ABSTRACT The detection of buried or obscured archaeological features remains a central challenge in landscape archaeology, particularly in the irrigated floodplains of Mesopotamia where levees and canals formed the basis of complex agrarian systems. This study presents a deep learning–based approach for the large‐scale, automated detection of ancient levees in central Iraq, integrating big multitemporal and multisource satellite datasets with advanced instance segmentation models. Datasets were assembled from multitemporal Landsat 5, Sentinel‐1 SAR, Sentinel‐2 multispectral imagery and the TanDEM‐X Edited DSM, combined with vegetation and moisture indices, PCA reductions of seasonal variability and Multi‐Scale Relief Model (MSRM) outputs. Training labels were generated through both threshold‐based automatic extraction and detailed manual digitization. Three architectures—U‐Net, Attention U‐Net and Swin UNETR—were evaluated on datasets containing 53, 48 and 36 bands. Results demonstrate that Swin UNETR consistently outperformed other models, particularly when trained on the 48‐band dataset with manually digitized levees. Unlike wide automatic annotations, which produced irregular and noisy patches, thin manual annotations yielded clearer, more linear predictions. Post‐processing further refined the outputs, enabling the model to achieve pixel‐level precision of 0.6555 and recall of 0.5107 and vector‐level precision of 0.7554 and recall of 0.7743 after additional post‐processing. Although pixel‐level metric scores remain modest, reflecting the irregularity of the archaeological features, the model successfully predicted levee networks across ~31 250 km 2 , extending from the Ba'qubah region to Ad Rumaythah, with detected ~13 680 km potential levees located outside the dataset (20 660 km 2 ). Comparative analysis with independent palaeochannel reconstructions confirmed that the model identified many of the most prominent irrigation features while avoiding misclassification of modern infrastructure. The results of the Monte Carlo simulation indicate a clear relationship between identified levees and archaeological sites of different periods, particularly Old Babylonian, Parthian, Sasanian and Early Islamic. The results highlight both the challenges and promise of deep learning in archaeological remote sensing. Automated predictions cannot yet replace interpretative digitization, but they provide reproducible, standardized and scalable outputs that can accelerate archaeological mapping and support regional‐scale analysis. By leveraging multitemporal, multisource datasets and advanced AI architectures, this study demonstrates a pathway towards reconstructing irrigation systems of different historical periods and landscapes. The approach opens new possibilities for documenting, preserving and interpreting water management legacies in some of the world's most significant ancient landscapes.
Buławka et al. (Wed,) studied this question.