Quaternary International xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Quaternary International

journal homepage: www.elsevier.com/locate/quaint

Archaeological, geomorphological and cartographical evidence of the sea level rise in the southern Levantine Basin in the 19th and 20th centuries

∗ E. Tokera,b, , J. Sharvitc, M. Fischerd, Y. Melzere, O. Potchterf a Kibbutzim College, Mordechai Namir 149, -Yafo, b Porter School of Environmental Studies, Tel-Aviv University, Israel c Underwater Unit of the Israel Antiquities Authority, Rockfeller Museum, , Israel d Institute of Archaeology of Tel-Aviv University, Israel e Survey of Israel, Lincoln St 1, Tel Aviv-Yafo, Israel f Department of Geography, Man and Environment, Beit Berl College, 44905, Israel

ARTICLE INFO ABSTRACT

Keywords: Sea level variability affected by sea water mass is strongly associated with global, regional and local climate. In Sea level this context, the eastern has been intensively investigated in recent decades because of its Archaeological sea level indicators sensitivity to climatic and environmental variables, due to the influence of the Eastern Mediterranean Transient Historical maps (EMT). The sea level in Israel during the Crusader period (12th-13th centuries CE) was found to be 19th century −0.5 ± 0.20 m relative to the present mean sea level (MSL). The difference between the Crusader sea level and Acre the present-day MSL raises some questions which bring us to the aim of this study: estimating the timeline of the Jaffa changes in sea level elevation in the eastern Mediterranean over the last two centuries. Archaeological evidence from areas of low tidal range, such as the Mediterranean Sea, can provide significant information on sea level changes for times when instrumental measurements where not yet available (e.g., before 1955 in Israel). The method employed in this study integrates two dimensions: The vertical—estimating the changes in sea levels relative to the present MSL, based on archaeological evidence; and the horizontal—determining the coastline changes, based on coastal architectural and geomorphological structures appearing in historical maps. Both the structural and the cartographic evidence for sea level changes date to the 19th Century, and indicate a rise of 0.36 m over the last two centuries. Findings attesting to horizontal changes, indicate a gradual migration of the coastline landward, to the east since 1863 and a rapid change in the coastal geomorphology at the beginning of the 20th century. The sea level increase from the 19th century might be, in part, a consequence of regional trends and, in part, a result of a gap between method accuracy (archaeology and modern measurements). Nevertheless, the drastic change in the geomorphology of the coastline may indicate an extreme meteorological event, such as a storm at sea, accom- panied by a local rise of sea level, but further research is required to verify this.

1. Introduction Sea level measurements reflect a vertical movement of sea water in relation to the adjacent land, which can be influenced by both vertical Sea level variability that is a result of changes in sea water mass is land movements and variability in ocean volume and density (Lambeck strongly associated with global, regional and local climate changes and Purccel, 2005; Peltier, 2004). (Jevrejeva et al., 2006; Milne et al., 2009; Church and White, 2011; The majority of coastlines in the eastern Mediterranean have been Alley et al., 2005; Douglas, 2000; Fairbanks, 1989). In this context, the identified as tectonically and isostatically unstable in the last two mil- Mediterranean Sea has been intensively investigated in recent decades lenia (Lambeck et al., 2011; Antonioli et al., 2007; Ferranti et al., 2006; since it is sensitive to climatic and environmental variables (Tsimplis Pirazzoli, 2005; Anzidei et al., 2010; Morhange et al., 2006). This and Baker, 2000; Tsimplis and Josey, 2001; Klein et al., 2004; Vigo makes it complicated to monitor and assess the land's movements in et al., 2005). order to associate sea level variability with climate changes. In the

∗ Corresponding author. Kibbutzim College, Mordechai Namir 149, Tel Aviv-Yafo, Israel. E-mail addresses: [email protected] (E. Toker), [email protected] (J. Sharvit), fi[email protected] (M. Fischer), [email protected] (Y. Melzer), [email protected] (O. Potchter). https://doi.org/10.1016/j.quaint.2019.05.015 Received 11 February 2019; Received in revised form 12 May 2019; Accepted 13 May 2019 1040-6182/ © 2019 Published by Elsevier Ltd.

Please cite this article as: E. Toker, et al., Quaternary International, https://doi.org/10.1016/j.quaint.2019.05.015 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

another (Tsimplis et al., 2006; Vigo et al., 2005). Scholars relate these findings to the Eastern Mediterranean Transient (EMT) and to other environmental and meteorological factors. It is these negligible land movements and the sensitivity to climatic changes that make the Israeli Mediterranean coastline a significant lo- cation for the research of sea levels and climate changes. Sea level measurements taken in Israel continuously since 1958 show that the mean sea level (MSL) is 0.055 m above the Israel Geodetic Datum (Shirman, 2004; Toker et al., 2012). Based on this data, ac- cording to Auriemma and Solinas’ (2009) method, the present MSL along the Israeli coastline is calculated as 0.06 m (Toker et al., 2012). Estimates of past MSL rely largely on archaeological indicators along the Israeli coastline (Flemming, 1969; Sneh and Klein, 1985; Nir and Eldar, 1986; Sneh, 2000; Sivan et al., 2001, 2004; Galili et al., 2007; Galili and Nir, 1993; Nir, 1997; Toker et al., 2012). Archaeological evidence from areas of low tidal range, such as the Mediterranean Sea, can provide significant information on sea level changes; namely, coastal structures that were constructed with deliberate relation to the sea can be used as indicators of the sea level at the time of their con- struction (Lambeck et al., 2004).Archaeological evidence of past sea level (SL) dating as far back as ∼6900 BCE (∼8850 BP) (Glili and Nir, 1993) is scattered along the Israeli coastline. There are two methods of determining changes in sea level using archaeological evidence (Vunsh et al., 2017). The first based on direct evidence provided by man-made structures along the coastline, which were originally directly related to sea level at the time they were constructed and functioning (e.g. sea fortress foundations and drainage outlets) (Auriemma and Solinas, 2009; Toker et al., 2012). The second indirect evidence involves in- termediate factors, between man-made structures and sea level. For instance, the coastal wells required water table levels which lean and adjust to sea-level (Sivan et al., 2004; Marcus, 1991; Nir and Eldar, 1986). Our study focuses on direct findings, which provide evidence for sea levels over the last 200 years when the land adjacent to the sea has been relatively stable. Frequent changes of ruling powers during this period Fig. 1. Map of study area. enable dating the archaeological SL evidence to relatively short time ranges of around a century each. The two exceptions are evidence from – 1970's, these difficulties were discussed regarding the tectonic and neo- the Mamluk period (1291 1517), which was not sea oriented and left tectonic stability of the Mediterranean coastline of Israel. Discourse on little archaeological SL evidence, and from the days of the Ottoman determining stability or instability became intense and focused in the Empire, which ruled in the Holy Land for 400 years, often making it ffi fi 1990s on the coastline of Caesarea (Neev et al., 1973; Mart and di cult to date its remains to a speci c time within this span. The Perecman, 1996). However, recent studies show that for at least the last paucity of coastal evidence from these two adjacent periods forms a gap two millennia land movements along the Israeli Mediterranean coast- in SL measurements between the end of the Crusader period (1291) and line (as a result of both isostatic and tectonic movement), have been employment of modern measurements in Israel (1958). The exceptions ff negligible (Schattner et al., 2010; Gvirtzman et al., 2008; Sneh, 2000; are two wells in Ja a and Acre dating to ca.1700 (Toker et al., 2012; Sivan et al., 2001, 2004; Toker et al., 2012). It is for this reason that the Vunsh et al., 2017). Israeli coastline, located on the southeastern edge of the Mediterranean The sea level in Israel during the Crusader period was found to be − Basin (Fig. 1)reflects sea level and climate variability, with land 0.5 ± 0.20 m relative to the present MSL (Toker et al., 2012). A movement being a minimal factor (Sivan et al., 2001, 2004; Anzidei study of the Crusader period MSL in produced similar results fi ff et al., 2010; Toker et al., 2012). Moreover, although nine earthquakes (Furlani et al., 2013). The signi cant di erence between the Crusader have been documented off the eastern Mediterranean coastline since and the present-day MSL raises two questions: how was this 1837 none of them impacted the Israeli coastline (Salamon et al., 2007, 0.5 ± 0.20 m gap closed and when did this change take place? 2010). Furthermore, two waves were documented in the Le- The aim of this study is therefore to determine the course of the vant one in 1759 as a result of earthquakes in , and the second changes in MSL in the eastern Mediterranean Sea during the last two fl off the Israeli coastline in 1856 (Zohar, 2017). Neither of these tsunami centuries, as re ected in archaeological evidence scattered along the events have been shown to have altered the Israeli coastline. In addi- Israeli coastline, and the timeframe of these changes. tion, the standard continuance sea level measurements of the office of Survey of Israel examined this issue by monitoring the displacements of 2. Methodology the tide-gauges using satellite observations and concluded that the ef- fect of land movements on sea level measurements has been negligible The methodology of this study is based on an examination of two fi (Rokhlin, 2014). dimensions of change. The rst is vertical: an evaluation of MSL Previous studies found that sea level does not rise uniformly around changes over the last two centuries. The second is horizontal: an eva- the globe (Church and Wight, 2011). For instance, sea levels in the luation of the changes to the coastline location. In this study, evidence fl eastern Mediterranean change differently from the global oceanic ones, of the vertical change was used to evaluate sea level uctuations and and even the Mediterranean's basins differ in their sea levels one from evidence pertaining to horizontal shifts was used to determine the date of the MSL change over the last two centuries.

