB)	‰ δ18O (VPDB)	T (°C) δ18Osw: −1.2	T (°C) δ18Osw: −1	Laboratory
1	E1-LBF	1.73	−2.36	20.34	21.29	METU
2	E4-LBF	2.54	−1.96	18.43	19.38	METU
3	E5-LBF	2.38	−1.18	14.72	15.67	METU
4	E6-LBF	2.14	−1.44	15.96	16.91	METU
5	E7-LBF	1.48	−2.33	20.19	21.14	METU
6	E4-rock sample	2.20	−2.77	22.29	23.24	METU
7	T1-LBF	0.78	−2.19	19.53	20.48	METU
8	T2-LBF	2.40	−1.32	15.39	16.34	METU
9	T3-LBF	2.08	−1.01	13.91	14.86	METU
10	T4-LBF	2.03	−1.22	14.91	15.86	METU
11	T5-LBF	2.06	−1.05	14.11	15.06	METU
12	T6-LBF	2.14	−1.53	16.39	17.34	METU
13	T7-LBF	2.00	−2.40	20.53	21.48	METU
14	N1-LBF	2.16	−0.85	13.15	14.10	METU
15	N2-LBF	1.79	−1.40	15.77	16.72	METU
16	N3-LBF	1.66	−2.52	21.10	22.05	METU
17	N4-LBF	2.36	−2.76	22.24	23.19	METU
18	N5-LBF	1.41	−5.26	34.14	35.09	METU
19	N6-LBF	1.59	−4.02	28.24	29.19	METU
20	N7-LBF	1.42	−3.03	23.53	24.48	METU
21	G1-LBF	2.34	−0.92	13.48	14.43	METU
22	G2-LBF	2.03	−0.79	12.86	13.81	METU
23	G3-LBF	2.21	−0.93	13.53	14.48	METU
24	G4-LBF	2.35	−1.02	13.96	14.91	METU
25	G5-LBF	2.03	−2.07	18.96	19.91	METU
26	G6-LBF	2.55	−1.59	16.67	17.62	METU
27	G7a-LBF	2.19	−0.87	13.24	14.19	METU
28	G8-rock sample	1.28	−2.80	22.43	23.38	METU
29	S1-rock sample	−1.19	−2.81	22.48	23.43	METU
30	S10a-rock sample	1.47	−2.36	20.34	21.29	METU
31	S13-rock sample	1.93	−3.01	23.43	24.38	METU
32	Ü1-LBF	1.07	−6.20	38.62	39.57	METU
33	Ü3-LBF	1.39	−3.72	26.81	27.76	METU
34	Ü6-LBF	1.53	−2.95	23.15	24.10	METU
35	Ü7-LBF	1.83	−3.18	24.24	25.19	METU
36	Ü10-LBF	1.21	−4.42	30.14	31.09	METU
37	Ü13-LBF	0.97	−3.88	27.57	28.52	METU
38	Ü16-LBF	1.27	−4.14	28.81	29.76	METU
39	Ü19-rock sample	1.63	−5.68	36.14	37.09	METU
40	Bu1-rock sample	1.42	−4.48	30.43	31.38	WSS
41	Bu3-rock sample	0.28	−5.21	33.9	34.85	WSS
42	Bu4-rock sample	0.67	−6.06	37.95	38.90	WSS
43	Bu5-rock sample	0.73	−5.95	37.42	38.37	WSS
44	Öl-1-LBF	1.74	−4.40	30.05	31	WSS
45	Öl-2-rock sample	1.75	−4.74	31.66	32.61	WSS
46	Öl-3-rock sample	2.08	−3.89	27.62	28.57	WSS
47	Öl-4-rock sample	1.77	−5.18	33.87	34.82	WSS
48	Öl-5-rock sample	1.96	−5.62	35.85	36.80	WSS

,

Table 4. Stable isotope analysis results of rudist samples. E represents samples taken from the Eskişehir Hill measured stratigraphic section, and T represents samples taken from the Tepeköy measured stratigraphic section.

Number	Sample Name	‰ δ13C (VPDB)	‰ δ18O (VPDB)	T (°C) δ18Osw: −1.00	Laboratory
49	E4-Rudist	2.28	−2.35	21.84	METU
50	T3-Rudist	1.88	−2.64	23.14	METU

and

Table 5. Statistical values for stable isotope results and paleotemperature calculations of different sample types (Minimum: Min., Maximum: Max., Standard Deviation: SD).

