Urban flooding poses a growing threat to cities in arid and semi-arid regions, where rapid urbanization, limited drainage infrastructure, and scarce hydrological data exacerbate disaster risks. This study develops Hydro-TransformerNet, a novel hybrid deep learning framework for urban flood susceptibility mapping under data-scarce conditions. The model integrates a ResNet34-based convolutional encoder for fine-scale spatial feature extraction, a temporal transformer module for simulating sequential hydrological processes, and a hydrologically guided attention mechanism that embeds physical terrain and rainfall relationships into the learning process. The model uses nine remote sensing-based geospatial variables: Digital Elevation Model (DEM), Slope, Normalized Difference Water Index (NDWI), Normalized Difference Vegetation Index (NDVI), land use/land cover (LULC), Soil Type, Impervious Surface, Rainfall, and Topographic Wetness Index; as input features for the Sharjah region, UAE. Due to the absence of historical flood information, a synthetic flood mask was developed using expert knowledge of hydrological patterns and low-lying urban areas. The model demonstrates strong predictive performance, achieving an AUC of 0.945, with high alignment between predicted flood-prone areas and plausible hydrological zones. Quantitative evaluation using ROC, precision-recall, calibration, and confusion matrix metrics confirm the model's reliability and robustness. Feature importance analysis via SHapley Additive exPlanations (SHAP)-based feature importance analysis reveals elevation, slope, and rainfall as dominant factors in flood susceptibility. The model's predictions enabling integration with GIS platforms for urban planning and disaster mitigation. Hydro-TransformerNet represents a significant step in creating scalable, interpretable, and GIS-ready flood prediction tools for urban regions facing data scarcity. This study contributes a scalable, explainable nd data-efficient deep learning solution for early-warning systems, infrastructure planning, and climate resilience management in rapidly developing arid cities.Graphical AbstractThis presents a complete visual representation of the framework, designed to assess urban flood susceptibility in arid and semi-arid regions using advanced deep learning architecture that integrates convolutional and transformer-based modules. The model begins by ingesting a set of nine hydro-environmental input parameters, Elevation, Slope, Rainfall, NDVI, NDWI, TWI, Impervious Surface Area, LULC, and Soil Type. These input layers are first subjected to a multi-collinearity assessment using TOL and VIF equations to ensure independence among predictors and eliminate redundant variables, which can otherwise impair model learning. Once the input layers are validated, they are resampled and stacked into a multi-channel geospatial tensor that serves as the model's input feature block. This tensor is passed through a CNN encoder, which captures local spatial patterns and feature hierarchies across the urban landscape. The encoded spatial features are then processed by a Transformer module, which is responsible for modeling long-range dependencies and contextual relationships, which are crucial in urban flood modeling. To further enhance the model's domain-awareness, a hydro-attention mechanism is applied that re-weights spatial features based on their hydrological relevance, such as prioritizing low-elevation or impervious regions prone to flooding. The model is trained using synthetic binary flood masks, representing probable inundation areas derived through expert-guided heuristics in the absence of ground truth flood data. These masks allow the model to learn flood patterns in data-scarce environments like Sharjah in UAE. Model evaluation metrics are also visualized in the graphical abstract. The Receiver Operating Characteristic (ROC) curve displays a strong Area Under Curve (AUC) score of 0.945, indicating high discrimination ability. In addition, feature importance analysis, derived from post-hoc SHAP explanations, reveals that Elevation, Slope, and Rainfall are the most influential predictors, while features like NDVI and Soil Type contribute less.