Environmental Impact on Fossil Record for Palaecological Reconstruction Studies

Paleoecological studies have an important role in understanding past environmental, dietary and/or societal changes however require the authentic signature of fossil materials. Therefore, a significant part of these studies concerns the isolation of the material authentic matrix. Bone hydroxyapatite from different animal species from the archaeological site of Dispilio in Kastoria Lake basin in northern Greece has been subjected to mineral analysis in order to detect if there are suitable for palaoecological studies. Calcium, phosphorus, oxygen and hydrogen are the main components of bones resulting rigidity, hardness and compressive strength of their structure. However different bone structure resulting different calcium- phosphate phases and different compositions, including Ca/P ratios. These disparities may be attributable to different physiological characteristic, conditions under which the bones were formed or burial environment. Trace element analysis (Ca/P, Sr/P, Fe/Mn) concluded that treated fossil bones retained their biochemical signal without any strong influence by soil remains however without suggesting that no chemical alteration have been occurred.


Introduction
"Fossil bones" do not always constitute "archaeological bones", with respect to reconstruction terms, as many alternations are taking place through their passage from the biosphere to the lithosphere and their finally fossilization. In order to reconstruct information related to physiology or diet habits, environmental conditions or migration episodes, a significant part of paleoecological studies have been focused in the isolation and the preservation of the authentic chemical structure of bones. On the other hand, chemical, mineralogical and histological alterations during diagenesis process could also be a considerable information reservoir leading to an interpretative tool for taphonomic characteristics and diagenetic environment. One of the main objectives of this study is to read the reliability or not of skeletal samples for paleoenvironmenal and paleoecological reconstructions with mineralogical methods.

Trace elements in bone and teeth samples
Calcium, phosphorus, oxygen and hydrogen are the main components of bones resulting rigidity, hardness and compressive strength of their structure. However different bone structure resulting  4 3-or OHin the lattice resulting in slight differences to bone ratios [1][2][3][4]. Measured weight ratio values for modern deer were between 2.07 and 2.19 (Table 1) with mean value 2.13 which coincided with that of fossil bones. Moreover, literature data [5] for modern Vulpes vulpes and measured human bones exhibit mean Ca/P values about 2.1 and 2.07 respectively. These values are relatively close to those of hydroxylapatite indicating the biochemical stability of hydroxyapatite deposit environment, without excluded any fluctuations due to different skeletal samples. On the other hand, measured data of raw fossil bone samples (untreated samples) exhibited mean Ca/P ratio 2.33 (Table 1), a value that turns away its print from both modern bones and hydroxylapatite. So considering that Ca/P ratio of treated samples is close to that of hydroxyapatite, it could be concluded that the pretreatment of the samples (acetic acid) managed a sufficient secondary calcite removal.