2 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

According to the Israeli law (1978), “an antiquity … refers to an artifact produced before the year 1700 CE”.1 This study focuses mostly on materials from the 19th century, the only exception being the Cae- sarea high-level , whose appearance in historical maps is used as an indicator of changes in coastline position. Though the 19th cen- tury materials are not legally defined as archaeological finds, archae- ological methods were nonetheless employed in their study. References to “archaeological indicators” throughout this paper thus relate to the method of research rather than to the structures themselves.

2.1. Estimation of the vertical change in sea level

Archaeological indicators: The estimation of sea levels in this study was based on coastal structures (drains incorporated into jetties, ram- Fig. 2. Flowchart of the study. parts and wooden piles used as foundations for sea walls), which can provide significant information concerning sea levels at the time of fi their construction because of their doubtless relation to the sea 3. Empirical ndings (Lambeck et al., 2004; Auriemma and Solinas, 2009; Toker et al., 2012; Vunsh et al., 2017). Based on Lambeck et al. (2004) and on Auriemma Following this methodology, the results are presented in two di- and Solinas (2009). This study employs the following method to esti- mensions: the vertical and the horizontal. mate the past MSL based on archaeological evidence: 3.1. Evidence of vertical changes 1. Identifying the coastal structure and determining its original func- tion. The evidence of sea level elevations in the last 200 years is based 2. Understanding the structure's relative position to the past MSL, in- upon the remains of coastal structures. cluding seasonal sea level elevations and the tidal changes at the According to the Israel Antiquities Authority (IAA, research work- time of construction, and finding the element in each structure that shop, Acre and Jaffa), there are similarities in the history and archae- can serve as an indicator (marker) of the past MSL. ology of Jaffa and Acre in the 19th century during the Ottoman period. 3. Measuring the height of the past sea level marker in relation to the In both cities, historical events led to the intensive building of off-sea Israel Geodetic Datum. fortification and marine infrastructure (IAA). An examination of these 4. Calculating the past MSL according to the following equations: structures indicates mean sea level at the time of their construction. Drainage outlets in Jaffa port: Drainage outlets must be located When the marker represents upper-bound sea level: above high tide elevation to guarantee sewage disposal (Fig. 3); thus, the bottom level of the outlet can indicate the height of the upper Past MSL = marker height - amplitude + interval boundary of the sea level at the time of its construction (Auriemma and ff When the marker represents lower-bound sea level: Solinas, 2009). In a survey undertaken for this study in Ja a port, two drainage outlets were found traversing an Ottoman jetty which building Past MSL = marker height + amplitude + interval was completed in 1888 (the building time is estimated in this study to The marker height is the height of the indicator relative to Israel 10 years, hence this measurement of past MSL has an uncertainty of 10 Geodetic Datum. The amplitude is one half of the spring tide variability years). In 1931, a British customs house was built on top of the jetty. (0.4 m) (Rosen, 2002) and the seasonal variability (0.2 m) (Goldsmith Today the sill of the jetty drain lines is covered by sea water (Fig. 3). and Gilboa, 1986) Amplitude = (spring tide variability + seasonal The bottom of one of the outlet drainage lines is located 0.16 m above variability)/2 = (0.2 + 0.4 m)/2 = 0.3 m. Therefore, the amplitude the Israel Geodetic Datum and the second is 0.23 m above it (measured along the Israeli coast is 0.3 m. The interval is the difference between by the Survey of Israel in February 2015). A calculation of the MSL at the Israel Geodetic Datum and the long-term MSL (60 years) in Israel. the time of construction of the drainage system (see above, section 2.1) − – According to Shirman (2004) and Toker et al. (2012) the interval is shows that the MSL was at least 0.2 ± 0.10 m (0.16 0.3 + [-0.06]) − – 0.06 m (Fig. 2). According to Vunch et al. (2017), after reducing the and 0.13 ± 0.10 m (0.23 0.3 + [-0.06]), respectively, relative to tidal efect, by averaging long term data, measurement of past MSL the present MSL (Fig. 3). based on wells has an uncertainty of ± 0.10m.,. We have adopted this The foundation of the southern sea wall in Acre: The southern method for our estimation of MSL uncertainty. The uncertainty of each sea wall in Acre was severely damaged during 's 1799 cam- – vertical evidence was fixed according to archaeological dating methods. paign and was reconstructed during the years 1831 1840 using a variety of buildings methods (Sharvit et al., 2013; Schaffer et al., 2014). Hence the date uncertainty was fixed to 9 years. Two of these methods, 2.2. Estimation of the horizontal change of the coastline the wooden-post foundation (tolpi), and the artificial embankment of sea fortress, can be used as markers of past sea levels (Fig. 4-6). An estimation of the horizontal change of the coastline can also It is quite communis opinio that sea water strengthens the solidifi- indicate that there was a change in sea level elevations (Galili and Nir, cation of concretions, as it was pointed out already by Vitruvius (De 1993). In this study, these estimations were based on an analysis of arch. 5.7.3, 4), at the end of the 1st century BCE. Similar evidence of an historical maps in a two-step process: (1) reviewing the maps and artificial embankment helped to determine the past sea level in the identifying changes of coastline position in these maps in relation to the cases of Caesarea and , where the tops of the embankments present-day terrain (Shtienberg et al., 2014) and (2) using archae- were at an elevation of a little over the MSL at the time of their con- ological remains as an additional indicator to measure the distance struction, and the hewn stones of the sea walls were placed above them from the coastline (Fig. 2). (Nun, 1988; Raban, 2004). Following these previous insights, the arti- ficial embankment in Acre may indicate the Ottoman period sea level. 1 Israel Ministry of Foreign Affairs (2018). http://mfa.gov.il/MFA/ Nowadays, the highest elevation of the artificial embankment is at PressRoom/1998/Pages/Antiquities%20Law-%201978.aspx. −0.3 m relative to the Israel Geodetic Datum. The first two layers of

