Fig. 9.1. Location of the site (A) and drone image showing individual excavated features (B). As the DaleSice waterworks forms an integral part of the Czech power grid network, and the func- tioning of the Czech Republic electric grid has priority over archaeological excavations, the salvage archaeological works are only possible when the area is not flooded, i.e. only during short time windows during by maintenance breaks (cf. Skrdla et al. 2018). A short visit of the site is often possi- ble in the early morning (ca 6am), when maximum volumes of water from the lower (VD Mohelno) reservoir are pumped into the upper reservoir, and the water level in VD Mohelno is at its minimum. Sunrise occurs later in the day during wintertime and filling of the upper reservoir cannot be com- pleted each day. In addition, the water-logging of sediments just after reduced water levels make movement over the terrain difficult. The above-mentioned factors make all visits, surveys, and exca- vation complicated with uncertain results. Excavation methodology Stone structures KSD and KSE in Mohelno-Plevovce were discovered in early 2019. While KSE was excavated in 2019, KSD was excavated during 2020-21. Both structures were excavated over several subsequent excavation campaigns permitted by the time constraints of the short maintenance breaks throughout 2019-2021. The excavation area is almost always under the water surface so opportunities for archaeological excavations are very limited. While KSA and KSB were excavated during longer maintenance breaks that lasted for several days, opportunities for excavation of KSD and KSE were more limited. The excavation was realized during a series of short breaks over three consecutive days. This unusual set of circumstances required a tailored excavation methodology. The field work must be swift, but precise at the same time — we propose to name this technique ‘salvage-systematic’ excavation. Digging is generally possible for only several hours during the early morning immediately after the area drains, and until the area floods again towards the middle part of the day. The excavated sediments are water-logged and not easy to dig, store, and transport for wet sieving (Fig. 9.3A). The excavation continues until the trenches flood. The following day, water in the trench must be bucketed out (Fig. 9.3B) before the excavation resumes and rising water (from sedi- ment) complicates digging. Before transport to the wet-sieving station, the excavated sediments are placed in buckets (101 each) and individual sectors (0.5 « 0.5 m) and layers (the Ist layer is sediment above the stones and from gaps between individual stones, the second layer is sediment below the stones) are labelled. Some of the sediments are wet-sieved during the excavation on the edge of the water reservoir in the area of the fluctuating water level, and some of the sediment is transported ca 250m out of reach of the fluctuating water level to be wet-sieved later. The shape of the stone structure has been reconstructed from photographs using photogrammetry (Fig. 9.3C, 4). No strati- graphic observations, or identification of possible pits or post holes below the pavement, or artefact bearing horizon are possible in these excavation conditions. Fig. 9.3. Excavation of KSD. Photo by P. Skrdla and J. Bartik. Fig. 9.5. Spatial distribution of finds. Graphics T. Rychtafikova and P. Skrdla. A new question asked for the KSD lithic material was artefact numbers and weight per raw material. Raw materials imported from greater distances including erratic flint and radiolarite follow the expected pattern predicting more economical use — their proportion is greater in numbers than weight, suggesting the raw material was reduced to complete exhaustion. While local quartz artefacts show higher weight values, other local raw materials (including rock crystal, smoky quartz, and plasma) show different patterns. While the number of items and weight proportions for plasma and Krumlovsky les-type chert are similar, the proportion of artefact numbers is twice that of the weight proportion for rock crystal and smoky quartz (perhaps the influence of higher cleavage resulting in more tiny chips?). —_. There are 16 tools (3.2% of assemblage, counted without microlithic tools, but not all have been identified vet), that include burins (5 items), endscrapers (7 items), a borer (Fig. 9.6: 19), a combined tool (a truncation with a notch, Fig. 9.6: 20), and four retouched flakes (Fig. 9.6: 8, 9). The burins include a multiple burin (Fig. 9.6: 32), a burin on a truncation (Fig. 9.6: 22), a burin on a retouched point (Fig. 9.6: 23), a transversal burin (Fig. 9.6: 21), and a burin on a broken bladelet (Fig. 9.6: 31). The endscrapers include two atypical carinated endscraper/cores (Fig. 9.6: 11, 12), two endscrapers ona short flake (Fig. 9.6: 15, 16), an endscraper on a blade (Fig. 9.6: 13), an endscraper on a broken blade (Fig. 9.6: 17), and an endscraper combined with a notch (Fig. 9.6: 14). The last five items are thin in cross section. An atypical carinated endscraper/core (Fig. 9.6: 24) was found during surface preparation before excavation (not included in the inventoried artefacts list). In addition, two flakes (Fig. 9.6: 18) and a bladelet (Fig. 9.6: 3) are partially retouched. L1ICaVvVY QULy StOHe HIGUSUL Y Stone structure D (hereafter KSD) yielded 235 pieces interpreted as heavy duty implements These artefacts are characterized by traces of pounding, grinding, or polishing. These artefacts were identified among the floor stones (however, some of them were located stratigraphically higher, seemingly above the pavement — Fig. 9.3E, F), where orthogneiss predominates. Despite the weathered surface of the orthogneiss, it cannot be conclusively determined whether it was used as part of the floor, or as a functional artefact. Use of pavement stones for a wide range of other activities has already been noted at a similarly dated site in Grubgraben (Montet-White, Williams 1994). In KSD, fragments of slabs with working surfaces made of granulite and amphibolite, as well as fragments with negatives and clearly visible bulbs, were observed and analysed. Tab 9.3. Radiocarbon dating for individual stone structures, calibrated using CalPal software (Weninger, Joris 2008) on the IntCal20 curve (Reimer et al. 2020). Discussion
