A global near real-time filtered 500m 10-day FPAR dataset from MODIS and VIIRS instruments, suited for operational agricultural monitoring and crop yield forecasting

Climate change and extreme weather events pose challenges to food security, emphasizing the need for reliable and timely monitoring of crop conditions. This monitoring relies in part on long-term consistent datasets on vegetation conditions. However, sensor failure of existing systems like MODIS results in challenges in obtaining long-term operational records that can effectively shows anomalous crop and rangeland status. To address this challenge, we present a near-real-time (NRT) global dataset of Fraction of Photosynthetically Active Radiation (FPAR) at 500m resolution, optimized for agricultural applications. Our dataset combines daily MODIS-FPAR (Collection 6.1) and VIIRS-FPAR (Collection 2) data, ensuring continuity from 2000 to well beyond 2030. We applied a robust filtering approach based on the Whittaker smoother to produce reliable FPAR estimates in NRT, accounting for sparse observations due to cloud cover. The dataset is composed of two 10-day filtered time series: 1) MODIS-FPAR for 2000 to 2022 and 2) intercalibrated VIIRS-FPAR for 2018 onward. While several methods can smooth FPAR data (i.e., using observations before and after the estimation date), filtering in NRT (using only prior observations) is more challenging. Our approach generates six FPAR estimates: the first immediately after the reference 10-day period, followed by four estimates every 10 days using new observations, and a final consolidated estimate 90 days later. The implemented filtering ingests all available FPAR observations, weighing them using their original quality assessment (QA) layers. To avoid unrealistic extrapolation when observations are sparse, we impose a constraint to the possible FPAR values for each 10-day period. For each pixel, we then intercalibrated the VIIRS-FPAR with the MODIS-FPAR filtered time series, using a mean difference correction approach, to ensure consistency between both series. This paper describes the filtering and intercalibration method used, the quality assessment of resulting timeseries, and a detailed description of the product and corresponding QA layers. The NRT FPAR dataset is publicly available online and supports operational vegetation monitoring and crop yield forecasting.