Archaeological Reconnaissance in Tarbagatai (East Kazakhstan) Using Satellite Imagery Data: methodology and results of research
The paper deals with the results of an archaeological reconnaissance based on satellite imagery in the Tarbagatai mountain range and its foothills (Abai region, Republic of Kazakhstan). One of the key aspects of the study was the detailed development of a survey methodology, which was divided into t...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
State institution «Tatarstan Аcademy of Sciences»
2025-04-01
|
| Series: | Археология евразийских степей |
| Subjects: | |
| Online Access: | https://www.evrazstep.ru/index.php/aes/article/view/1571 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The paper deals with the results of an archaeological reconnaissance based on satellite imagery in the Tarbagatai mountain range and its foothills (Abai region, Republic of Kazakhstan). One of the key aspects of the study was the detailed development of a survey methodology, which was divided into three stages: preparatory, searcg, and verification. During the preparatory stage, mapping of archaeological sites known from archival sources and publications was carried out. This made it possible to gather essential data on how these sites appear in satellite imagery, which was later used to search for new archaeological objects. The search phase involved a visual examination of satellite images covering a total area of 17,309 km². The search was conducted systematically using a specially designed grid, ensuring comprehensive coverage of the area. At the verification stage, all identified objects were reviewed in accordance with the experience accumulated during the survey. As a result, 1166 objects were identified as archaeological sites: 664 of them are very likely to be archaeological sites (category 1) and 502 require further verification (category 2). The distribution of the identified objects by type is as follows: burial sites – 1132, settlement ones – 33. The identified objects contain a total of 4582 elements (barrows, masonry, etc.). The article also examines the relationship between the applied archaeological reconnaissance methodology and the rapidly developing technologies of neural networks and machine learning for site detection. It is concluded that the proposed methodology, due to its free of charge and greater availability, holds an independent value, at the same time, the data obtained with its help can be used in the future to train neural networks to search for previously unknown sites and other archaeological research. |
|---|---|
| ISSN: | 2587-6112 2618-9488 |