Evidence for centennial scale sea level variability during the Medieval Climate Optimum (Crusader Period) in Israel, eastern Mediterranean
Introduction
Global mean sea level has risen by about 18 cm over the past century (IPCC, 2007). The two major contributing mechanisms for such a sea-level rise are thermosteric expansion and mass addition to the oceans by ice-sheet melting (Jevrejeva et al., 2006, Milne et al., 2009). However, the spatial distribution of these factors is not uniform, and additional factors, including salinity changes, local and regional land movements and oceanic response, and changes in meteorological forcing, all contribute to local sea-level change. The coasts of Israel are located within the Mediterranean Sea, a semi-enclosed basin linked through the Strait of Gibraltar with the Atlantic Ocean. In general, observed sea level change along a coast can be due to land movements, to changes in the local water mass, or in the water volume.
The presently available observational records that can give us some indication of natural variability over such periods of time usually span less than 150 years, and in the Mediterranean tide-gauge records have been available only since 1875 (Lichter et al., 2010, Raicich, 2007).
By comparison, sea level has risen and fallen by more than 100 m between glacial maxima and minima, and most studies of past sea-level changes provide information on rates of which largely depend on the part of the glaciation–deglaciation cycle in which they occur. Rates of change during deglaciation periods have reached 5 m per century (Fairbanks, 1989, Blanchon and Shaw, 1995). Even within interglacial periods, rates in excess of 1 m per century have been found (Rohling et al., 2008). For the past 2500 yr, after taking into account global isostatic adjustment (GIA) and local vertical land movements, in most cases the established view is that sea level has remained unchanged.
The exact dating of the diminishing rates of sea-level change varies from place to place, and is not synchronous across the globe (Gehrels, 2010). The small rates of change that have dominated the last few thousand years make the identification of the corresponding forcing mechanisms more difficult. Tectonic and isostatic vertical movements, thermal expansion, and other steric effects, such as salinity changes, the effect of gravitation on local levels, and eustasy in terms of ice-equivalent volumes, may all be involved, but are difficult to separate.
For the Mediterranean coast of Israel (Fig. 1a and b) the Holocene data before 2000 yr ago are clustered in a few periods of time, and therefore cannot be used as the basis of exploring variability at centennial time scales (Sivan et al., 2001). For the last few thousand years, previous research indicates that sea level reached its present-day elevation about 3600 yr ago (Porat et al., 2008), fell to about − 1.5 m 2400 yr ago, and reached its present level once again about 2000 yr ago (Anzidei et al., 2010, Sivan et al., 2004). Sea level estimates based on archaeological indicators suggest that during the Roman period (Table 1) sea level in the studied area, as well as in other parts of the Mediterranean, was not significantly different from what it was at the beginning of the observational period, (i.e. around 1900) (Anzidei et al., 2010, Lambeck et al., 2004a). Sivan et al. (2004) suggest, on the basis of coastal well observations, a possible high-stand of less than 0.5 m between 300 and 600, followed by decreasing levels that had already started to fall in the 9th century, and reached a minimum during the 13th century. These oscillations are based on 34 wells from the Early Moslem period (7th to the end of the 11th centuries) that were excavated in Caesarea (Sivan et al., 2004), most indicating a sea level of − 0.24 m in the 9th century (Table 2). Two Dendropoma samples (Sivan et al., 2010), dated to the Moslem period also indicate levels of − 0.12 m and − 0.13 m (Table 2 and Fig. 2). The Crusader sea levels were obtained from the coastal wells in Caesarea, which indicate − 0.36 to − 0.62 m (Sivan et al., 2004), and from one Dendropoma sample from Atlit dated to 1090–1300 (Table 2), which suggests a mean sea level of − 0.28 m for the period (Sivan et al., 2010). Data from the following periods are scarce and not continuous (Toker et al., 2010). Model predictions for the last 2000 yr have been proposed by Sivan et al., 2004, Anzidei et al., 2010.