3 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

Fig. 3. (A) Sea gate to Jaffa; two drainage outlets can be seen from the sea (1895–1905); (B, C) photos and (D) schematic chart of the drain channels in the Ottoman jetty (2016). hewn stones on top of the embankment of the sea wall in Acre were position. The outline and location of some archaeological remains ap- constructed from headers—a building method known for withstanding pear differently on different maps, reflecting horizontal change. In this intensive wave energy (Schaffer et al., 2014). The construction of paper we demonstrate these changes by examining two case studies: header's as protection against wave energy supports the assumption Caesarea and Jaffa. It is notable that texts accompanying the maps, that the past MSL was as high as the top of the embankment. Due to the which were also consulted for this study, are often more reliable than above and the correction (see section 2.1, the MSL in 1840 was the maps themselves (Biger, 1996). −0.36 ± 0.10 m (−0.3 + [-0.06]) relative to the present MSL (Fig. 6). 3.2.1. Evidence of coastline changes in Caesarea Eight maps prepared between 1799 and 1932 show the position of 3.2. Evidence of horizontal changes the coastline relative to the intersection between the Roman high-level aqueduct and the Byzantine wall (erroneously identified as a Roman Coastline migration can be deduced by comparing historical maps wall in some of the maps) This intersection point was selected as a demonstrating a change of distance between coastal structures and the reference for presenting the chronology of the coastline's horizontal coastline. This enables us to estimate the horizontal change of the latter; changes during the last 200 years as they were recorded (Fig. 7): however, maps of the lacked precision until the end of the 18th century, mainly because they were based on rare, sporadic and in- 1. Jacotin Map of Caesarea (Jacotin, 1810)(Fig. 7a). This map shows accurate measurements (Goren, 2002). It was only in the 19th century the high-level aqueduct still intact, running parallel to the coastline that a change occurred in cartographic methods, which were based on until it intersects with the Byzantine city wall in the northern part of relatively reliable and accurate measurements. For the first time, the the shore of Caesarea. The city wall is inaccurately marked on the process of mapping the Holy Land was carried out by geographers and map as the “enceinte qui renferme des ruines” ("City wall which archaeologists for research and military intelligence goals (Goren, contains ruins"), and in later maps erroneously assigned as being 2002). The first topographic maps of based on measurements from the Roman period (Vann, 1992). Although this map is based on were prepared by Colonel Pierre Jacotin, a French army surveyor, fi eld measurements, its accuracy is not satisfactory, and it is there- during Napoleon's campaign in 1799. These were followed by maps fore presented here only as evidence of the existence of an intact made by 19th century European researchers. Cartography in Palestine aqueduct and not as an indication of its distance from the coastline. was fully developed by the British Mandate that followed the First 2. Mansell's Map of Caesarea (after Mansell, 1862)(Fig. 7b): Prepared World War (1921–1948); detailed and accurate maps were made based by Commander Arthur Lukis Mansell for the British navy, presents in on a trigonometric survey that was carried out across the entire country high resolution four sites along the Israeli coastline, including Jaffa (Gavish, 2004). and the bay of Caesarea. The high-level aqueduct of Caesarea is In order to evaluate the horizontal change of the Israeli coastline presented in full, intersecting with the Byzantine wall, 65 m east of position over the last 200 years, we surveyed some of these historical the coastline. maps. In this study, we compare the distance between the coastline and 3. Guérin map of Caesarea (after Guérin, 1870)(Fig. 7c): The high-level the same structures that appear in these maps as markers of coastline aqueduct appears here, but the map is very schematic and roughly

4 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

Caesarea” (von Mülinen, 1908: 363). 6. Caesarea map of the Antiquities District (, 1921) (Fig. 7f): This is a schematic map prepared by a team of archae- ologists that were working at the site. A photo of this map that is available at the Israel Antiquities Authority's Scientific Archive did not retain the map's original dimensions and can therefore not be consulted for matters of scale. Nevertheless, the map shows both aqueducts intact and intersected by the Byzantine city wall east of the coastline. 7. Cadastral Map of Caesarea (Survey of Palestine, 1925a)(Fig. 7g): This is the first map that describes Caesarea without presenting the high aqueduct that used to run parallel to the coastline and its in- tersection point with the Byzantine wall, since the aqueduct had been ruined by the sea by the time of the map's preparation. The new bay that appears north of the city still exists. 8. Topographical Map of Caesarea (Survey of Palestine, 1932)(Fig. 7h): This map was prepared on the basis of a triangulation network and the above mentioned cadastral map (Fig. 7g). It is similar in its description to the 1925 map.

