Marleen Vermoere

Man and environment in the territory of Sagalassos, a classical city in SW Turkey

Abstract Since 1990 archaeological research by the Catholic University of Leuven (Belgium) has been carried out at the ancient site of Sagalassos (Aǧlasun, Burdur province, Southwestern Turkey). At first, research focused on the excavation of the city and the study of the immediate vicinity which provided it with raw materials. The main objective was to obtain a clear picture of the history and development of the city. Since 1993 research has also incorporated a study of the territory of the Roman city, from prehistoric to modern times, in order to understand why the site was selected for settlement, why it developed into a middle-sized town, its economy and subsistence, how it affected and exploited the environment, its decline, and what changes have taken place in the district subsequently. The focus has now shifted towards obtaining a better understanding of the linkages between human and environment systems so that inter-relations between the two can be more readily understood. As a result, a number of environmental topics concerning the territory of the Roman city are presently being studied. This territory extended from Lake Burdur in the West to the Aksu canyon in the East, from the Aǧlasun Daǧlari in the North to Mt. Kestel in the South. Interdisciplinary research revealed that for the early Neolithic and the Roman period there was a slightly warmer climate, a richer vegetation and more fertile soils for agricultural practice.

Pollen analysis of two travertine sections in Basköy (southwestern Turkey): implications for environmental conditions during the early Holocene

Two travertine sections in the valley of Baskoy, near the archaeological site of Sagalassos have been palynologically investigated. The base of one of these sections has been dated to 9000±600 yr B.P. by the U/Th dating technique. The pollen diagrams of both sections reveal the importance of deciduous Quercus and Betula woodlands during deposition of the travertine. The development of woodlands in the early Holocene implies an increase in humidity after the Late Pleistocene. An annual precipitation of at least 500 mm/yr is suggested. A study of the δ18O values of the travertines in the sections indicates a change to higher precipitation temperatures of the travertines and hence a climatic change to higher air temperatures around 9000±600 yr B.P. This increase in temperature is not documented in the pollen diagrams. The evaluation of the local pollen curves provides additional information on the depositional environment. The facies and pollen analyses indicate that the travertines were deposited in a fluvial–barrage system. High abundances of Cyperaceae pollen suggest a marshy environment and fluctuating water levels of this early Holocene fluvial-barrage system. High percentages of Poaceae and/or Aster and Mentha–Thymus pollen types point to drier local conditions at the beginning and the end of the travertine accumulation of both sections.

Palynological evidence for late-Holocene human occupation recorded in two wetlands in SW Turkey

Pollen diagrams from mountain lakes and marshes in SW Turkey show evidence of intensive anthro pogenic influence in the landscape between 3500 and 1300 BP. Three cores from within the territory of the classical city Sagalassos (Western Taurus, Pisidia) were palynologically analysed to make a reconstruction of the past vegetation in the territory of Sagalassos and to estimate the impact of its inhabitants on the landscape in Pisidia. Two cores originate from an intramontane marsh (Gravgaz; elevation 1215 m) and one from a seasonal intramontane lake (ÇanaklK; elevation 1030 m). Human acitivity is apparent from c. 2530 BP in the pollen diagrams. A deforestation phase/‘disturbance’ phase (from c. 2530 BP/c. 2480 BP till c. 2280 BP/c. 2270 BP), anthropogenically or climatologically driven, precedes a cultivation period (arboricultural phase) (from c. 2280 BP/2270 BP till c. 1480 BP/c. 1270 BP). The results of the pollen analyses of the three cores are compared and show many similarities. There are also similarities with other cores from SW Turkey, although the arboricultural phase (mainly olive cultivation) occurs later in the cores from the Sagalassos territory than in other cores from SW Turkey.

A comparison between modern pollen spectra of moss cushions and Cundill pollen traps

Pollen spectra of 23 Cundill pollen traps from 23 different sampling sites in Southwest Turkey are compared with the corresponding pollen spectra of moss cushions from the same sites. The Cundill pollen traps represent the modern pollen rain data from one year whereas the moss cushions represent the pollen rain of several years. The comparative study reveals some main differences between the two pollen entrapment media. The one-year entrapment medium (pollen trap) appears to be more sensitive for local (releve area 10x10 m) and regional (100-500 m, or a few kilometres for Olea europaea) vegetation. Pollen spectra of moss cushions are dominated by high pine pollen percentage values and hardly sense fine vegetation structures. The conclusion of this comparative modern pollen study allows to interpret fossil sediment spectra from the Near East in a more critical way. It is concluded that one should preferably sample rapidly deposited sediments for palynological analyses, as the resulting highresolution pollen diagrams will be most informative about the former vegetation patterns.

Modern pollen studies in the territory of Sagalassos (Southwest Turkey) and their use in the interpretation of a Late Holocene pollen diagram

Modern pollen precipitation studies were carried out on a regional scale in the territory of the classical city of Sagalassos (Southwest Turkey). 57 moss cushions and 7 soil samples from different vegetation types were analysed palynologically. The resulting pollen spectra were grouped into 14 vegetation groups, which were examined using discriminant analysis and principal components analysis. These numerical procedures are applied to display the structure in the data set in a comprehensive way. Some of the modern spectra groups seem to be well-defined, whereas others are difficult to distinguish from each other. After the relationships between the modern pollen spectra had been analysed, the fossil data from a core from the marsh of Gravgaz were added to the data set so that any similarities between fossil and modern spectra could be seen. A joint principal components analysis of the modern and fossil pollen spectra revealed that the pollen spectra from the earliest (ca. 2600-2500 BP; Cal. 830-590 BC till 790-520 BC) and the latest (starting at ca. 1300 BP; Cal. 660-780 AD) pollen assemblage zones have modern counterparts. On the other hand, the pollen spectra from the pollen assemblage zone dominated by Artemisia (ca. 2500-2300 BP; Cal. 790-520 BC till 410-210 BC) and from the pollen assemblage zone that indicated the practice of arboriculture (ca. 2300-1300 BP; Cal. 410-210 BC till 660-780 AD) have no close modern analogues, although individual tabulated pollen percentages revealed some similarities with some modern spectra groups.

Late Holocene local vegetation dynamics in the marsh of Gravgaz (southwest Turkey)

Late Holocene local vegetation succession is reconstructed in twodifferent sites in a small-scale open marsh ecosystem in southwest Turkey.This is done by comparison of the fossil local pollen assemblage zones in twocores with the local pollen data of 40 surface samples from the marsh. Thepollen data are supplemented with sedimentological and archaeological data. Theinsertion of the mean pollen data of the local pollen zones as passive samplesinto the canonical correspondence analysis triplot of the modern samples allowsus to detect modern analogues for the fossil pollen zones. From this numericalcomparative approach it is concluded that the marsh area was relatively dryuntil ca 2500 BP. After 2500 BP the area shifts towards a wet area dominated bySparganium and/or Typha angustifolia. A diversification of the marsh vegetationstarts at ca 2400/2300 BP. The area around one core site seems to have beensituated in an area with slowly flowing source water, whereas the other coresite is likely to have been characterised by damp conditions. The steadilyincreasing dryness of the marsh area starts after ca 680 BP. The drying upappears to be associated with recent agricultural and grazing pressure.