Soil chemistry and ionic interaction with soil solution
The contamination from the surrounding soil through the interaction of bone/teeth samples with the soil solution it should be further considered. Silicon, iron, manganese and aluminum are enriched in sediments but normally absent in living bone. So their incorporation is possible through secondary processes in burial environment as new crystals of phosphate mineral formed during diagenesis. Soils reflect the parent rock material since are the result of biological, chemical, and physical weathering. Inorganic and organic signature of any soil type influences the growth of plant roots supplying them with minerals. Moreover, diagenetic fluids reflect the soil profile in which they circulate [6,7], therefore, soil chemical profile was constructed. Sediment core related to the taphonomic conditions of fossil bones is characterized by ten horizons from the surface to 2.00m with 0.20m depth step. High concentrations of Fe 2+ and Mn 2+ were detected which related by a strong correlation coefficient (r 2 =0.87) ( Figure 3). In more detail (Table 1) Fe/Mn ratio ranged from 2.84 to 4.06 (mean 3.6). Their ratio remained relatively constant with respect to depth however the two horizons of 1.40 and 1.80 differentiated displaying lower values ( Figure 1). These layers accompanied by a grey-dark black color as well as by diatom bloom (Figure 2), in contrast to rest layers which characterized by a brown-grey color, implying an elevated primary productivity [8,9]. Previous studies have come up on conclusions regarding the soil zonation in Dispilio excavation based on micromorphological analysis of soil. Detailed, soil layer of 2.01-1.80 m corresponds to lacustrine sediments and anthropogenic materials that have deposited through physical processes within the lake. This soil layer represents a coastal lacustrine environment of high energy with strong waves and currents [10] in direct interaction with the open lake.
The strong presence of anthropogenic material confirms the occupation of the site by humans with constructions above the water [11]. The soil layer of 1.80-1.00 m is characterized by alternate depositions from waving to stagnant water corresponding to an environment with undisturbed ponds, rich in reeds, matching to the present picture in the coastal zone of the lake in front of the excavation site [12]. The interaction with the well-mixed open lake and the physical deposits charged the soil with large amounts of silica allowing high biological productivity (which is evident, as previously reported, on the color shade of soil) and the undisturbed environment gave a boost to the development of diatoms as identified by microscopic and elemental analyzes of soil core under this study. Finally soil layer of 1.00-0.40m correspond to a period without strong interaction with the lake water except wet seasons [11]. The picture of Kastoria Lake with a multiple interaction with sediments overtime is supported by the general observations that been documented by Magny et al. [13] about the instability in European lake-level in Holocene rather driven by climatic oscillations. Soil samples as well as raw and treated bone samples pictured in Figure 3. In this diagram two conditions of soil are presented. The first (black long dash dot line) concerns the entire soil layers while the second (black solid line) excludes soil layers of 1.80 m and 1.40 m as have been influenced by the diatom bloom and disrupt the territorial ratio of Fe/Mn. The resulted correlation is much stronger (r 2 =0.96) with the trend line to end at the bone area. Furthermore, attention was paid on the slopes correspond to each sample group as presented to Figure  3. Entire soil layers exhibited slope "2.5" while without soil layers of 1.80 m and 1.40 m (diatom bloom) exhibited slope "3.4". The fist circumstance described by trivalent cations where due to their greater charge are held more tightly in soil surface than divalent or monovalent. The second circumstance described by cations with the same valence but smaller radius resulting in strong bonds because of the shortest distance between positive and negative charges. As a result, it is expected a greater cation concentration in soil particle surface and lower concentrations to soil solution [14]. Therefore, raw bone samples present similar slope with soil slope reflecting soil particle contamination while the completely deferent slope of treated bones implies the effective treatment of samples for bioapatite extraction. The fact that Fe/Mn slope of soils does not retained in bones however trendlines are met, possible reflects the process though which plants enriched in trace elements from soils and then transferred to the bones through diet. A competitive relationship has been reported between Fe 2+ and Mn 2+ in plants where Fe 2+ prevents Mn 2+ accumulation, or vice versa, either during uptake by the roots, or during translocation from the roots to the leaves or other above ground parts [15][16][17]. This is also evident in this study where the correlation factor between Fe 2+ and Mn 2+ ions is really weak (r 2 =0.05) influenced by higher Fe 2+ concentrations correspond to lower Mn 2+ concentrations and inverse. Unlike raw fossil bones exhibit a better correlation (r 2 =0.37) and a slope closer to that of soil implying their influence from soil remains. Finally, a negative correlation between Ca 2+ and Fe 2+ was detected, however wasn't strong (r 2 =0.4), implying that Fe 2+ concentration isn't governed by calcite dissolution [18]. In mammal physiology strontium is considered to substitute the essential element of calcium and concentrate in skeletal bioapatite [19]. The interaction between plants and soil reservoir and the passage of trace elements to bones are also indicated by Sr/Ca ratio. Many studies [20][21][22][23][24] have been focus on major and trace components of inorganic phase of bone as they constitute reliable tracers for the quality and quantity of ingested food with the trace element of Sr to gather a great interest. About the 99% of strontium corresponding to vertebrates found in bones with less than 10% originate from water, as plants take up strontium concentrations mainly by soil reservoir. The biochemical path of strontium begins from soil to plant and then to bones of herbivores through their diet. Plants absorb strontium along with calcium in proportions roughly equal to its presence in the environment therefore Sr/Ca ratio of plants should respond to the respective Sr/Ca ratio of soil. Discrepancies often attributed to different kind of plant and the part of the plant consumed or studied. Higher Sr concentrations accumulated through woody vegetation in contrast to grasses, therefore, browsers (e.g., animals eating leaves, shoots, etc.) exhibit higher concentrations of strontium than do grazers (e.g., animals eating grasses, etc.). Sr/Ca ratio ranges between 0.71 and 1.98 (mean 1.25) for treated fossil bones while soil samples characterized by higher ratio ranging between 1.35 and 3.48 (mean 2.15) ( Table 1). A reduction of Sr/Ca ratio through the system soil-bone it is observed which is consistent with the fact that strontium amount decreases up the food chain as animals preferentially retain calcium while excreting strontium; however only a ratio of ingested strontium accumulate in bones and teeth of herbivores as this trace element is not fully excreted by organisms [25]. Modern deer bones exhibit Sr/Ca ratio 1.10, close to the ratio of treated fossil bones (1.25) as well as the ratio of raw samples (0.93). Based on this observation it could be assumed that biogenic signal is retained in treated fossil bone in Displio excavation however without suggesting that no chemical alteration have been occurred. The good condition and conservation of bones have also been documented by Nellie Phoca-Cosmetatou et al. [26].

Conclusions
Major pathways that diagenetic mechanisms take place are the precipitation of secondary calcite salts or mineral phases in the carbonate matrix of hydroxyapatite, burial contamination from the surrounding soil through the interaction of bone/teeth samples with the soil solution and recrystallization of biogenic apatite to larger crystallized form. Mineralogical analysis managed to address sufficiently the first two mechanisms. Through Ca/P ration in raw and treated bones was concluded the sufficient removal of secondary calcite. Moreover, through Fe/Mn ratio identified the soil particle contribution to raw material and their efficient pre-treatment. The trace element analysis based on Fe/Mn, Ca/P and Sr/Ca ratios concluded that treated fossil bones retain their biochemical signal without any strong influence by soil remains. However mineralogical analysis exhausted its limits in detecting recrystallization processes.