Calculating the changing distance between the coastline and the Fig. 4. Location of archaeological remains (marked in red) on Acre port map: intersection between the aqueduct and the Byzantine city wall reveals (1–1) wooden-post foundation; (2–2) artificial embankment foundation. the horizontal change of the shoreline during the last 200 years. 1. Wooden-post foundation (tolpi method): An 80m segment of the According to the graph presented in Fig. 8, it appears that, during this southern sea wall was constructed using the tolpi method (Fig. 4:1). period, the shoreline migrated 65 m eastward. Additionally, these maps Wooden posts were inserted into the seabed and a horizontal wooden frame prove that the destruction of the high aqueduct occurred between 1921 was fixed to their upper sections with iron nails (Fig. 5). The first stone layer and 1925. of the wall, consisting of large, ashlar stone blocks, was placed over the wooden frame (Schaffer et al., 2014). This section of the sea wall was de- 3.2.2. Evidence of coastline changes in Jaffa signed by the engineer Declaretto from Italy (Dichter, 2000), who brought Four maps from 1865 to 1929 reveal the horizontal changes in the the post-and-frame foundation system that was used in Venice to Acre (Goy, position of the Jaffa coastline: 2006). The upper wooden frame was found intact with no signs of rot (Fig. 5B). Since wood rots when exposed to air and water alternatively 1. Jaffa Map (Mansell, 1862)(Fig. 9a): In this map, prepared by (Blanchette, 2003), this shows that the frame was placed below the low tide Commander Arthur Lukis Mansell for the British navy, a sandbank elevation, indicating the lower boundary of sea level at its construction ff time. The elevation of the upper wooden frame was −0.84 m relative to the appears on the coastline of Ja a, south of the city wall. ff Israel Geodetic Datum. According to the calculation (see section 2.1), the 2. Plan of Ja a (Survey of , 1918)(Fig. 9b): This map was drafted lower boundary of the MSL in 1841 was 0.6 ± 0.1 m (−0.84 + 0.3 + [- during the First World War and issued as part of a pocket guide- 0.06]) (Fig. 5B). book. The shores north and south of Jaffa are wide and the sandbank can be identified. ff drawn using thick lines. The precise distance of the aqueduct from 3. City map of Ja a (Survey of Palestine, 1925b)(Fig. 9c): Despite the the coastline is therefore hard to determine; however, the written massive development of the harbor, the southern shore remained fi description of a way leading to Caesarea from the north shows wide and the sandbank can still be identi ed. ff clearly that there was, in fact, a road that separated the shore from 4. Topo-cadastral map of Ja a (Survey of Palestine, 1929)(Fig. 9d): the aqueduct: “To our right is the coastline and to our left remains of This is the latest map that describes the coastline before the con- an ancient aqueduct that once ran water from Zabarin to Caesarea. struction of the port under the British Mandate. The shores are … The arches that carried it, are now almost completely buried in narrower than they were in previous maps and the sandbank has the sand …. it was probably a unique decoration for the north road disappeared. along the coast leading to Caesarea.” (Guérin, 1875). 4. PEF map of Caesarea (after Wilson and Conder, 1890)(Fig. 7d): The Photos from 1917 indicate a seasonal change in the accumulated high-level aqueduct is intact and intersected by the Byzantine wall sand on the above mentioned beach with the sandbank, which Haddad 12.5 m east of the coastline. It is described as follows: “After passing (2013) believes represents changes that occurred in other years as well. through this ridge, which is of soft stones easily tunneled, the Though it is possible that this might also be the case in the 1929 maps, aqueduct turns due south, and runs along the shore for rather more it is less likely that a seasonal change would have been incorporated than a mile, its cross being marked by a ridge of loose sand blown into a map. The decision to relocate the coastline in the 1929 map over it and entirely hiding it. Near the north-west corner of the therefore more likely indicates a permanent change that occurred when fl Roman enceinte of Caesarea it is, however, visible, and was here the sandbank had been washed away by the sea or ooded. also examined” (Conder and Kitchner, 1881 Vol. II: 23). fi 5. Von Mülinen's Map of the Carmel (after von Mülinen, 1907)(Fig. 7e): 3.3. Summary of the ndings This map of the Carmel, which includes Caesarea, was atached to von Mülinen's 1908 Beiträge zur Kenntnis des Karmels. The high-level Finds attesting to the vertical changes (Table 1) indicate that during − aqueduct seems intact, reaching the Byzantine city wall, 6.5 m east the 19th century the lower boundary of the MSL was 0.6 m relative to − to the coastline. The connection of the aqueduct to the city wall is the present MSL, whereas the upper boundary was 0.2 m relative to described in the book as follows: “The aqueduct re-emerges west of the present MSL. Archaeological sea level evidence leans on the past fi the sandstone () ridge, covered today with piles of sand in average sea level elevation as indicated by the arti cial embankment, − fi certain places, and continuing south, along the coastline, to which shows that the MSL was 0.36 m in 1840. The ndings per- taining to horizontal changes indicate a gradual migration of the

5 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

Fig. 5. (A) schematic section (see Fig. 4: 1-1), and (B) photo of the wooden posts of the foundation of the sea wall in Acre (Fig. 4: 1-1).

2. Artificial embankment of sea fortress: Part of the was founded on an artificial embankment (Fig. 4:2) made from a concretion of pebbles, field- stones, animal remains (slaughterhouse waste and bones) and mortar (Fig. 6)(Schaffer et al., 2014).

Fig. 6. (A) Schematic section and (B) photo of artificial embankment on the foundation of the sea wall in Acre (see Fig. 4: 2-2). coastline eastward, toward the land, since 1863 and a rapid change in 4. Discussion the coastal geomorphology in the beginning of the 20th century (see Table 2). There is a gap in the archaeological evidence of sea level elevations Fig. 10 displays the sea level estimation between 1750 and 1900 along the Israeli coastline over the last 700 years between the end of the deduced from previous and current research and Israel Geodetic Datum, Crusader period (1291) and 1958 when analogical measurements were based on the Survey of Palestine under the British Mandate and on the first employed in Israel (Shirman, 2004). Since 1291 sea level in Israel measurements taken along the Israeli Mediterranean coastline in the has risen by 0.50 ± 0.20 m (Toker et al., 2012). A similar rise of sea Survey of Israel in 1921–1995 (Toker et al., 2019). In addition, esti- level during that time frame was found in Malta (Furlani et al., 2013). mated sea levels from wells in Acre and Jaffa based on Vunsh et al. Consequently, the aims of this study were to determine the timetable of (2017) were added on the this graph, in order to strengthen our find- the changes in the sea level elevation of the Levantine Basin during the ings. last 200 years as demonstrated by the Israeli coastline. Based on ar- chaeological and geomorphological evidence from Acre, Caesarea and Jaffa and on sea level measurements from Gaza taken by the British Survey of Palestine during 1921–1922, this study suggests that over the