Sample Type	Age Sample Number/ Number	‰ δ13C (VPDB)	‰ δ18O (VPDB)	T (°C) δ18Osw: −1.2	T (°C) δ18Osw: −1
LBF	late Campanian Ü/7	Min: 0.97 Max: 1.83 Mean: 1.32 SD: 0.27	Min: −6.20 Max: −2.95 Mean: −4.07 SD: 0.99	Min: 23.15 Max: 38.62 Mean: 28.48 SD: 4.72	Min: 24.10 Max: 39.57 Mean: 29.43 SD: 4.72
	early Maastrichtian E, T, N, G/26	Min: 0.78 Max: 2.55 Mean: 1.99 SD: 0.41	Min: −5.26 Max: −0.79 Mean: −1.85 SD: 1.05	Min: 12.86 Max: 34.14 Mean: 17.90 SD: 4.98	Min: 13.81 Max: 35.09 Mean: 18.85 SD: 4.98
	late Maastrichtian Öl/1	1.74	−4.40	30.05	31
Rock	late Campanian Ü/1	1.63	−5.68	36.14	37.09
	early Maastrichtian E, G/2	Min: 1.28 Max: 2.20 Mean: 1.74 SD: 0.46	Min: −2.80 Max: −2.77 Mean: −2.79 SD: 0.01	Min: 22.29 Max: 22.43 Mean: 22.36 SD: 0.07	Min: 23.24 Max: 23.38 Mean: 23.31 SD: 0.07
	late Maastrichtian S, Bu, Öl/11	Min: −1.19 Max: 2.08 Mean: 1.17 SD: 0.94	Min: −6.06 Max: −2.36 Mean: −4.48 SD: 1.24	Min: 20.34 Max: 37.95 Mean: 30.45 SD: 5.89	Min: 21.29 Max: 38.90 Mean: 31.40 SD: 5.89
Rudist	early Maastrichtian E, T/2	Min: 1.88 Max: 2.28 Mean: 2.08 SD: 0.20	Min: −2.64 Max: −2.35 Mean: −2.50 SD: 0.15	-	Min: 21.84 Max: 23.14 Mean: 22.49 SD: 0.65

, Figure 5, Figure 6, Figure 7 and Figure 8).

The mean value of δ¹³C is 1.36 in the late Campanian, 1.98 in the early Maastrichtian, and 1.22 in the late Maastrichtian. The mean δ¹⁸O values are −4.27 in the late Campanian, −1.95 in the early Maastrichtian, and −4.48 in the late Maastrichtian. Based on all measured sections, a positive correlation (0.59) is observed between δ¹³C values and δ¹⁸O values (Table 3, Table 4 and Table 5, Figure 5, Figure 6, Figure 7 and Figure 8).

Paleotemperature values were calculated based on different formulas and coefficients for benthic foraminifera and rudist, according to the studies and equations given in

Table 6. Equations used in calculating paleotemperature.

Reference	Material	Calibration Range (°C)	Equation
Lynch-Stieglitz et al. [25] reorganized by Cramer et al. [28].	In situ Cibicidoides and Planulina (Benthic foraminifera)	4–26	T (°C) = 16.1 − 4.76 [δ18Occ − (δ18Osw − 0.27)] δ18Osw: −1.2
Anderson and Arthur [24]	Rudist	-	T (°C) = 16.0 − 4.14 (δ18Occ − δ18Osw) + + 0.13 (δ18Occ − δ18Osw)2 δ18Osw: −1 [27]

[24,25,26,27,28,29].

Different formulas and different coefficients for δ¹⁸Osw are used in the calculation of paleotemperature values [24,25,26,27,28,29]. For the late Cretaceous period, although values between −1 and −1.2 are used for δ¹⁸Osw, a value of −1 is widely assumed (Table 6). In the calculation of paleotemperature values in this study, δ¹⁸Osw values of −1 and −1.2 were used for benthic foraminifera, while a value of −1 was used for rudist. The results obtained for the −1 and −1.2 values for benthic foraminifera are presented below, separated by a slash (/).

Based on paleotemperature calculations using only δ¹⁸O values from larger benthic foraminifera, the mean temperatures were 29.43/28.48 °C in the late Campanian, 18.85/17.90 °C in the early Maastrichtian, and 31/30.05 °C in the late Maastrichtian. When considering only the rock samples, the mean temperatures were 37.09/36.14 °C in the late Campanian, 23.31/22.36 °C in the early Maastrichtian, and 31.40/30.45 °C in the late Maastrichtian. Based on rudist, the mean temperature in the early Maastrichtian was 22.49 °C (Table 5).

