Optimum temporal and spectral window for monitoring crop marks over archaeological remains in the Mediterranean region

This paper aims to explore the formation of crops marks over archaeological features using remote sensing data. Crop marks are very important to archaeological research since they can reveal in an indirect way buried archaeological remains. Several researches have been able to locate subsurface relics via interpretation of multispectral and hyperspectral satellite images. However the best phenophase (time-window) where these crop marks are enhanced and which spectral bands are the most capable for this enhancement are still an open question for the scientific community. This study aims to address these fundamental aspects of Remote Sensing Archaeology. For this reason an extensive dataset was used: two control archaeological sites were constructed in two different areas of Cyprus (Alampra and Acheleia). In these areas both barley and wheat, were cultivated. In situ spectroradiometric measurements were taken over a whole phenological cycle (2001–2012) for both areas. More than 30 in situ campaigns were performed and more than 2600 ground measurements were collected. The phenological analysis of these measurements have shown that a period of only 15 days is best suitable for monitoring crop marks which is connected with one phenological stage; during the beginning of the boot stage of the crops. This period can be used in areas having similar climatic condition as Cyprus (Mediterranean region). Also as it was found, this phenological stage of crops might be also influenced by the climatic conditions of the area. The evaluation of a proposed Archaeological Index was performed at the Thessalian plain (central Greece), where several Neolithic settlements are established. The evaluation was made using both Hyperion and Landsat images. The results have shown that the proposed Archaeological Index is suitable for the enhancement of crop marks in satellite imagery.

► The optimum spectral and time-window for detecting crop marks is presented.
► One phenological stage of the crops is best suitable for detection of crop marks.
► An Archaeological Index which enhance crop marks was developed.
► The Archaeological Index was evaluated successfully using hyperspectral images.