6 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

created a new bay (Nir, 1981). Nir (1981) suggested that the high-level aqueduct was supported by a sea-wall which prevented the erosion of sand along the north beach. This sand erosion/could be expected as the result of massive building in the sea (Herod harbor), south of the high level aqueduct by Herod. According to Nir (1981) during the Early Is- lamic conquest in the 7th century the supporting sea-wall fell into ruin, leaving the north shore of Caesarea, without defense from erosion. Thus according to Nir (1981) erosion of the coast caused the destruction of the high-level aqueduct. Others have asserted that the high level aqueduct fell into ruin due to an earth quake (Galili et al., 2011)or natural erosion (Reifenberg, 1950–1951; Karni, 1982). But archae- ological research of the water channels on top of the high-level aque- duct revealed that the highest of these three water channels was built in the Crusader period (1101–1291) (Olamy and Peleg, 1977; Peleg, 1989). Hence the high-level aqueduct would have to have been de- stroyed after the Crusader period in Caesarea, during the Mameluk conquest in the 13th century (, 1964; Olamy and Peleg, 1977; Peleg, 1989). However, the high-level aqueduct of Caesarea appears in all the topographic maps of the second half of the 19th century and the beginning of the 20th century, contradicting all proposals of an earlier demise. Moreover, the high-level aqueduct was witnessed and de- scribed by Guérin (1870),byConder and Kitchner (1881) and by von Mülinen (1908). The coastline shift in Caesarea between 1863 and 1927 was pre- Fig. 7. Historical Maps of Caesarea: (a) Jacotin's map of Caesarea sented by Shtienberg et al. (2014), who claimed that it had eroded by (Jacotin,1810); (b) Mansell's Map of Caesarea; (Mansell, 1862); (c) Guérin's 10–65 m. Shtienberg et al. (2014) related this erosion to natural pro- map of Caesarea (Guérin, 1870; (d) PEF map of Caesarea (Wilson and Conder, cesses, which were amplified by quarrying of beach sand and of coastal 1890); (e) von Mülinen's Map of the Carmel (von Mülinen, 1907); (f) Caesarea, Map of the Antiquities District (Survey of Palestine, 1921); (g) Cadastral map of kurkar for building the Bosnian colony in 1878, while the high-level Caesarea (Survey of Palestine, 1925a); (h) Topographical Map of Caesarea aqueduct was damaged by scavenging material for re-use. According to (Survey of Palestine, 1932). Zviely and Klein (2002), the stabilization of coastal geomorphological processes due to anthropogenic interference takes only a short period of time (5–8 years), and since the colony was built in 1878, while von Mülinen's map (1907) shows an intact aqueduct, it is doubtful that the Bosnian colony interfered with the coastal geomorphology. The last appearance of the high-level aqueduct of Caesarea as an intact feature is in a map dating to 1921 (Survey of Palestine, 1921). The map of Survey of Palestine (1925a) shows only a bay indenting the coast. It seems that the aqueduct was destroyed between these years. Since coastal structures are sensitive to sea level amplitude and the tides in the Mediterranean are generally small (Androulidakis et al., 2015), damage to coastal structures is more likely to be associated with extreme meteorological events that cause the sea level to rise (Pugh and Woodworth, 2014). Hence, a high sea level elevation combined with an extreme high sea level event might bring about a geomorphological change to the coastline. A report in Doar Hayom a newspaper published in Hebrew in the 1920s, offers evidence for extreme winter events. On Fig. 8. The change in distance between the aqueduct and the coastline of Caesarea. February 5, 1924 it was reported that extreme weather caused inland flooding and broke a bridge over the , north of Tel Aviv. This weather event fits the time of the geomorphological change that last two centuries sea level elevation rose by 0.36 m. was identified in this study and it may have been what damaged and So far, no study has addressed this issue. However, previous studies destroyed the high-level aqueduct of Caesarea (Fig. 11). have dealt with some of the evidence presented in this study, yet they Additionally, the fact that coastline shifts occurred in Jaffa during employed different approaches to examining the evidence and came to the same period, when the British authority banned building projects different conclusions. For example, the cause and time of the destruc- along the coastline until 1932 (Gilor, 2016), strengthens the assump- tion of the high-level aqueduct in Caesarea are disputed (Nir, 1981; tion that the changes might have happened on a regional scale rather Reifenberg, 1950–1951; Karni, 1982; Olamy and Peleg, 1977; Peleg, than being local-anthropogenic ones in Caesarea alone (Zviely et al., 1989; Porath and Siegelmann, 2002; Negev, 1964; Shtienberg et al., 2007). This is more likely to have been related to a larger-scale phe- 2014). The aqueduct which is now partially in ruin has served as an nomenon like sea level rise. indicator of coastal changes in several studies (Galili et al., 2011; Therefore, it seems likely that horizontal changes of the coastline Shtienberg et al., 2014). Our study claims that the high-level aqueduct observed in historical maps of Caesarea and Jaffa were a result of an was destroyed at the beginning of the 20th century due to sea level rise. extreme high sea event, which triggered geomorphological processes. However, there are studies that argue that the cause of its destruction These processes of extreme weather events accompanied by sea level was anthropogenic interference in the coastal environment, while rise might have created the coastline erosion between 1921 and 1929. others assign this change to environmental factors other than a rise in The conclusions of this study are based on direct and indirect evi- sea level. Some of these claimed that this happened after the con- dence: the direct evidence is based on the unequivocal relation of struction of the pier of Herod's harbor in the 1st century BCE, which coastal structures to changing sea levels. Wells constitute indirect

7 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

Fig. 9. Historical maps of Jaffa: (a) Jaffa Map (Mansell, 1862); (b) Plan of Jaffa(Survey of Egypt, 1918); (c) City map of Jaffa(Survey of Palestine, 1925b); (d) Topo- cadastral map of Jaffa(Survey of Palestine, 1929).

Table 1 Findings attesting to vertical changes.

SL Evidence (m) Site Evidence Method Date

−0.36 Acre Artificial embankment Archaeological indicator 1830 −0.6 lower boundary Acre Wooden-post foundation Archaeological indicator 1830 −0.2 upper boundary Jaffa Drainage outlets Archaeological indicator 1888

Table 2 Findings attesting to horizontal changes.

Distance of aqueduct from coastline (m) Reference Site Method Date

65 Mansell's map; (Mansell, 1862) Caesarea Map analysis 1863 12.5 PEF map (Wilson and Conder, 1890) Caesarea Map analysis 1878 6.5 von Mülinen’ s map (von Mülinen, 1907) Caesarea Map analysis 1907 0 Cadastral map (Survey of Palestine, 1925a) Caesarea Map analysis 1925 0 Topographical map (Survey of Palestine, 1932) Caesarea Map analysis 1932

Fig. 11. A weather report from Doar Hayom daily newspaper; February 5, 1924 Fig. 10. The sea level over the last 250 years, deduced from previous and (Hebrew). current observations along the Israeli Mediterranean coastline. research of Vunsh et al. (2017) are in line with the upper boundary evidence as their function relies on the water table, which is affected by established via the direct evidence in the current study, and therefore fi the level of the adjacent sea water (Vunsh et al., 2017). According to reinforce our ndings. Fig. 10, the sea levels that are reflected by the wells presented in the In order to explain the 0.36 m rise in sea level elevation during the

8 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

5. Conclusion

This study found an elevation in sea level and a narrowing of the coastal strip along the Israeli coastline over the last two centuries. Evidence of a rapid rise of sea level was obtained using two sources: structural archaeological remains and historical maps. Both sources point to the same years when changes in sea level occurred. The sea level rise determined for the 19th century may be in part a consequence of regional trends and in part the result of a gap between the accuracy provided by the different data sets (archaeological remains vs. modern measurements). It is also important to point out that the notable change in the geomorphology of the coastline may indicate an extreme me- teorological event, such as a storm surge, which resulted in a local rise of sea level. Further research is required to verify this.

