Nonmarine Ostracoda as proxies in (geo‐)archaeology — A review

Ostracods as bioindicators are extremely useful for reconstructing palaeoenvironment and palaeoclimate and can also indicate the provenance of sediments and materials, for example, in studies on ancient commercial networks. Ostracods are small crustaceans that live in almost all aquatic habitats, both natural and man‐made. Due to their calcitic carapace, they have high fossilization potential, and their use in geoarchaeology has been steadily increasing during the last decades. Their small size needs mean that only small volumes of sediment samples are needed, and species‐specific ecological tolerances and preferences allow detailed palaeoenvironmental reconstructions. Typical methods of their application are palaeoecological analyses of associations based on ecological information and taphonomy, morphometric variability and stable isotope and chemistry analyses of their shells. The present paper aims to present an overview of applications of non‐marine ostracods in (geo‐)archaeological research, recommending sampling and analytical techniques for addressing archaeological research questions on palaeoclimate, habitat and landscape changes, water availability and quality, land use and other anthropogenic impacts, the provenance of materials and commercial networks to promote the application of Ostracoda in geoarchaeology/environmental archaeology.

highest density and diversity are found in lakes and ponds, that is, standing or slowly flowing waters. However, waters with low ion concentrations and low pH are problematic for ostracods because of their calcitic shell (Ruiz et al., 2013). There are only a few ostracod species in such waters, and the fossil record is destroyed chiefly by dissolution. The opposite is true for carbonate-rich lacustrine deposits, where ostracod shells may form a dominant part of the sand-sized sediment fraction.
The tiny size of ostracods allows high numbers in small sample volumes; for example, from sediment cores or sediment attached to artefacts, diversity is mostly higher than for macrofossils, and preservation is often better because transport results in lower levels of damage.
The presence of indicator species, the species composition of assemblages, morphological variability or chemical and isotopic signatures of their valves document living conditions and anthropogenic impacts through time. These indications can be very specific and quantitative or summarize several environmental factors and show trends. The temporal resolution may range from centuries over decades and years (e.g., Palacios-Fest, 1994) or even seasonal effects (e.g., Palacios-Fest, 1997) depending on the investigated taxonomic groups and methodology. Another interesting aspect of using microfossils in geoarchaeological studies is provenance analysis. They may indicate locations or regions of origin because of their specific distribution in space and time, thus enabling reconstruction of trade connections, exchanges or distributions, mobility/migration and settlement shifts (e.g., Quinn, 2008Quinn, , 2013Wilkinson, 2017;Wilkinson et al., 2016Wilkinson et al., , 2017.