3.2. Radiogenic Isotope Analysis Findings

Radiogenic isotope analyses (⁸⁷Sr/⁸⁶Sr) were performed on a total of 37 samples at the METU and WSS. Analyses of the same samples (E4-rudist and T7-rudist) in different laboratories (Türkiye and China) served as an external check of the accuracy of the data. Radiogenic isotope analyses were performed on 29 rudist, Orbitoides, and rock samples (

Table 7. Evaluation of radiogenic isotope (⁸⁷Sr/⁸⁶Sr) analysis results in terms of age (see Table 2 for explanations of the sample numbers) (Minimum: Min., maximum: Max., standard deviation: SD).

Number	Sample Name	87Sr/86Sr	Age According to the Analysis Results and Larger Benthic Foraminifera	Laboratory
1	E4-rock sample	0.707694		METU
2	E4-Rudist	0.707866		METU
3	E4-Rudist	0.707593		METU
4	E4-Rudist	0.707694		WSS
5	E4-Rudist	0.707692		WSS
6	E6-rock sample	0.707923		METU
	Eskişehir Hill	Min.: 0.707593 Max.: 0.707923 SD: 0 Mean: 0.707744 Number: 6	early Maastrichtian
7	T1-Rudist	0.707827		METU
8	T1-Rudist	0.707685		WSS
9	T1-rock sample	0.707771		METU
10	T3-LBF	0.707854		METU
11	T3-LBF	0.707715		WSS
12	T3-Rudist	0.707716		WSS
13	T3-Rudist	0.707596		METU
14	T3-Rudist	0.707589		METU
15	T3-Rudist	0.707664		WSS
16	T3-rock sample	0.707672		METU
17	T7-Rudist	0.707870		METU
18	T7-Rudist	0.707577		WSS
19	T7-Rudist	0.707699		WSS
20	T7-Rudist	0.707697		WSS
21	T7-rock sample	0.707746		METU
	Tepeköy	Min.: 0.707577 Max.: 0.707870 SD: 0 Mean: 0.707712 Number: 15	early Maastrichtian
22	N1-rock sample	0.707808		METU
23	N2-LBF	0.707703		WSS
Nohutlupınarı		Min.: 0.707703 Max.: 0.707808 SD: 0 Mean: 0.707756 Number: 2	early Maastrichtian
24	S1-rock sample	0.707904		METU
25	S10-rock sample	0.707804		METU
26	S13-rock sample	0.707656		METU
Silsilekaya Hill		Min.: 0.707656 Max.: 0.707904 SD: 0 Mean: 0.707788 Number: 3	late Maastrichtian
27	Öl-1-Rudist	0.707670		WSS
28	Öl-1-Rudist	0.707701		WSS
29	Öl-1-Rudist	0.707687		WSS
Ölbe Stream		Min.: 0.707670 Max.: 0.707701 SD: 0 Mean: 0.707686 Number: 3	According to analysis result: early Maastrichtian (diagenesis effect) According to the larger benthic foraminifera: late Maastrichtian

, Figure 9). The analysis results for eight samples were excluded from the evaluation because a very strong diagenesis effect was observed. Radiogenic isotope (⁸⁷Sr/⁸⁶Sr) values varied between 0.707577 and 0.707923 (Table 7, Figure 9). Ages corresponding to the mean of the ⁸⁷Sr/⁸⁶Sr values in the measured sections were evaluated by comparing with ages given according to the larger benthic foraminifera content. Considering the means of the other sections, the mean of the ⁸⁷Sr/⁸⁶Sr values is consistent with the ages given by the larger benthic foraminifera (Table 7, Figure 9).

3.3. ICP-MS Analysis Findings

ICP-MS analyses of 10 LBF and rudist samples were performed at WSS; the value of trace elements and rare earth elements determined by the analyses are shown in Tables S1 and S2 and Figure 10A,B. Table 6 shows the trace element contents of Li, Be, V, Cr, Co, Ni, Cu, Zn, Ga, Rb, Sr, Zr, Nb, Sn, Cs, Ba, Hf, Ta, Tl, Pb, Th, and U, and trace element ratios such as V/Cr, V/V+Ni, Sr/Cu, Rb/Sr, Sr/Rb, Sr/Ba, and Ga/Rb. The rare earth element contents for La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Y, and Sc, and the rare earth element rates for Y/Ho, Eu/Sm, La/Yb, Er/Nd, Eu/Eu*, and Ce/Ce* are reported in Table S2. These rare earth element values were normalized according to PASS [31] (Table S2). The ratios of trace and rare earth elements such as V/Cr, V/V+Ni, Sr/Cu, Rb/Sr, Sr/Rb, Sr/Ba, Ga/Rb, Y/Ho, and Eu/Sm, which are important in interpreting paleoenvironmental conditions, are given in

Table 8. Trace element values used in the interpretation of paleoredox conditions [33,34] and the values in this study (Minimum: Min, maximum: Max., standard deviation: SD).

