Based on the analysis of the aquifer system feeding the Yperia Krini spring, it is likely that the strong decrease of water discharge during the last decade was caused by the drilling of numerous new boreholes for the water supply. In fact, these bore-holes pump significant volumes of water directly from the aquifer, thus exceeding the capacity of the reservoir to recover (Fig. 10).

On the other hand, it seems clear that the present-day Yperia Krini crisis should be included in the more regional hydrogeological problem that affects the broader Larissa Plain. However, as discussed, the spring has already suffered periods of partial (or complete) dryness in historical times, though it is not possible to cite the drilling of boreholes as a possible cause of previous dry events.

In general, the quality of life and the periods of expansion and reduction of the old town were directly related to the amount of water discharge. Moreover, the disappearance of at least two ancient towns at the end of the Hellenistic epoch or at the beginning of the Imperial epoch, matches that of Pherae. Therefore, although the local conditions are different and there is not perfect chronological coincidence between the three events, the causes responsible for this phenomenon are likely to be regional ones. Indeed, in the case of Pherae, it is likely that the loss of the spring Yperia induced the abandonment of the town.

Two different, but not alternative, explanations for this phenomenon are presented in the following. Firstly, a possible unique natural cause occurring at

the regional scale and generating instability in this sector of the plain, is the active tectonism which has affected the area since the Middle-Late Pleistocene (Caputo 1990, 1995; Caputo & Pavlides 1993). Although no specific data are available to support this hypothesis for the study area, a close relationship between the variations of the stress field, as induced by earthquakes, and the underground water level is well documented world-wide (Roeloffs 1988; Oki & Hiraga 1988; Kumpel 1992; Asteriadis & Contadakis 1994).

Secondly, it is well known that from the third century BC to the fourth century AD the northern hemisphere underwent the so-called 'Roman climatic optimum' (Panizza 1985; Veggiani 1994). However, if the term 'optimum' can certainly refer to Europe where ice-caps retreated, in contrast in the Mediterranean regions, this warm period evidently generated drier climatic conditions and thus a generally reduced amount of precipitation (Greig & Turner 1986). For this reason, we attempted to analyse the response of our hydrogeo-logical system in the case of dry periods.

Let us consider the relationships between mean annual precipitation and mean water discharge (Fig. 8). For example, just one year of reduced precipitation (1977), with 220 mm/a corresponding to about 50% of the average value for the period 1971-1984 (465 mm/a), caused a decrease of more than 15% in water discharge for as long as two years (1978 and 1979).

Accordingly, whatever the real amount is, it is clear that a decrease of precipitation, of less than 50% but lasting several centuries, such as occurred during the 'Roman' epoch (c. third century BC to fourth century AD; Lespez 2003; Bottema 1979; Digerfeldt et al. 2000; Dragoni 1998), certainly generated a progressive depletion of the reservoir and a consequent strong reduction in water discharge of the Yperia Krini spring. Based on the flourishing environmental conditions still existing during 196 BC, the degradation phenomenon was necessarily very slow and probably not perceivable during a human lifetime. Nevertheless, in order to induce a migration of the population from Pherae and consequent decay of the old town, it is obviously not necessary for a drastic reduction of the water discharge to have occurred. Indeed, during the first century BC, after 300 years of warm climate, the spring probably did not provide enough water for a large number of inhabitants.

0 0

Post a comment