F I G U R E 1
Adult examples of two recent ostracod species from the huge lake Tangra Yumco on the Tibetan Plateau. Left: The ornamented Limnocythere inopinata (Baird, 1843) in external view of a left valve and a dorsal view of two valves still articulated. Right: The smooth shelled Candona candida (Müller, 1776) with an internal view of a right valve and a dorsal view of two valves still articulated. Scale: 0.5 mm.
F I G U R E 2 The ontogenic stages of Cyprideis torosa, a brackish water ostracod, sampled from the early Holocene palaeolake of Tayma, Saudi Arabia. Measurements were taken from a single sample to avoid high morphological variances, except for the measurements of adult valves, due to the overall very low numbers of adults.
Most methods of ostracod-based palaeoenvironmental reconstruction are either palaeoecological approaches or shell chemistry investigations, including stable isotope analysis. Because ostracods are growing by moulting, the chemical and stable isotopic signature of a shell represents a snapshot of the ambient water composition.
Palaeoecological approaches are diverse (Table 1). A fast but rough method is to rely on indicator species (or taxa) reflecting specific conditions or influences such as increased salinity or oxygen deficiency (e.g., Geiger, 1993;Pint et al., 2012). More detailed and reliable results are produced by evaluating the proportion of ecologically classified groups of taxa over a series of samples, either over time or along a transect (De Deckker & Forester, 1988;Frenzel, 2019;Frenzel et al., 2010). However, this method does not provide quantitative reconstructions of environmental variables despite requiring taxa counts within the samples. Quantitative reconstructions of environmental variables are possible by the mutual tolerance method comparing known ecological tolerances of a set of living species using the overlap of these ranges as reconstruction (e.g., Horne et al., 2012b). Transfer functions are more sophisticated in producing quantitative reconstructions with error estimations (e.g., Mezquita et al., 2005;Mischke et al., 2007). They, however, require large training data sets for the regions investigated; such data sets are available only for a limited number of regions. Some species react via changes in morphology like forming nodes or changing the ornamentation and shape of the valves or sieve pores (e.g., Bodergat et al., 1991;Boomer et al., 2017;Frenzel et al., 2017Frenzel et al., , 2012Yin et al., 2001). These environmentally induced morphological changes can be used for palaeoenvironmental reconstructions. Shell chemistry approaches use trace element or stable isotope signatures of ostracod shells for reconstructing past salinity, water chemistry, redox conditions, temperature, and so forth (Börner et al., 2013;Holmes & Chivas, 2002;Holmes & De Deckker, 2012).
Taphonomical data provide additional information about the depositional environments (Boomer et al., 2003). A more detailed overview of all these methods is given in Griffiths and Holmes (2000), Frenzel et al. (2010) and Horne et al. (2012a). This article will present an overview of the state of nonmarine ostracods in (geo-)archaeological research, highlighting applications and methods that are already well used and promoting the further use of ostracod proxies in this field.
2 | APPLICATIONS 2.1 | A short history of ostracod-based palaeoenvironmental reconstructions The history of ostracod research in a palaeo-science context started with descriptions of fossil taxa in the early 19th century, with the first stratigraphic investigations appearing around 50 years later (Hartmann, 1966). Their usefulness in geological, especially biostratigraphic research, made them one of the classical groups of micropalaeontology.
Quaternary palaeoecological studies are especially useful for the geoarchaeological context, and many of them answer overlapping research questions in palaeoclimatology, palaeolimnology, pollution records and archaeology. While the first studies on Quaternary ostracods reach back to the mid-19th century (Jones, 1850), these merely listed species without systematic attempts to reconstruct the palaeoenvironments. After some rare studies applying ostracods in analysing Quaternary environments (Griffiths & Holmes, 2000 and review therein), there was an initial phase of Quaternary studies in the 1960s and the 1970s that has continued up to the present day.
Geoarchaeological studies based on non-marine ostracods are more numerous today, as we will show below. Nevertheless, their potential is still estimated as under-utilized (e.g., Kenward, 2009).