Element Ratio	Oxic	Dysoxic	Anoxic	Euxinic	Authors	This Study
V/Cr	<2	2–4.5	>4.5		Jones and Manning [34]	Min.: 0.71 Max.: 2.56 SD: 0.68 Mean: 1.44 Number: 10 Oxic
V/V+Ni	<0.46	0.46–0.60	0.54–0.82	>0.84	Hatch and Leventhal [33]	Min.: 0.15 Max.: 0.59 SD: 0.16 Mean: 0.40 Number: 10 Oxic

and

Table 9. Trace element values used in the interpretation of paleoclimate [35,36,37,38,39,40,41,42] and the values in this study.

Element Ratio	Authors	This Study
Sr/Cu	>5 Arid and <5 Humid [36,37,39,42]	Min.: 7 Max.: 698 SD: 225.67 Mean: 205 Number: 10 Arid
Rb/Sr	<0.5 Arid and >0.5 Humid [41,42]	Min.: 0.0002 Max.: 0.0145 SD: 0.01 Mean: 0.0058 Number: 10 Arid
Sr/Ba	>1 Arid and <1 Humid [40,42] >1 Sea or Saline Water [35,38]	Min.: 10 Max.: 530 SD: 154 Mean: 110 Number: 10 Arid and Sea or Saline Water

, S1, and S2 and Figure 10A,B. The trace element concentrations were normalized to marine carbonate (MC) [32].

3.4. XRF Analysis Findings

XRF analysis of 10 LBF and rudist samples was performed at WSS. According to the results, CaO is the dominant component, ranging from 49.49% to 54.87% (

Table 10. Major oxide values (%) obtained from XRF analysis of the samples (Minimum: Min, maximum: Max., standard deviation: SD).

Sample Number	CaO	SiO2	TiO2	Al2O3	Fe2O3	MnO	MgO	Na2O	K2O	P2O5	LOI	SUM
E4-Rudist	54.08	0.82	0.014	0.21	0.17	0.009	0.93	0.07	0.062	0.015	43.40	99.77
E6-Rudist	54.59	0.28	0.008	0.03	0.07	0.002	1.04	0.06	0.010	0.009	43.62	99.73
T1-Rudist	49.49	7.29	0.058	1.40	0.52	0.031	0.55	0.11	0.409	0.024	39.82	99.70
T3-LBF	52.24	3.52	0.042	0.87	0.75	0.015	0.57	0.08	0.220	0.020	41.44	99.76
T3-Rudist	52.26	2.73	0.053	0.76	0.31	0.010	0.66	0.09	0.204	0.016	42.36	99.45
T7-Rudist	51.96	0.47	0.012	0.14	0.88	0.031	2.56	0.05	0.038	0.017	43.56	99.71
T7-Rudist	54.87	0.36	0.008	0.05	0.19	0.023	0.64	0.03	0.010	0.012	43.59	99.79
N2-LBF	52.48	3.10	0.032	0.76	0.70	0.019	0.61	0.05	0.169	0.020	41.77	99.71
Öl-1-Rudist	52.22	1.60	0.021	0.36	1.96	0.101	1.12	0.06	0.03	0.034	42.52	100
Öl-1-Rudist	52.33	1.60	0.021	0.36	1.97	0.103	1.11	0.06	0.03	0.035	42.54	100
Min.	49.49	0.28	0.008	0.03	0.07	0.002	0.55	0.03	0.010	0.009	39.82	99.45
Max.	54.87	7.29	0.058	1.40	1.97	0.103	2.56	0.11	0.409	0.035	43.62	100
SD.	1	2	0	0	1	0.03	0.57	0.02	0.12	0.01	1.15	0.15
Mean	52.65	2.18	0.03	0.49	0.75	0.03	0.98	0.07	0.12	0.02	42.46	99.76
Number	10	10	10	10	10	10	10	10	10	10	10	10

, Figure 10C). The major oxide contents of the samples ranged from 0.28 to 7.29% for SiO₂, 0.01 to 0.06% for TiO₂, 0.03 to 1.40% for Al₂O₃, 0.07 to 1.97% for Fe₂O₃, 0.00 to 0.10% for MnO, 0.55 to 2.56% for MgO, 0.03 to 0.11% for Na₂O, 0.01 to 0.41% for K₂O, and 0.01 to 0.04% for P₂O₅ (Table 10, Figure 10C). The mean major oxide contents of the samples is as follows: CaO (52.65%) > SiO₂ (2.18%) > MgO (0.98%) > Fe₂O₃ (0.75%) > Al₂O₃ (0.49%) > K₂O (0.12%) > Na₂O (0.07%) > MnO = TiO₂ (0.03%) > P₂O₅ (0.02%).