Fig. 12. Sea level variability over the last 165 years according to multi-method Acknowledgements measurements. The authors wish to thank Yossi Melzer, Meir Malin, and Ilia last 200 years it is necessary to distinguish between variability and Perelman from the Survey of Israel (SOI), for the measuring the heights long-term trends of rising sea levels. Sea level variability is affected by of the archaeological evidence. This research was funded through the local and regional meteorological factors (such as storm surges) and Smaller-Winnikow Scholarship Fund in cooperation with the Keren modes of climate variability (such as the North Atlantic Oscillation), as Keyemet L’Israel (—JNF) 2015, and in addition, well as anthropogenic effects (such as the construction of the Aswan generous support and cooperation were provided by The Porter School Dam). Long-term trends are brought about by global climate factors of Environmental Studies 2014-2019. (Church and Wight, 2011). Fig. 12 shows sea level elevations between 1760 and 2017. The elevations from 1760 to 1900 are based on ar- References chaeological evidence; elevations between 1921 and 1929 are derived from mechanical sea level measurement; and since 1955 they are de- Gilor, S., 2016. Jaffa Port at the end of the Ottoman period and the beginning of the termined by continuous tide gauge measurements. The analysis of sea British Mandat period. Merhavim 7, 189–220. Reifenberg, A., 1950. Caesarea: a study in the decline of a town. Isr. Explor. J. 1 (1), level elevations since mechanical tide gauge measurements were 20–32. available (1921) shows a variability of ± 0.17 m. Based on archae- Alley, R.B., Clark, P.U., Huybrechts, P., Joughin, I., 2005. Ice-sheet and sea-level changes. ologically analyzed evidence, this study shows that between the years Science 310 (5747), 456–460. Androulidakis, Y.S., Kombiadou, K.D., Makris, C.V., Baltikas, V.N., Krestenitis, Y.N., 1760 and 1900 the sea level along the Israeli coastline was 0.36 below 2015. Storm surges in the Mediterranean Sea: variability and trends under future the present MSL with an uncertainty of −0.26 (based on the lower climatic conditions. Dyn. Atmos. Oceans 71, 56–82. boundary deduced from the wooden piles) and +0.16 m (based on the Antonioli, F., Anzidei, M., Lambeck, K., Auriemma, R., Gaddi, D., Furlani, S., Orrù, P., č ć upper boundary deduced from drainage outlets). The range of values Solinas, E., Gaspari, A., Karinja, S., Kova i , V., 2007. Sea-level change during the Holocene in Sardinia and in the northeastern Adriatic (central Mediterranean Sea) for sea levels is larger than could be explained by periodical changes from archaeological and geomorphological data. Quat. Sci. Rev. 26 (19–21), caused by local and regional meteorological factors that have been 2463–2486. measured on the Israeli coastline throughout the 20th century. A si- Anzidei, M., Antonioli, F., Benini, A., Lambeck, K., Sivan, D., Serpelloni, E., Stocchi, P., fi 2010. Sea Level change and vertical land movements since the last two millennia milar sea level change was not identi ed in the long-term measure- along the coasts of southwestern and Israel. Quat. Int. 232, 13–20. ments in Marseille (representing the western Mediterranean Basin); Auriemma, R., Solinas, E., 2009. Archaeological remains as sea level change markers: a review. Quat. Int. 206, 134–146. hence, the data gap between the 19th century and 1920s is probably a – ff Biger, G., 1996. Judicial viewpoint concerning boundary maps. Horiz. Geogr. 44 45, consequence of the discrepancy between the accuracy a orded by ar- 91–96. chaeological data and that of instrumental measurements. Nevertheless, Blanchette, R.A., 2003. Deterioration in historic and archaeological woods from terres- drainage outlets were found to be a reliable indicator for ancient sea trial sites. In: Koestler, R.J., Koestler, V.H., Charola, A.E., Nieto-Fernandez, F.E. (Eds.), Art, Biology, and Conservation: Biodeterioration of Works of Art. The levels (Toker et al., 2012). It can therefore be stated that between Metropolitan Museum of Art, New York, pp. 328–347. 1888—when the drainage outlet was built—and the 1920s, when the Bonaduce, A., Pinardi, N., Oddo, P., Spada, G., Larnicol, G., 2016. Sea-level variability in sea level was first measured, the latter rose by at least 0.2 m (Toker the Mediterranean Sea from altimetry and tide gauges. Clim. Dyn. 47 (9–10), 2851–2866. et al., 2019). Church, J.A., White, N.J., 2011. Sea-level rise from the late 19th to the early 21st century. A comparison of the trend of rising sea levels according to the Surv. Geophys. 32 (4–5), 585–602. findings in this study to global and Mediterranean trends of rising sea Conder, C.R., Kitchener, E.H.H.K., 1881. The Survey of Western Palestine: Memoirs of the level for the last few centuries shows the following: the rises in Israeli Topography, Orography, Hydrography, and Archaeology (No. 2). Committee of the Palestine Exploration Fund, London. sea level is 2.7 mm/year which is slightly higher than the entire Dichter, B., 2000. Akko: sites from the Turkish period. In: Carmel, A., Baumwoll, Z. (Eds.), Mediterranean trend which is 2.4 mm/year (Bonaduce et al., 2016), and Gottlieb Schumacher Institute for Research of the Christian Activities in 19th Century slightly lower than the global trend, 3.2 mm/year (Quartly et al., 2017). Palestine. University of , Haifa. Douglas, B.C., 2000. An introduction to sea-level. In: Douglas, B.C., Kearney, M.F., Hence it can be concluded that the sea level rise along the Israeli Leatherman, S.P. (Eds.), Sea-level Rise History and Consequences. Academic Press, coastline may have been caused by global climatic factors that were New York, pp. 1–11. fl more pronounced in the eastern Mediterranean Basin relative to the Fairbanks, R.G., 1989. A 17,000-year glacio-eustatic sea level record: in uence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342, entire Mediterranean Sea. As for the coastline migration found in 637–642. Caesarea and Jaffa, the presumed extreme weather event, and the re- Ferranti, L., Antonioli, F., Mauz, B., Amorosi, A., Dai Pra, G., Mastronuzzi, G., Verrubbi, lationship to the rise in sea level elevation might suggest a more V., 2006. Markers of the last interglacial sea-level high stand along the coast of Italy: tectonic implications. Quat. Int. 145, 30–54. complex explanation; however, further research is required on this Flemming, N.C., 1969. Archaeological evidence for eustatic change of sea level and earth point. movements in the western Mediterranean during the last 2,000 years. Geol. Soc. Am. Spec. Pap. 109, 1–98. Furlani, S., Antonioli, F., Biolchi, S., Gambin, T., Gauci, R., Presti, V.L., Anzidei, M., Devoto, S., Palombo, M., Sulli, A., 2013. Holocene sea level change in Malta. Quat.