Salinity
For example, at several sites in the United Kingdom, ostracod assemblages were used to reconstruct the environments and to characterize the nature of freshwater sources in the area, for example, a slow-flowing river with adjacent marshlands during the late Middle Pleistocene (for an overview of geological time scales, see Figure 5) interglacial site at Purfleet in Essex (Schreve et al., 2002), small permanent groundwater-fed ponds at the Boxgrove site in West Sussex Roberts & Pope, 2009;Whatley & Haynes, 1986;Whittaker & Parfitt, 2017) or shallow groundwaterfed ponds with minor fluvial input in the White Peak in Derbyshire (Taylor et al., 1994).
At the Youfangbei Locality, North China, a small Middle Palaeolithic flake industry in the Nihewan Basin of Yangyuan County, Hebei Province, ostracods and bivalves indicated a wetland environment on a river terrace, and showed a climatic change from a drywarm to a cold-wet environment around the end of the last interglacial (Zhao et al., 2021).
Ostracods from a sediment core were also used to reconstruct the environmental history since the Late Glacial of Lake Ochaul, in the Upper Lena region of Eastern Siberia, known for its Palaeolithic, Mesolithic, Late Neolithic and Iron Age archaeological sites. The ostracod assemblages indicate lake-level changes, but show that the lake never dried out and, therefore, were in all likelihood an important source of water and hunting ground for hunter-gatherers (Kobe et al., 2021). Stepanchuk & Moigne, 2015).
In the Anagni basin in central Italy, abundant tool industries and fossils indicate favourable conditions for the Lower Palaeolithic hominin occupation of the area, which developed when large lacustrine basins and alluvial plains started to progressively shallow and temporarily emerged. This was seen in a combination of ostracod and palynological records and facies analysis (Florindo et al., 2021).
Ostracod abundances and taphonomic factors (abrasions, iron stains and calcium carbonate infillings) were used in the study of the Palaeo-American Gail Stone Site in Wisconsin, USA, where they, together with an aeolian lag deposit, indicated that the Palaeo-Americans deposited artefacts on a low terrace surface (Newman, 2001).
In the arid southern Great Basin, eastern California and Nevada, USA, prehistoric human occupations seem to be linked to episodic lake high-stands in the Ivanpah basin. Spaulding et al. (2021) concluded with remote imagery and several excavations that pluvial palaeolakes must have existed for decades or centuries in the Ivanpah basin during the Holocene. The freshwater ostracod Limnocythere ceriotuberosa indicated cold, freshwater conditions and, therefore, a large, open lake rather than a paludal habitat.
The development of marshlands in connection with a shallow lake in the Lake Bonneville Basin in western Utah's Great Salt Lake Basin was thought to have attracted human occupations. The lake's formation, which was indicated by ostracod species as cold and fresh, was concurrent with the long-term use of the basin between 13,000 Hampshire, UK, were described by Davies and Griffiths (2005).
Ostracods reflected the hydrological conditions and showed that these tufas were groundwater-fed systems that covered considerable areas of the major river valley, in contrast to the other, usually springfed, tufa systems in the United Kingdom.
Environmental changes were also important at the transition from the Mesolithic to Neolithic and were noted both in the Holocene of the Polish lake site of Dąbki and at Queens Sedgemoor in Somerset, UK. At Dąbki, silting up of the lake and subsequent development of more lake shore vegetation worsened the accessibility of the lake shore (Laschke et al., 2015), while in Queens Sedgemoor, the whole lake environment changed to a raised bog (Hill et al., , 2016. At the nearby UK localities Moorland House and Saltmoor near the River Parrett, brackish water ostracod species indicated a connection to the coast. The Parrett valley was therefore interpreted as an estuarine environment with strong tidal action and likely used for foraging by Mesolithic groups, but not for their camps (Wilkinson et al., 2021).
In the Neolithic, the development of farming and permanent settlements was influenced by environmental characteristics (e.g., At the palaeolake, Haijad in the Central Sahara that was occupied during the Neolithic, morphological changes of ostracods reflected cycles of disturbances in the carbonate balance, which correspond to low-amplitude climatic cycles (Carbonel, 1991).
Even the accessibility of sites has been investigated through ostracod analysis. Ostracods from the Grotta del Lago cave in Italy showed that it was only accessible at low lake levels during the Atlantic (cf. Figure 5), which correlated with the time of human occupation (Taliana et al., 1996).
With the technological development during Ancient and Postclassical history, research moreover focuses on the increasing human impact on the environment (cf. Ruiz et al., 2013). In the Levant, Chalcolithic to Early Bronze Age settlements at the eastern tributaries in the Upper Khabur basin in Syria were, however, still dependent on humid phases and water accessibility. There, a steady flow of water in the tributaries was demonstrated by the ostracod assemblages, while today, they run dry during summer (Deckers & Riehl, 2007).
At the Bronze Age Crabble Paper Mill site in Dover, UK, ostracods from tufa springs were studied to reconstruct the palaeoenvironments. While the site was assigned to the 'burned mounds' sites that are always located adjacent to water bodies, the ostracods indicated that the tufa springs were much less developed and extensive than similar springs, such as the complex tufa pools at The Grove site near Watford (Bates et al., 2008).
With Bronze to Iron Age finds in the overlying clay pits, tufa deposits in the Ancholme Valley in Lincolnshire, UK, were also analysed for their ostracod fauna. For the three sites in the valley, the assemblages only reflected small local changes in the depositional environments. All assemblages were relatively similar and dominated by spring-dwelling species (Preece & Robinson, 1984).
In Turkey, the palaeoenvironmental history from the Lycian to the Late Roman periods for the city area of Limyra in SW Anatolia was studied using ostracods. Freshwater ostracods indicated a lacustrine environment in a sheltered area, probably a former lake in the Finike plain. In the mid-1st millennium (Late Roman Period), the lake started silting up, river channels evolved and people settled in the area, which later became the city of Limyra (Stock et al., 2020).
Ostracods have been used to reveal the palaeoenvironmental history, for example, at a Roman bridge in Fréjus, France, that today is located far from any rivers. The ostracods showed a transition from a river mouth setting to a closed lagoon with increasingly fresh water during the Roman times, and finally, a fully continental setting with subsequent desiccation (Allinne et al., 2006).