3.5. SEM-EDS Analysis Findings

SEM-EDS analysis was performed on six samples (Table 2) at the MTA laboratory.

Table 11. EDS analysis results of samples taken along the measured sections.

	Percentage of Elements by Weight (Weight %)
Elements	Nohutlupınarı Section (1) Range Mean	Ürünveren Section (2) Range Mean	Silsilekaya Hill Section (3) Range Mean	Eskişehir Hill Section (4) Range Mean	Guremağalan Section (5) Range Mean	Tepeköy Section (6) Range Mean
OK	26.72–54.67 45.87	23.64–54.40 46.31	43.53–56.93 49.82	25.79–54.47 46.86	37.49–59.86 50.79	32.24–54.69 46.08
CaK	4.20–45.83 34.36	0.32–44.75 24.94	2.76–47.17 36.06	0.85–44.08 28.50	2.29–42.61 29.03	0.75–45.42 21.17
CK	5.64–42.48 12.97	6.89–13.02 10.35	7.06–13.76 10.88	8.04–13.63 11.27	6.90–13.51 11.60	9.28–12.63 10.97
FeK	0.48–54.81 11.82	0.40–67.39 11.77	-	0.57–65.46 15.93	0.55–47.68 8.84	0.67–60.44 20.78
SiK	0.26–10.96 2.94	0.33–40.35 8.27	0.08–51.56 8.05	0.28–52.48 7.55	0.24–56.27 10.26	0.20–54.62 14.67
AlK	0.26–6.47 2.06	0.35–13.81 3.64	0.04–1.44 0.70	0.28–10.32 2.07	0.30–6.27 1.93	0.15–12.70 2.99
MgK	0.48–1.43 0.83	0.52–7.60 1.70	0.15–1.11 0.76	0.47–7.25 1.37	0.35–2.77 1.20	0.42–13.37 3.05
NK	8.33	-	-	-	-	-
NaK	1.18 -	0.44–12.74 6.13	-	-	-	3.47 -
SK	3.15	-	-	-	-	-
KK	0.75 -	0.20–0.59 0.29	-	0.19–5.21 2.04	0.28–0.31 0.30	0.21–4.69 2.14
TiK	-	1.20–2.82 2.03	-	0.30–1.98 0.9	0.34–0.93 0.64	-
MnK	-	0.31	-	-	-	-
VK	-	-	-	0.43	-	-
CrK	-	-	-	-	1.47	-
ClK	-	-	-	-	-	0.44

shows the percentage weight variations and means of the elements along the measured sections. EDS analyses revealed that the samples predominantly contained O (45.87%–50.79%), Ca (21.17%–36.06%), Fe (8.84%–20.78%), Si (2.94%–14.67%), and C (10.35%–12.97%). Al (0.70%–3.64%), Mg (0.76%–3.05%), S (3.15%), N (8.33%), Na (1.18%–6.13%), S (3.15%), K (0.29%–2.14%), Ti (0.64%–2.03%), Mn (0.31%), V (0.43%), Cr (1.47%), and Cl (0.44%) were also found (Table 11, Figure 11 and Figure 12). These values are similar to the major oxide values.

4. Discussion

The latest Cretaceous was characterized by a long-term cooling trend of several degrees [43,44,45] that followed the warm mid-Cretaceous super-greenhouse climate with tropical sea-surface temperatures that were about 10 °C higher than today (e.g., [46]) and intermediate-to-deep-water temperatures above 20 °C (e.g., [43]).

The late Cretaceous period is well known for its general greenhouse conditions, characterized by relatively high atmospheric pCO₂ concentrations, a high global mean sea level, and largely absent ice sheets [43,47,48]. The Campanian–Maastrichtian interval witnessed a general global downward trend in temperatures, accompanied by anomalies in the carbon cycle [49,50]. The combination of lower pCO₂ levels and the opening of ocean passages has been proposed as possible triggers for paleoclimatic events [44,51]. This cooling trend during the late Cretaceous was not monotonous, and the late Campanian climate optimum was interrupted by a series of warming events such as cooling in the early Maastrichtian, warming in the mid-Maastrichtian, and warming in the late Maastrichtian [52].

An integrated interpretation of the micropaleontological and geochemical characteristics of the Harami Formation outcrops northeast and southwest of Elazığ city provides a detailed reconstruction of paleoenvironmental evolution during the late Campanian–Maastrichtian period.