9 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

Int. 288, 146–157. Pugh, D., Woodworth, P., 2014. Sea-level Science: Understanding Tides, Surges, Galili, E., Nir, Y., 1993. The submerged Pre-Pottery Neolithic water well of Atlit-Yam, and Mean Sea-Level Changes. Cambridge University, Cambridge, UK. northern Israel, and its palaeoenvironmental implications. Holocene 3, 265–270. Quartly, G.D., Legeais, J.F., Ablain, M., Zawadzki, L., Fernandes, M.J., Rudenko, S., Galili, E., Zviely, D., Ronen, A., Mienis, H.K., 2007. Beach deposits of MIS 5e high sea Carrère, L., García, P.N., Cipollini, P., Andersen, O.B., Poisson, J.C., 2017. A new stand as indicators for tectonic stability of the Carmel coastal plain, Israel. Quat. Sci. phase in the production of quality-controlled sea level data. Earth Syst. Sci. Data 9 Rev. 26 (19), 2544–2557. (2), 557–572. Galili, E., Zviely, D., Salomon, A., Rosen, B., 2011. Harbor and anchorage installations in Raban, A., 2004. The history of Caesarea's harbors (Hebrow). Qadmoniot 37, 2–22. Caesarea and the destruction of the Roman harbor: a re-evaluation based on finds Reifenberg, A., 1950–51. Caesarea: a study in the decline of a town. Isr. Explor. J. 1, from underwater and coastal research. In: Izdarekhet, A., Ayalon, E. (Eds.), Treasures 20–32. of Caesarea: Summary and Studies of Caesarea and its Surroundings, in Memory of Rokhlin, E., 2014. Monitoring Vertical Displacements of TGBM Using GNSS Amnon Kidron. ACFA ancient Caesarea, Jerusalem, pp. 41–66. Measurements and Connected Them to the Permanent GNSS Stations in Israel (MSc Gavish, D., 2004. The Survey of Palestine under the British Mandate, 1920–1948 Dissertation). Technion, Haifa, Israel. Routledge Curzon, London. Rosen, D.S., 2002. A review of sea level monitoring status in Israel. In: Rosen, D.S., Gilor, S., 2016. Jaffa port in the late Ottoman period and the British Mandate. Merchavim Thorkild, A. (Eds.), Intergovernmental Oceanographic Commission & International 7, 189–222 (in Hebrew). Commission for the Scientific Exploration of the Mediterranean Sea. MedGLOSS Pilot Goldsmith, V., Gilboa, M., 1986. Analysis of Israeli tidal data. Horiz. Geogr. 15, 21–46. Network Workshop and Coordination Meeting, Israel Oceanographic & Limnological Goren, H., 2002. Sacred, but not surveyed: nineteenth‐century surveys of Palestine. Imago Research, IOC Workshop Report 176, pp. 49–62 Annex Ivj pp. Mundi 54 (1), 87–110. Salamon, A., Rockwell, T., Ward, S.N., Guidoboni, E., Comastri, A., 2007. Tsunami hazard Goy, J.R., 2006. Building Renaissance Venice: Patrons, Architects, and Builders, C. evaluation of the eastern Mediterranean: historical analysis and selected modeling. 1430–1500. Yale University, New Haven. Bull. Seismol. Soc. Am. 97 (3), 705–724. Guérin, M.V., 1875. Description, géographique, historique et archéologique, de la Salamon, A., Rockwell, T., Guidoboni, E., Comastri, A., 2011. A critical evaluation of Palestine. (Ben Amram H., Trans.). Yad Ben Zvi, Jerusalem. tsunami records reported for the Levant Coast from the second millennium BCE to the Guérin, V., Guérin, V., 1870. map of the holy land, map 1:10,000 m). Description present. Isr. J. Earth Sci. 58, 327–354. géographique, historique et archéologique de la Palestine: accompagnée de cartes Schaffer, Y., Sharvit, Y., Planer, D., 2014. Structural and conservation solutions for the détaillées, vol. V: Samarie Oriental Press, Paris. treatment of the sea front walls of old Akko (Acre), Israel. In: Peña, F., Chávez, M. Gvirtzman, Z., Zilberman, E., Folkman, Y., 2008. Reactivation of the Levant passive (Eds.), SAHC2014: 9th International Conference on Structural Analysis of Historical margin during the late Tertiary and formation of the Jaffa basin offshore central Constructions. Mexico City, pp. 1–23. Israel. J. Geol. Soc. 165, 563. Schattner, U., Lazar, M., Tibor, G., Ben-Avraham, Z., Makovsky, Y., 2010. Filling up the Haddad, E., 2013. The Ascendance and Decline of the Jaffa Sea Port from the Time of the shelf—a sedimentary response to the last post-glacial sea rise. Mar. Geol. 278 (1), Crusaders Period to the Beginning of the First World War. Ph.D. thesis. Haifa 165–176. University, Haifa, Israel. Sharvit, J., Planer, D., Buxton, B., 2013. Preliminary findings from archaeological ex- IAA Research Workshop, Acre and Jaffa (in Hebrew): http://www.antiquities.org.il/ cavations along the foot of the southern seawall of Akko, 2008–2012. Michmanim 24, article_heb.aspx?sec_id=17&sub_subj_id=621. 39–40. Jacotin, P., 1810. Carte topographique de l'Egypte, map. 1:100 000, Sheet 45: Ce'sare'e. Shirman, B., 2004. East Mediterranean sea level changes over the period 1958–2001. Isr. Panckoucke, Chevalier de la Legion d'Honneur. Atlas géographique. Eran Laor J. Earth Sci. 53, 1–12. Cartographic Collection, the Hebrew University of Jerusalem, Paris. Shtienberg, G., Zviely, D., Sivan, D., Lazar, M., 2014. Two centuries of coastal change at Jevrejeva, S., Grinsted, A., Moore, J.C., Holgate, S., 2006. Nonlinear trends and multiyear Caesarea, Israel: natural processes vs. human intervention. Geo Mar. Lett. 34 (4), cycles in sea level records. J. Geophys. Res.: Oceans 111 (C9), 1978–2012. 1–15. Karni, H., 1982. A comment on the destruction of the aqueduct in Caesarea. Cathedra 25, Sivan, D., Wdowinski, S., Lambeck, K., Galili, E., Raban, A., 2001. Holocene sea-level 172–175 (in Hebrew). changes along the Mediterranean Coast of Israel, based on archaeological observa- Klein, M., Lichter, M., Zviely, D., 2004. Recent sea-level changes along Israeli and tions and numerical model. Palaeogeogr. Palaeoclimatol. Palaeoecol. 167, 101–118. Mediterranean coasts. Contemp.Isr.Geogr. 61–62, 167–176. Sivan, D., Lambeck, K., Toueg, R., Raban, A., Porath, Y., Shirman, B., 2004. Ancient Lambeck, K., Purcell, A., 2005. Sea-level change in the Mediterranean Sea since the LGM: coastal wells of , Israel: an indicator for relative sea level changes model predictions for tectonically stable areas. Quat. Sci. Rev. 24, 1969–1988. during the last 2000 years. Earth Planet. Sci. Lett. 222, 315–330. Lambeck, K., Anzidei, M., Antonioli, F., Benini, A., Esposito, A., 2004. Sea level in Roman Sneh, A., 2000. Faulting in the coastal plain of Israel during the late Quaternary, reex- time in the central Mediterranean and implications for recent change. Earth Planet. amined. Isr. J. Earth Sci. 49, 21–29. Sci. Lett. 224 (3), 563–575. Sneh, Y., Klein, M., 1985. Holocene sea level changes of the coast of Dor, Southeast Lambeck, K., Antonioli, F., Anzidei, M., Ferranti, L., Leoni, G., Scicchitano, G., Silenzi, S., Mediterranean. Science 226, 831–832. 2011. Sea level change along the Italian coast during the Holocene and projections Survey of Egypt, 1918. Plan of Jaffa, 1:6,000, Vol. 1. 34 V 1245. Palestine Pocket Guide- for the future. Quat. Int. 232 (1), 250–257. Books. The Palestine News, Cairo. Mansell, R.N., 1862. Mediterranean: Coast of , map, 1:9260, Sheet 3: Ras en-Nakara Survey of Palestine, 1921. Caesarea, Antiquities District, Map, 1:20,000. The Israel to El Arish. Admiralty, London. Antiquities Authority, the Scientific Archive 1919–1948. Caesarea I, SRF_42(217/ Marcus, E.S., 1991. Tel Nami: a Study of a Middle Bronze IIa Period Coastal Settlement. 217). http://www.iaa-archives.org.il/zoom/zoom.aspx?id=28965&folder_id=276 Master Thesis,. Haifa University, Haifa. &type_id=&loc_id=9. Mart, Y., Perecman, I., 1996. Neotectonic activity in Caesarea, the mediterranean coast of Survey of Palestine, 1925a. Caesarea Cadastral Map, 1:5000. Survey of Palestine, Tel Central Israel. Tectonophysics 254, 139–153. Aviv. Survey of Israel Maps Archive. Milne, G.A., Gehrels, W.R., Hughes, C.W., Tamisiea, M.E., 2009. Identifying the causes of Survey of Palestine, 1925b. City Map of Jaffa, 1:7,500. Survey of Palestine, Tel Aviv. sea-level change. Nat. Geosci. 2, 471–478. Survey of Israel Maps Archive. Morhange, C., Pirazzoli, P.A., Marriner, N., Montaggioni, L.F., Nammour, T., 2006. Late Survey of Palestine, 1929. Topo-cadastral Map of Jaffa, 1:20,000. Survey of Palestine, Tel Holocene relative sea-level changes in Lebanon, eastern Mediterranean. Mar. Geol. Aviv. Survey of Israel. 230 (1), 99–114. Survey of Palestine, 1932. Caesarea Topographical Map, 1:20,000: Sheets 14–21. Survey Neev, D., Bakler, N., Moshkovitz, S., Kaufman, A., Magaritz, M., Gofna, R., 1973. Recent of Palestine, Tel Aviv. Geography Department, . faulting along the Mediterranean coast of Israel. Nature 245, 254–256. Toker, E., Landman, H., Potchter, O., 2019. Who Are You Calling a Zero? - Sea Level Negev, A., 1964. The high level aqueduct at Caesarea. Isr. Explor. J. 14, 237–249. Measurements in Israel from the British Mandate Period through the Present Day in Nir, Y., 1981. The destruction of the Roman high level aqueduct as a result of the building Reletion to the National Orthometric Heights Network. Cathedra: For the History of of the Herodian harbour of Caesarea. Cathedra 18, 50–56 (in Hebrew). Eretz Israel and Its accepted, (accepted) (in Hebrew). Nir, Y., 1997. Middle and late Holocene sea-levels along the Israel Mediterranean coast: Toker, E., Sivan, D., Stern, E., Shirman, B., Tsimplis, M., Spada, G., 2012. Evidence for evidence from ancient water wells. J. Quat. Sci. 12, 143–151. centennial scale sea level variability during the Medieval Climate Optimum (Crusader Nir, Y., Eldar, I., 1986. Ancient ground-water table in old wells as an indicator of paleo period) in Israel, eastern Mediterranean. Earth Planet. Sci. Lett. 315, 51–61. sea-levels and neotectonic changes along the coastal plain of Israel. Geol. Surv. Israel Tsimplis, M.N., Baker, T.F., 2000. Sea level drop in the Mediterranean Sea: an indicator of Rep. GSI/34/86, 28 (in Hebrew). deep water salinity and temperature changes? Geophys. Res. Lett. 27 (12), Nun, M., 1988. Tiberias: city of The sea of . In: Hirschfeld, Y. (Ed.), Tiberias: from 1731–1734. its Foundation to the Muslim Conquest. Yad Ben Zvi, Jerusalem, pp. 140–150 (in Tsimplis, M.N., Josey, S.A., 2001. Forcing of the Mediterranean Sea by atmospheric os- Hebrew). cillations over the north atlantic. Geophys. Res. Lett. 28 (5), 803–806. Olami, Y., Peleg, Y., 1977. The water supply system of Caesarea Maritima. Isr. Explor. J. Tsimplis, M.N., Zervalis, V., Josey, S.A., Peneva, E.I., Struglia, M.V., Stanev, E.V., 27, 127–137. Theocharis, A., Lionello, P., Malanotte-RizzoliI, P., Artale, V., Tragou, E., Oguuz, T., Peleg, J., 1989. The water system of Caesarea. In: In: Amit, D., Hirschfeld, Y., Partich, J. 2006. Changes in the oceanography of the Mediterranean Sea and their link to cli- (Eds.), The Aqueducts of Ancient Palestine, vol. 9. ESI, pp. 130–131 115–122. mate variability. In: Lionello, P., Malanotte-Rizzoli, P., Boscolo, R. (Eds.), Peltier, W.R., 2004. Global glacial isostasy and the surface of the Ice-Age Earth: the ICE- Mediterranean Climate Variability: Developments in Earth and Environmental 5G (VM2) model and GRACE. Annu. Rev. Earth Planet Sci. 32, 111–149. Sciences 4. Elsavier, London, pp. 227–282. Pirazzoli, P.A., 2005. A review of possible eustatic, isostatic and tectonic contributions in Vann, R.L., 1992. Early travelers and the first archaeologists. In: Vann, R.L. (Ed.), eight late-Holocene relative sea-level histories from the Mediterranean area. Quat. Caesarea Papers: Straton's Tower, Herod's Harbour and Roman and Byzantine Sci. Rev. 24 (18), 1989–2001. Caesarea (JRA5). Ann Arbor. University of Michigan, Michigan, pp. 275–290. Porath, Y., Siegelmann, A., 2002. The water-supply to Caesarea: a re-assessment. In: Amit, Vigo, I., Garcia, D., Chao, B.F., 2005. Change of sea level trend in the mediterranean and D., Hirschfeld, Y., Pitrich, J. (Eds.), The Aqueducts of Israel. Journal of Roman black seas. J. Mar. Res. 63, 1085–1100. Archaeology, Portsmouth. Von Mülinen, E., von Kartenskizze der Beiträge zur Kenntnis des Karmels von Dr. E.Graf