Morphological variation
Salinity reconstructions based on ostracod analyses are relatively well used in geoarchaeological studies. However, analyses of morphological variations have not been considered much yet, as the methodology is relatively new. An example, using sieve pore analysis in Cyprideis torosa amongst other proxies, are the studies on the palaeolake of Tayma in north-western Saudi Arabia to identify humid periods that favoured Neolithic settlements. The ostracods indicated an early Holocene saline, but permanent lake with substantial seasonal hydrological variations Engel et al., 2012Engel et al., , 2017Pint et al., 2011;Pint, Engel et al., 2017).

Relative abundance
It is much more common to use ostracods as salinity proxy through the relative abundance of indicator species and ecological groups or conductivity transfer functions relying on ostracod assemblages. Also, in the Hula valley, ostracods from the Jordan River Dureijat archaeological site were analysed for palaeoenvironmental reconstruction. They indicated a relatively large lake with probably slightly colder water temperatures than today. The lake was permanent, but probably with seasonal fluctuations, and the conductivity transfer function of Mischke, Ashkenazi et al. (2014) suggested strictly fresh water. Further ostracod taxa indicated that in a much younger sample, from the Late Holocene, the freshwater near-shore setting changed into a lake more similar to the modernday Sea of Galilee (Eyþórsdóttir, 2019;Valdimarsson, 2017).
Besides providing information about the suitability of a territory for early humans and thus indicating possible routes for early dispersals, ostracods have also been used to explain early human adaptations. At the Early Pleistocene Barranco León site in southern Spain, which is one of the most important sites for the human evolution and early occupation in Europe, ostracod assemblages, stable isotopes and trace elements indicated that the human occupation took place when the marginal zone of the adjacent lake was characterized by an oligohaline to mesohaline through-flow open lacustrine system, fed by groundwater and meteoric stream water. The section yielding archaeological remains was deposited during water-level changes with corresponding erosion and redeposition, as reflected by a high abundance of imbricated ostracod valves, and the tools are thus not in situ (Anadón & Gabàs, 2009;Anadón et al., 2003;De Marfà, 2007 (Biddittu et al., 2020).
The interglacial ostracod fauna found in the sediments of the Middle Palaeolithic Geiseltal area in Neumark-Nord, Germany, was markedly different from other Eem Age (cf. Figure 5) ostracod assemblages, as it, for example, included species that today are restricted to Africa. The ostracod assemblage indicated an increase in salinity in the water body, probably due to increased evaporation during a climatic change (Fuhrmann & Pietrzeniuk, 1990;Mania, 1992).
Ostracods were used in a multi-proxy study from a saline wetland in 'Nagla Lu', Xizang, southern Tibetan highlands, which shows a continuous fire record attributed to humans since the Last Glacial Maximum. Freshwater species showed that the catchment of the area was less arid at the end of the Last Glacial Maximum than it is today due to higher winter precipitation (Miehe et al., 2021). Valley in Romania appears to have been an attractive environment for human settlement. In a multiproxy study of the lake, most of the ostracods found were cosmopolitan species, but they still indicated a stable lacustrine environment, which accords with the high availability of resources like fish and molluscs (Nowacki et al., 2019).