The dissolved Sr content in the world’s oceans has changed over time from ⁸⁷Sr/⁸⁶Sr [30,53,54]. Strontium isotope stratigraphy is a good chronostratigraphic method, and ⁸⁷Sr/⁸⁶Sr isotopic compositions are widely used in radiometric dating [53,55,56,57,58]. Chronostratigraphic interpretations have been made using this isotope ratio [59,60,61,62,63]. While the Sr isotope ratio of the ocean is generally considered homogeneous during all geological periods, variations can occur over short periods [55]. The ⁸⁷Sr/⁸⁶Sr value data from well-preserved and well-dated limestones has been used to create a marine reference curve of geological history and is continuously updated [30,53,54,56,57]. A geological time interval characterized by a steep Sr isotope curve can be accurately dated chronostratigraphically and globally correlated using Sr isotope stratigraphy [30,53,54,56,57].

The mean analysis values of ⁸⁷Sr/⁸⁶Sr obtained from larger benthic foraminifera, rudist, and rock samples taken from the Maastrichtian sequence of the Harami Formation are between 0.707686 and 0.707788. Except for the Ölbe Stream section, the ages obtained from the larger benthic foraminifera content in the Eskişehir Tepe, Nohutlupınarı, Tepeköy, and Silsilekaya Tepe sections (early/late Maastrichtian) support the ages obtained using the ⁸⁷Sr/⁸⁶Sr data. In the Ölbe Stream section, while larger benthic foraminifera data indicate the late Maastrichtian, the analysis results (0.707686) show the early Maastrichtian; this difference is thought to be due to diagenesis.

Table 5 presents the mean temperature values calculated based on samples of larger benthic foraminifera, rudist, and rock in this study. Based on these values, it can be inferred that there was a decrease in mean temperature from the late Campanian to the early Maastrichtian, followed by an increase from the early Maastrichtian to the late Maastrichtian; this result is consistent with those of previous studies [43,44,47,48,49,50,51,52,64]. The stable isotope results obtained from the larger benthic foraminifera, rudist, and rock samples in this study suggest that the larger benthic foraminifera samples were relatively less affected by diagenesis compared to the rock and rudist samples. Paleotemperature calculations are controversial, and various studies employ different δ¹⁸Osw coefficients and formulas [24,25,26,27,28,29]. In this study, two paleotemperature values were calculated for benthic foraminifera using δ¹⁸Osw coefficients of −1 and −1.2, revealing a difference of 0.95 °C between them. This difference is negligible when interpreting mean temperatures.

Although diagenetic processes can alter stable isotope ratios, foraminifera and other organisms can maintain their initial isotope ratio for millions of years [65,66]. However, since recrystallization occurs at the sea bottom temperature where the benthic foraminifera are buried, the benthic foraminifera are not significantly affected by recrystallization [65,67]. Studies on Eocene benthic foraminifera have concluded that δ¹⁸O values obtained from rocks are lower than those obtained from larger benthic foraminifera [66].

The major oxide and elemental compositions of the samples are compatible. The negative correlation observed between CaO and major oxides such as SiO₂ (−0.83), TiO₂ (−0.76), Al₂O₃ (−0.82), Fe₂O₃ (−0.36), MnO (−0.27), MgO (−0.02), Na₂O (−0.70), K₂O (−0.77), and P₂O₅ (−0.55) indicates that CaO is dominant and the other components have a diluting effect, as also stated by Zhang et al. [68] and Tribovillard et al. [69] (Figure 13). Depositional implications are consistent with a shallow platform/ramp setting in which alternation between more carbonate-rich and more marly intervals may result both from changes in terrigenous supply and from variability in carbonate production/preservation linked to energy, turbidity, and nutrient availability [70,71]. The strong positive correlation observed between Al₂O₃ and SiO₂ (0.98), Na₂O (0.80), K₂O (0.97), and TiO₂ (0.95) indicates that Si, Na, K, and Ti are largely obtained from clay minerals. The strong (r = 0.98) positive correlation between SiO₂ and Al₂O₃ suggests that feldspar and quartz are derived from the same source area (Figure 14A–D).

The negative correlation between Fe₂O₃ and Eu/Eu* (−0.23) and Zr and Eu/Eu* (−0.46) (Figure 14E,F) indicates that iron is largely of terrestrial origin and that the Eu anomaly decreases due to increasing terrestrial material [72]. The high Fe concentration implies that the marine environment is close to land and rather shallow. The red color, particularly seen in the larger benthic foraminifera and rudist, is due to the Fe content. In the Silsilekaya Hill section, the larger benthic foraminifera are white and, as can be seen from the EDS analysis, do not contain Fe. In the Ürünveren section, the fact that the shells are not red compared to the shells in other sections is due to the presence of other elements aside from Fe. As Muray [73] and Boudagher-Fadel [74] have also noted, the high Fe content indicates that the environment is very close to land and quite shallow.

