Data — IrrEM (Irrigation in the eastern Mediterranean): Two and a half decades of seasonal high-resolution (30m) irrigated land maps of the eastern Mediterranean (2000-2025)

Dataset Documentation: Eastern Mediterranean Irrigation Extent (2000–2025) Overview This dataset contains annual irrigation extent maps, detection event counts, valid observation metadata, and related modeling parameters for the Eastern Mediterranean from 2000 to 2025. The data is derived from Landsat imagery using the Optical Trapezoid Model (OPTRAM) to map irrigation extent based on relative moisture changes. The dataset consists of the following primary categories: (Final) temporally smoothed irrigation extent for summer season and supplementary irrigation, (Raw) Irrigation extent for summer season and supplementary irrigation, Count of detected irrigation event for summer season and supplementary irrigation, Number of valid satellite observation for summer season and supplementary irrigation, post-processed, and Supplementary Tabular Data. All geospatial images are multi-band composite files, where each band represents a specific year. 1. Temporally Smoothed (Final) Irrigation Data These files contain the final, refined irrigation extents after applying a 5-year weighted majority temporal filter. This process utilizes the event counts and valid observations to fill physical gaps and eliminate isolated noise/anomalies. File Name Parent File Description File Names Data Range / Values Ir_Summ_Wsmooth_YEAR Final_IrrigationEM_YearlyImages.zip Smoothed Summer Extent. The temporally filtered binary classification for summer irrigation. 26 Tiff file single band image, each image represents a specific year. YEAR ∈ [ 2000 -2025] 0 = Non-irrigated 1 = Irrigated Ir_Supp_Wsmooth_YEAR Final_IrrigationEM_YearlyImages.zip Smoothed Supplementary Extent. The temporally filtered binary classification for wet season supplementary irrigation. 26 Tiff file single band image, each image represents a specific year. YEAR ∈ [ 2000 -2025] 0 = Non-irrigated 1 = Irrigated 2. Unsmoothed (Raw) Irrigation Data + Detected events count + raw Observations These files contain the direct outputs of the spatial detection algorithm prior to any multi-year temporal smoothing. File Name Parent File Description Band Names Data Range / Values Ir_Summ_All Raw_Data_Irr_Summ.zip Summer Irrigation Extent. Binary classification of irrigation presence during the dry summer season. Multi-band composite files, where each band represents a specific year. Ir2000 to Ir2025 0 = Non-irrigated 1 = Irrigated Ir_Summ_Count_All Raw_Data_Irr_Summ.zip Summer Event Count. The absolute number of distinct irrigation events detected per pixel during the summer season. Multi-band composite files, where each band represents a specific year. CIr2000 to CIr2025 [0, 1, 2, 3, ...] (Integer) Ir_Summ_V_All Raw_Data_Irr_Summ.zip Summer Valid Observations. The number of cloud-free, valid Landsat scenes available and processed per pixel during the summer. Multi-band composite files, where each band represents a specific year. VIr2000 to VIr2025 [0, 1, 2, 3, ...] (Integer) Ir_Supp_All Raw_Data_Irr_Supp.zip Supplementary Irrigation Extent. Binary classification of supplementary irrigation presence during the in-season (wet season). Multi-band composite files, where each band represents a specific year. Ir2000 to Ir2025 0 = Non-irrigated 1 = Irrigated Ir_Supp_Count_All Raw_Data_Irr_Supp.zip Supplementary Event Count. The absolute number of distinct irrigation events detected per pixel during the wet season. Multi-band composite files, where each band represents a specific year. CIr2000 to CIr2025 [0, 1, 2, 3, ...] (Integer) Ir_Supp_V_All Raw_Data_Irr_Supp.zip Supplementary Valid Observations. The number of cloud-free, valid Landsat scenes available and processed per pixel during the wet season. Multi-band composite files, where each band represents a specific year. VIr2000 to VIr2025 [0, 1, 2, 3, ...] (Integer) 3. Supplementary Tabular Data These files provide the underlying model parameters and the validation data used for the accuracy assessment. File Name Format Description Params_final.csv CSV OPTRAM Parameters. Contains the calculated wet/dry edge linear parameters (e.g., intercept, slope) for each distinct Landsat footprint (Path/Row) per year. Reference_data.csv CSV Accuracy Assessment Reference. Contains the spatial coordinates, manually assigned ground-truth labels, and corresponding Landsat NDVI timeseries data used to validate the classification models.