Stable isotope analysis
Humid periods have been identified in the Neolithic wetland sites in Unfriedshausen and Pestenacker-Nord in southern Germany, based on ostracod taxa and stable oxygen and carbon isotope signatures from shells of abundant species. These data showed that climatic changes lead to larger permanent water bodies, which, in earlier research, were discussed as having attracted the possibly contemporaneous human settlement (Janz & Matzke-Karasz, 2001;Mayr et al., 2015). America. Mg/Sr and Sr/Ca values were used to infer two major climatic events that caused two major floods and one drought ( Figure 6).

Mutual ostracod salinity range
Ranges with reconstructed maximum and minimum salinity values can be displayed with the Mutual Ostracod Salinity Range F I G U R E 6 Ostracod data from the Hohokam irrigation canals in central Arizona, and the corresponding interpretation of the palaeoenvironmental history and human occupational history. The displayed ostracod data are the number of ostracod valves per gram of sediment (Population Index), the number of different species found (Species Richness Index), the relative abundance of six ostracod species as indicator species for salinity tolerance and a thereof derived quantitative salinity index, the palaeosalinity based on trace elements (Sr/Ca) and the palaeotemperature based on trace elements (Mg/Ca). The shaded areas show major palaeoenvironmental events that are reflected in the different ostracod proxies. Modified from Palacios-Fest, (1994,1997).
(MOSR) (Pint, Schneider et al., 2017). The method is strongly dependent on the presence of index species with narrow salinity tolerance ranges, but has a better coverage of species in comparison to conductivity transfer functions that rely on large training data sets.

Relative abundance
Temperature reconstructions with ostracod proxies are, similar to salinity reconstructions, often based on the relative abundances of indicator species and ecological groups.
At the Hoxnian (cf. Figure 5 In past River Thames channels at Syon Park in Brentford, correlating stratigraphically to the Late Palaeolithic settlements of the area, ostracods seem to indicate a thaw episode during the Last Glacial Maximum, as the species found are today associated with summer tundra pools in periglacial environments (Corcoran et al., 2012).
The environmental transition from the late Last Glacial to postglacial at Lough Boora, one of the oldest Mesolithic sites in Ireland, was reconstructed based on ostracod assemblages. They showed a transition from a lake with carbonate sedimentation and poorly developed vegetation to a cool-temperate shallow lake with a swampy littoral zone and abundant lower plants (Griffiths, 1998).
The late Pleistocene-Holocene climatic transition was also seen in ostracods from a lake at the famous Early Mesolithic Star Carr site in North Yorkshire, UK. Late-glacial ostracod faunas reflected a calcareous, relatively deep lake with a low trophic state and temperatures, while ostracods from a Holocene sample showed the onset of warming, leaching and a shift in depositional regime (Holmes & Griffiths, 1998).
The Holocene environmental change also favoured human settlement at the Elbe river valley in Germany, where ostracods in a comprehensive multiproxy study were used for quantitative temperature reconstructions, showing a temperature rise of 4-6°C in the early Preboreal (cf. Figure 5), and describing a diverse landscape structure that was thought to have attracted human settlement (Turner et al., 2013).
Holarctic ostracod species were found in Late Quaternary lacustrine sediments from a dry lake basin in the south-eastern Pleistocene. While the Holsteinian (cf. Figure 5) Bilzingsleben site in Germany showed only slightly higher temperatures than modern day (Daniel & Frenzel, 2010;Diebel & Pietrzeniuk, 1980) (Figure 7), MOTR from the Hoxnian (cf. Figure 5) Beeches Pit site in Suffolk, UK, further showed that the human occupation not only took place during an interglacial but also continued into a succeeding cold phase (Benardout, 2015). Other Middle Pleistocene sites in the UK, like the Hoxnian (cf. Figure 5

| Radiocarbon dating
Radiocarbon dating has been shown to work on ostracod valves to some extent, but the valves of these water-dwelling animals have a reservoir effect similar to aquatic molluscs, and many adult valves are   (Taylor et al., 1994). In contrast, human modification of the landscape since the Late Mesolithic was presumably the reason for the transition of a freshwater lake to a raised bog at Queens Sedgemoor, Somerset, UK, as shown by a multiproxy study of the palaeoenvironment including ostracods (Hill et al., , 2016.