10 E. Toker, et al. Quaternary International xxx (xxxx) xxx–xxx

von Mülinen, 1:63,360, 1907. Geograph Anstalt von Wagner; Debes, Leipzig. Eran political and morphological aspects. Horiz. Geogr. 54, 68–77. Laor Cartographic Collection, the Hebrew University of Jerusalem. Zviely, D., Kit, E., Klein, M., 2007. Longshore sand transport estimates along the Von Mulinen, E.G., Geva-Klinberger, A., Ben-Artzi, Y. Trans, 1908. of Graf Mediterranean coast of Israel in the Holocene. Mar. Geol. 238, 61–73. http://www. Von Mulinen. Magnes, Jerusalem. soi.gov.il/page.php?id=sea_level/objectives. Vunsh, R., Tal, O., Yechieli, Y., Dean, S., Levanon, E., Sivan, D., 2017. Evaluating ancient coastal wells as sea‐level indicators from the coast of Israel. Geoarchaeology 33 (4). https://doi.org/10.1002/gea.21663. Further reading Wilson, W.C., Conder, C.R., 1890. The Palestine Exploration Fund Map, 1:63,360, Sheet 7. Stanford's Geographical Establishment, London. Eran Laor Cartographic Collection, Negev, A., 1966. The Caesarea, Mada 11, 136–144 (in Hebrew). the Hebrew University of Jerusalem. Peltier, W.R., 2002. On eustatic sea level history: last Glacial Maximum to Holocene. Zohar, M., 2017. A newly discovered historical earthquake or merely a chronological Quat. Sci. Rev. 21, 377–396. mistake? Report of Mary Eliza Rogers from October 10, 1856. J. Seismol. 21 (5), Sivan, D., Schattner, U., Morhange, C., Boaretto, E., 2010. What can a sessile mollusk 1285–1290. about neotectonics? Earth Planet. Sci. Lett. 296, 451–458. Zviely, D., Klein, M., 2002. The environmental impact of the Gaza coastal construction:

11