Agriculture is the largest global consumer of water. Irrigated areas constitute 40 % of the total area used for agricultural production. Using more effective irrigation practices can enhance carbon storage in soils through enhanced yields and residue returns. Irrigation management has a strong impact on the disease severity and epidemic progress rates of many plants, going from leaf blights to vascular wilts, additionally, plant virus vector population levels and vector dispersal are also affected by irrigation practices. The use of Earth observation (EO) to map irrigation practices in agricultural areas is key since it can be used to generate quantitative information about vegetation health in a consistent and systematic manner. Additionally, using climate and weather variables such as precipitation in a synergistic method together with EO data can improve identification and characterisation of irrigation practices. Here, we used this synergistic approach by using i) EO data to derive time series of a vegetation index as an indicator to identify active and healthy vegetation and ii) precipitation from climate reanalysis data to identify areas where photosynthetically active vegetation was present, but no precipitation was observed in the previous weeks, hence flagging those areas as potentially irrigated areas. The methodology relies on identifying a peak or peaks in the greenness indicator that is not associated with precipitation events. Therefore, we can identify agricultural areas where irrigation has been applied once or more times within a calendar year. We use the MODIS-Terra Vegetation Indices (MOD13Q1) Version 6.1 data that are generated every 16 days at 250m spatial resolution. After a Quality Assessment (QA) process, a harmonic analysis was performed to generate a gap-free time series. Then per-pixel ‘peaks’ and ‘valleys’' within a calendar year were identified. The ancillary variable selected was precipitation obtained from ERA5-Land. The ERA5-Land dataset is available for public use for the period from 1950 to 5 days before the current date. We created a time series of the monthly total precipitation and applied the “peaks” finder algorithm to identify which month corresponds to the one with the most accumulated precipitation in a calendar year. The latest step in the algorithm was to identify pixels that had peak in the EVI (enough to be considered photosynthetically active cropland) that did not have an associated peak in the precipitation. To validate the EO results, CABI conducted a search for available data on irrigation practices. Published datasets are readily available on a basic level and at a coarse geographic scale, but in order to assess the outputs, detailed data on the timing and location of irrigation activity was required. Published datasets were collated which cover numerous studies in the USA as well as information from Uzbekistan and Tajikistan. CABI and Cervantes carried out fieldwork in Boliva, Kenya and Pakistan to gather geographic information on irrigation practices in agricultural fields. This data was used to ground truth the EO results. The whole processing chain was developed in the Google Earth Engine at the original 250m spatial resolution. A global dataset for 2023 was generated. An offline resample to 0.16Deg was carried out to match the resolution of the climate models use in the project. Every grid cell provides the fraction of irrigated areas within that cell. The dataset is available in GeoTiff, NetCDF and Shapefile formats. Funding for the Allocated Work under this Agreement is made available by Science and Technology Facilities Council Grant: EO4AgroClimate Using Earth Observation data to improve datasets for biosecurity risk mapping of pest and disease and biocontrol suitability (Ref ST/Y00017X/1)