| Landscape changes by human activity
Anthropogenic activity was also revealed in the palaeoenvironment at the Early Celtic site on the mountain Ipf in Baden-Württemberg, Germany, by correlating periods of soil erosion and settlement history. Further, the palaeoenvironments of the site were reconstructed with ostracods, showing open standing water, followed by aridification and a transition from a marsh to a spring brook (Fischer et al., 2011).
Eutrophication of a lake can be detected with ostracods, as in Lake Coba on the Yucatan Peninsula, Mexico, at a time when the lake was at its deepest. This eutrophication was probably a result of agricultural forest clearance by early urbanization and development of farming villages at the lake during the Late Preclassic and Early Classic periods (ca. 600 BC-600 AC) and was demonstrated with ostracod assemblages and stable isotope ratios, combined with, for example, diatom and pollen analyses (Whitmore et al., 1996).
Besides assemblages and isotope data, Fleury et al. (2015) included ostracod abundances in their study of sediments from the | 721 deterioration of living conditions by high suspension loads and oxygen depletion in the lake. This also resulted in a reduction of the photosynthesis activity, which is reflected in a decrease in oxygensensitive ostracod species and 13 C depletion in the valves.
Palaeoclimatic studies have often attempted to explain the socalled Mayan collapse, as their swidden and wetland field agriculture depended on the climatic conditions. In northern Belize, this collapse happened at around 1000 AD, where ostracod assemblages from the Laguna de Cocos on Albion Island reflect a falling lake level coincident with a decline in agricultural activity (Bradbury et al., 1990). The importance of humidity for agriculture was also seen in Bronze Age Harappan Civilization at Lake Quinghai in China. δ 18 O values showed a minimum phase of increased humidity from ca. 11.6 to 9.4 cal ka BP, followed by a gradual decline in the mean annual temperature. The increasing aridity and dryness may have led to more efficient agricultural practices in the beginning and the eventual collapse of the Harappan Civilization between 3.5 and 3 cal ka BP (Leipe et al., 2014;Lister et al., 1991).
High erosion rates in connection with the reduction of woody vegetation were also seen in the Upper Khabur Basin in Syria around the 9th century AD, resulting in the aggradation of the floodplain. Deckers and Riehl (2007) attempted to include ostracods in their interpretation of the geomorphology of the basin, which presumably was partly human-influenced, but unfortunately, too few ostracods were sampled.
Deforestation may, besides increased erosion rates, also lead to increasing water yield, as the evapotranspiration will be reduced by a Furthermore, a decrease in the ostracod diversity can be used to determine the scale of anthropogenic influence. In Lake Sevan in Armenia, Holocene ostracod assemblages have been shown to be much more diverse compared to subrecent and recent assemblages, which is thought to be a result of the high anthropogenic activity in the area in the last decades. However, ostracods indicated that the lake inundated the Bronze Age landscape during more humid climate conditions. In addition, a major volcanic episode probably dammed an adjacent river, and the high lake levels remained until the early 20th century (Wilkinson, 2020;Wilkinson & Gulakyan, 2010).
This was also seen,for example, in Haarhausen, a Germanic site that shows a significant Roman influence (second half of the 3rd century AD), in Thuringia, Germany, where ostracods were analysed in two profiles: one from a stagnant lake with a swampy shoreline and one from a smaller marsh with proximity to a spring. Differences in the ostracod and mollusc assemblages and diversities from both profiles showed that the small marsh profile comprised evidence for anthropogenic influence, indicating that the settlement only extended towards the small marsh and not to the lake (Keding et al., 1995).