The means of V/Cr and V/V+Ni are 1.44 and 0.40 respectively. Redox-sensitive trace element ratios (V/Cr and V/V+Ni) indicate oxic condition [33,34,39,75,76].

Paleoclimate proxies (Sr/Cu, Rb/Sr, Sr/Ba) consistently point to arid climatic conditions, with fluctuating salinity [35,36,37,38,39,40,41,42,77]. The increases and decreases in the Sr/Cu and Sr/Ba ratios in the Harami Formation indicate that the salinity in the basin is controlled by climate changes [78]. When the paleoclimate was warm–humid, the salinity of the water was fresh/brackish, and when paleoclimate conditions were hot–arid, saline water prevailed in the basin [78].

The Sr/Ba ratio can be used to distinguish marine and terrestrial environments. Overall, ratios exceeding 1 indicate sea or saline water, and ratios lower than 1 represent freshwater conditions [35,38]. In this study, Sr/Ba ratios exceeding 1 indicate sea or saline water.

As a result of the evaluation of the samples on the Sr/Ba–Sr/Rb diagram [79], all the samples exhibited deposition in a continental margin shallow marine environment (Figure 15A). According to the SiO₂ and Al₂O₃+K₂O+Na₂O diagram (modified from [80]), except for one example, the samples were generally considered to be from arid regions (Figure 15B). According to the Ga/Rb and Sr/Cu diagram (modified from [39,75]), all but two of the samples generally fall into warm–hot regions (Figure 15C). Based on these two graphs, it can be said that the environment is warm–hot to arid.

The Y/Ho values vary between 29 and 44 and the mean value is 35. These values are close to those of carbonate rocks and seawater for the depositional environment and source rocks (Figure 15D). According to the Y/Ho and Eu/Sm diagram (modified from [81,82,83]), there is no hydrothermal effect, and the values are close to seawater composition (Figure 15E).

Ce anomalies (Ce/Ce*) are often used as indicators of depositional conditions [75,77]. Ce anomalies are associated with transgression and regression [84]. Ce anomaly values observed in marine limestones are important in determining redox conditions with terrestrial input [39,75,85,86]. Although Ce anomalies are commonly used to reconstruct past seawater redox conditions, there is currently no complete compilation of sedimentary Ce anomaly data [87]. The use of this data is controversial [84]. Generally, shallower water environments present more oxic conditions [39,75,88]. Limestones showing weak negative Ce anomalies indicate deposition in an oxic environment with low terrestrial input [81]. Therefore, the combination of trace element ratios and slightly negative Ce anomalies indicates an oxic environment in the Harami Formation during the early Maastrichtian.

In this study, the Eu anomaly (Eu/Eu*) values are 1.29–1.67 and the mean value is 1.44 (Table 7, Figure 10B). A positive Eu anomaly in the larger benthic foraminifera shows that it is associated with one or more parameters such as increases in O₂ and pH or increases in temperature [77,89]. Positive Eu anomalies are mostly due to hydrothermal fluids, a slight increase in detrital feldspar components, and diagenesis [85,90]. Positive Eu anomalies can result not only from hydrothermal fluids but also from the input of detrital feldspars and diagenesis. The fact that the Harami Formation samples have not undergone hydrothermal alteration is shown in the Y/Ho and Eu/Sm diagram in Figure 15E.

Strong correlations can exist between Ce/Ce* and Eu/Eu* in the diagenetic process, and changes in the values of these anomalies can be observed [89,91]. In this study, the low positive correlation of 0.21 between Ce/Ce* and Eu/Eu* indicates a low diagenetic effect (Figure 14G).

Figure 15. (A) Sr/Ba and Sr/Rb ratios indicating shallow marine environment for the Harami Formation (modified from [79]). (B) %SiO₂ and %Al₂O₃+K₂O+Na₂O indicating paleoclimatic conditions (modified from [80]). (C) Bivariate cross-plot of Sr/Cu and Ga/Rb ratios indicating paleoclimatic conditions for the Harami Formation (modified from [39,75]). (D) Comparison of the Y/Ho ratios in the study area (modified from [92]). (E) Y/Ho and Eu/Sm diagram showing no hydrothermal activity in the study area (modified from [82,83,93]).