The GIA-induced rate of sea level variation in Israel was found in previous research to be small during the last 2000 yr because the hydro- and the glacio-hydrostatic effects were found to be of approximately equal magnitude, but of opposite sign (Sivan et al., 2004) and close to zero in the eastern Mediterranean at present (Lambeck and Purcell, 2005, Stocchi and Spada, 2009). In addition, the study area has been found to be tectonically stable, not only for the last 2000 yr (Sivan et al., 2004), but also over longer periods (Galili et al., 2007, Galili and Sharvit, 1998, Sivan et al., 1999, Sivan et al., 2001, Sivan and Galili, 1999, Sneh, 2000). This conclusion of stability is based on both geological and archaeological records, strengthened by the comparison between the observations and the model predictions made for the Holocene (Sivan et al., 2001) and for the last 2000 yr (Sivan et al., 2004). Thus, sea level estimates from the region mainly reflect the other forcing parameters.
The tectonic stability of the area and the small Global Isostatic Adjustment (GIA) contribution to relative sea level (RSL) in recent periods provide the background against which new data are combined with published data. The new data provide support for a centennial variation of sea level during the last 2000 yr, with an emphasis on the Crusader period, dated to 1099–1291 (Table 1). A possible explanation for the observed sea level change is offered, based on published information on the prevailing climate over this period.
Section snippets
Data and methodology
Two types of sea level indicators are used in this study. Firstly, estimates had been obtained from coastal structures built above the highest mean sea levels, including drainage outlets of sewage channels and gutters, levelled floors, and rock-cut foundation trenches for coastal fortifications. These indicators are used as an upper limit for past sea levels (Table 2). Secondly, the study estimates sea level based on structures which depend on the height of the coastal water table for their
Results
The study uses published data, together with the new observations, all presented in Fig. 2a and b. The current paper focuses mainly on the Crusader period, first discusses the data, and then provides a tentative explanation for the observed sea-level fall during this period (11th to 13th centuries).
Discussion
The low Crusader levels suggested by the observations require some physical forcing for explanation. Rates of change of about 0.5 m within a few centuries are certainly within the limits of natural variability, considering that rates of change during deglaciation periods have reached 5 m per century (Blanchon and Shaw, 1995, Fairbanks, 1989), and rates of more than 1 m per century have been shown within interglacial periods (Rohling et al., 2008). To that extent the sea level estimates presented
Conclusions
New estimates of sea level during the Crusader period, based on archaeological indicators, suggest a sea-level fall during a period of a few centuries along the Mediterranean coast of Israel. The Crusader mean sea-level fall is in the range of − 0.13 to − 0.85 m, but with higher confidence for the lower values. A reassessment of the GIA effects confirms that these cannot be considered to be large enough to explain the suggested sea level fall. Because the area is believed to be tectonically stable
Acknowledgments
The authors would like to thank the Sir Maurice and Lady Hatter Fund of the Leon Recanati Institute for Maritime Studies (RIMS), University of Haifa, for supporting Eilat Toker's study, the data and results of which are the core of the current paper. Many thanks are also due to Shalom Yankelevich, Dani Doron, Yoram Efrat and Dani Sion for the unpublished data. We also wish to mention that this work was partly supported by COST Action ES0701 “Improved Constraints on Models of Glacial Isostatic
References (71)
- et al.
Holocene relative sea-level changes and vertical movements along the Italian and Istrian coastlines
Quat. Int.
(2009) - et al.
Archaeological remains as sea level change markers: a review
Quat. Int.
(2009) - et al.
Late Holocene climates of the Near East deduced from Dead Sea level variations and regional winter rainfall
Quat. Res.
(2003) - et al.
Beach deposits of MIS 5e high sea stand as indicators for tectonic stability of the Carmel coastal plain, Israel
Quat. Sci. Rev.
(2007) Extreme Nile floods and famines in Medieval Egypt (AD 930–1500) and their climatic implications
Quat. Int.
(2007)- et al.
Sea-level change in the Mediterranean Sea since the LGM: model predictions for tectonically stable areas
Quat. Sci. Rev.
(2005) - et al.
Sea-level change along the Italian coast for the past 10,000 yr
Quat. Sci. Rev.
(2004) - et al.
Sea level in Roman time in the Central Mediterranean and implications for recent change
Earth Planet. Sci. Lett.
(2004) - et al.
Holocene climate variability
Quat. Res.
(2004) - et al.
Climate deterioration in the Eastern Mediterranean as revealed by ion microprobe analysis of a speleothem that grew from 2.2 to 0.9 ka in Soreq Cave, Israel
Quat. Res.
(2009)