| Water use and water works
Ostracods can be used to detect past water works and water use. Non-marine ancient harbours have also been analysed through ostracods, such as the Magdala Harbour in the Sea of Galilee, Israel, which was surrounded by areas of intensive agriculture during the Hellenistic and Roman/Byzantine periods. The ostracods showed the transition from a preharbour beach to the harbour basin and that the harbour area was presumably protected artificially, which is seen in an increase of organic matter and alkali enrichment in the ostracod valves, a decrease in water energy and a change in salinity (Lena et al., 2017;Rossi et al., 2013Rossi et al., , 2015. On the Danube Delta in Romania, at the Roman fortress of Halmyris, ostracods were used to identify the presence of a fluvial channel north of the settlement of Halmyris, which was navigable throughout the occupational periods, from the Getic/Greek to the Roman period, and could have been used as a natural harbour (Giaime et al., 2019).
In the Roman sewer systems from Church Street, York, UK, two bottom-dwelling ostracod species were found that further indicated that the sewer systems contained shallow and not necessarily clear water, with low salinity and little or no living plant material. The water was probably moving only slowly enough to keep it from stagnating (Buckland, 1976).
Ostracods have also been found in high diversity in the medieval town moat of Greifswald in Germany, where they indicated high productivity and moderate eutrophication of the water because of human activity. The assemblage further showed that the flow direction in the moat presumably was dominantly into the river Ryck, while saline water from the Baltic would only be introduced at flood events (Frenzel et al., 2004).
In the Canadian Arctic, the Sadlermiut used small, shallow ponds to clean and prepare their subsistence harvest. A sediment analysis including ostracods of a freshwater pond close to the archaeological site 'Native Point' on Southampton Island, Nunavut, showed fundamental changes in the ostracod adult/juvenile ratios, species richness and frequency because of eutrophication. The anthropogenic influence was seen since the arrival of the Sadlermiut around 1250 CE and is still visible .

| Provenance studies
Ostracods found in, for example, pottery or building material can be useful in indicating provenance, thus distinguishing between locally produced and imported materials, thus pointing to trading and exchanges, or distributions, mobility/migrations and settlement shifts (see Quinn, 2008Quinn, , 2013Quinn & Day, 2007a;Wilkinson, 2017;Williams et al., , 2017. In this section, fossil marine ostracods will also be considered, as the ostracod-bearing materials may originate from very different settings than where they were sampled. Quinn and Day (2007b)  The presence or absence of ostracods was also crucial for a provenance study of clays in Garamantian ceramics from Jarma, south-west Libya. The different number of ostracods in specific red on white painted ware showed that it was both traded, as it, in contrast to local clays, contained ostracods, but that it also was imitated, as demonstrated by the absence of ostracods (Leitch et al., 2016).
The high abundance of ostracods in the sherds of Mediterranean amphorae from the Gulf of Hammamet, Tunisia, has been suggested as a primary determining feature of this production site, which can be used to distinguish it from other African amphorae (Capelli & Bonifay, 2014 (Toykailo, 2012).
Ostracods in ceramics can also be used for firing temperature reconstruction based on physical alteration. This was shown in ceramics from the Sasanian archaeological site Qizlar Qal'eh, Iran, where Early Cretaceous and Quaternary ostracods and other microfossils suggested that the material was derived from alluvial sediments taken from an adjacent site north-west of the Gorgan River plain, and the altered internal ultra-structures of the shells indicated firing temperatures of 650-850°C (Daghmehchi et al., 2015).
Ostracod studies have also determined the provenance of other materials such as microfacies in Roman mosaics and sculpture material from Friedberg and Unterbaar, Germany. Here, the microfacies of black mosaic stones containing ostracods were assigned to local sediments, while other microfacies were derived from other regional sources, which showed that the combination of local and regional sources for the mosaic stones was common (Flügel, 1999). Wilkinson et al. (2008) showed that it might be possible in some cases to identify ostracod species from mosaics as they were able to define two species in the chalks of Roman mosaic stones at Silchester, UK, but as these were long-ranging taxa, they provided no detailed biostratigraphic information.
Provenance studies of building materials have also sometimes focused on microfossils, including ostracods. One example is the building material from the late Iron Age of the Burrough Hill fort, which had its source in a local Pleistocene till, based on the identical microfossil signature (Wilkinson et al., 2013;Williams et al., 2015).
The same was the case for ceramics from Burrough Hill fort, and Wilkinson et al. (2017) showed that, based on the microfossil signature, ceramics from Romano-British sites in England were also derived from nearby sources.  applied microfossils, including ostracods, in a study of more recent building material from an English Civil War (mid-17th century) bastion at Wallingford Castle in Oxfordshire and showed that it also derived from a nearby source. They conclude that microfossils can be applied to a wide range of provenance studies for classical to modern building materials. at a maximum temperature of 60°C, the ostracods can be picked with a small brush or needle and collected in microcells using a stereomicroscope ( Figure 9). A comprehensive description of the processing of ostracod samples from archaeological sites is given by Griffiths et al. (1993).