Figure 15. (A) Sr/Ba and Sr/Rb ratios indicating shallow marine environment for the Harami Formation (modified from [79]). (B) %SiO₂ and %Al₂O₃+K₂O+Na₂O indicating paleoclimatic conditions (modified from [80]). (C) Bivariate cross-plot of Sr/Cu and Ga/Rb ratios indicating paleoclimatic conditions for the Harami Formation (modified from [39,75]). (D) Comparison of the Y/Ho ratios in the study area (modified from [92]). (E) Y/Ho and Eu/Sm diagram showing no hydrothermal activity in the study area (modified from [82,83,93]).

5. Conclusions

The conclusions obtained via the integrated interpretation of previously established micropaleontological characteristics of the Harami Formation and geochemical data obtained first for the same unit—derived in the present study—are described below.

The mean values of ⁸⁷Sr/⁸⁶Sr obtained from larger benthic foraminifera, rudist, and rock samples from the Maastrichtian sequence of the Harami Formation were determined to range between 0.707686 and 0.707788. Except for the Ölbe Stream section, the ages derived from the larger benthic foraminifera content in the other sections (early/late Maastrichtian) support the dates found with the ⁸⁷Sr/⁸⁶Sr data. In the Ölbe Stream section, while larger benthic foraminifera data indicate the late Maastrichtian period, analytical results (0.707686) correspond to the early Maastrichtian period; this difference is thought to be due to diagenesis.

Mean paleotemperature values calculated from the δ¹⁸O analysis results of larger benthic foraminifera, rudist, and rock samples of Harami Formation indicate a decline from the late Campanian to the early Maastrichtian, and an increase from the early Maastrichtian to the late Maastrichtian.

Comparing stable isotope data from larger benthic foraminifera, rudist, and rock samples from the late Campanian–Maastrichtian period revealed that larger benthic foraminifera samples were less affected by diagenesis than rock and rudist samples.

Geochemical findings regarding major elements indicate that the majority of Si, Na, K, and Ti originated from clay minerals, while feldspar and quartz were derived from the same source area. These findings also suggest a terrestrial origin for Fe, with high Fe concentrations implying the proximity of the terrestrial source area and a relatively shallow marine depositional environment.

Redox-sensitive trace element ratios (V/Cr and V/V+Ni) indicate an oxic environment. Paleoclimatic proxies (Sr/Cu, Rb/Sr, Sr/Ba) are indicative of persistent arid climate conditions and fluctuating salinity. An Sr/Ba ratio exceeding 1 indicates a marine or saline water environment. Similar variations in the Sr/Cu and Sr/Ba ratios in the Harami Formation samples indicate that salinity in the basin is controlled by climate changes. Some geochemical analysis results (i.e., SiO₂ and Al₂O₃+K₂O+Na₂O; Ga/Rb and Sr/Cu ratios) support the interpretation that the Harami Formation was deposited in a shallow marine environment within a region characterized by warm-hot to arid climatic conditions.

Although positive Eu anomalies can result from hydrothermal fluids, detrital feldspars introduced into the environment, or diagenesis, the Y/Ho and Eu/Sm values in Figure 15E indicate that the Harami Formation was deposited in a marine environment where there was no hydrothermal activity.

The weak negative Ce anomalies and trace element ratios of Harami Formation’s limestones indicate deposition in an oxic environment with low terrestrial input.

The positive Eu anomaly in larger benthic foraminifera can be explained by multiple parameters, such as an increase in O₂ and pH or an increase in temperature in the environment. Although positive Eu anomalies can result from hydrothermal fluids, detrital feldspars input into the environment, or diagenesis, the Y/Ho and Eu/Sm values in Figure 15E indicate the absence of hydrothermal activity in the environment.

Strong correlations can be observed between Ce/Ce* and Eu/Eu* due to diagenetic processes, and the value of these ratios may vary depending on the degree of diagenesis. In this study, the low positive correlation of 0.21 between Ce/Ce* and Eu/Eu* indicates a low diagenetic effect.

Consequently, the paleontological age assessment (late Campanian–Maastrichtian) of the carbonate-dominated Harami Formation depended on radiometric data on larger benthic foraminifera, rudist, and rock samples from five outcrops around Elazığ (Eastern Türkiye). Using the geochemical data (stable isotopes, XRF, ICP-MS, SEM-EDS), the depositional environment in an oxic, shallow-marine carbonate platform setting characterized by climate-controlled variations in salinity, no hydrothermal activity, low terrestrial input, and warm-to-hot, arid climatic conditions was also assessed. The diagenetic alteration is limited except for the Ölbe Stream.

The stable isotopes confirmed the general cooling trend from the late Campanian to the early Maastrichtian, followed by a renewed warming during the late Maastrichtian.