| METHODS FOR SAMPLING AND LAB ANALYSES
For species identification, classification literature is necessary. An overview of recent freshwater species of the world is given by Karanovic (2012), while regional literature such as books on regional ostracod faunas like, for example, Meisch (2000) or Fuhrmann (2012) for central Germany, or Henderson (1990) for Great Britain, is necessary as taxonomic references. A minimum of 300 individuals is recommended for statistically significant species abundances and for the use of mutual range methods. Multivariate statistics such as principal component analysis (PCA) and cluster analysis can be applied to compare samples and identify potential driving factors.
Shell chemistry of ostracod valves comprises analyses of stable F I G U R E 9 The process of sampling and sample processing: The sample is placed in a plastic bag to keep it moist; if necessary, it is processed with hydrogen peroxide to disintegrate the sediment, then sieved with water on mesh sizes 125 and 200 µm and dried at 60°C before ostracods can be picked under a low-power stereo-microscope. [Color figure can be viewed at wileyonlinelibrary.com] isotopes, trace elements and radiocarbon dating. One of the most common species whose shell chemistry is well understood is C. torosa and, is therefore easily comparable with other studies. To investigate the preservation state of the valves, which is very important to indicate reworking and erosional processes, a scanning electron microscope or a high-resolution light microscope can be used.

| CONCLUSIONS
Ostracods have been proven to be good tools for palaeoenvironmental reconstructions and have, especially for this reason, increasingly been especially where no species-rich association and no modern training data sets are available, while δ 13 C can be used for the analysis of the productivity or nutrient availability.
However, the mutual use of the different palaeoenvironmental proxies based on non-marine ostracods in geoarchaeological studies has still more perspectives. Other trace elements, such as trace metals (e.g., Cd, Ba, Zn), U/Ca ratios or 87 Sr/ 86 Sr and 143 Nd/ 144 Nd ratios, have not or only rarely been used in the geoarchaeological context, but could provide further information about the palaeoenvironment of a water body, such as the productivity, oxygenation and water source. Ostracods have been used for radiocarbon dating in very few geoarchaeological studies, but with a successful outcome where enough material was available.
Ecologically induced morphological changes like size differences, nodes or sieve pore variations in the valves of some species have been proven to be useful tools and additions to nonquantitative species and assemblage analyses for a more detailed palaeoenvironmental reconstruction, and they could be considered for many archaeological studies involving aquatic sediments.
Fossil ostracod associations can also be analysed taphonomically.
For example, the adult/juvenile ratio or the valve/carapace ratio can provide information on the water depth, sedimentation rate, energy regime of the water body or environmental stress.
The use of ostracods for provenance studies of many building materials or ceramics is less beneficial, as ostracods are usually damaged when attempting to extract them from the matrix. However, their mere occurrence is easy to recognize in thin sections and can play an important role in tracing the provenance of the material.
This overview of geoarchaeological studies, which all in some way used non-marine ostracods, gives an idea of the applicability of ostracod proxies for various (geo-)archaeological research questions.
At the same time, it also displays the sparsity of detailed nonmarine ostracod studies at archaeological sites, which, regarding the state of research and development of new and better ostracod proxies, may further increase in the coming years.