This review article discusses the use of optical remotely sensed time series data for land cover classification. Land cover information is essential for science, monitoring, and reporting, as land cover changes naturally over time and is influenced by human activities. Earth Observation (EO) data, particularly from Landsat and Sentinel-2, enable the production of annual, large-area, gap-free surface reflectance data products. These data support the development of annual land cover products informed by change detection. Time series data provide information on class stability and logical class transitions, aiding in the characterization of land cover. The review highlights the challenges and opportunities in generating and validating time series-informed annual land cover products. It identifies methods suitable for incorporating time series information and other novel inputs for land cover characterization. The availability of EO data, improved computing capacity, and novel image compositing approaches have enabled the production of annual, large-area land cover products. These products are essential for monitoring land cover changes, understanding environmental processes, and supporting policy decisions. The article discusses the development of methods for generating annual land cover products using time series data. It emphasizes the importance of temporal and spatial resolution in land cover mapping and the need for robust classification algorithms. The review also addresses the use of multi-temporal spectral data, which provides rich information for land cover characterization. It highlights the importance of incorporating ecological knowledge and temporal consistency in land cover classification. The review discusses the use of various classification algorithms, including unsupervised and supervised methods, for land cover classification. It emphasizes the need for accurate training samples and the importance of temporal and spatial resolution in land cover mapping. The article also discusses the use of multi-scale and multi-sensor data for improving land cover characterization. It highlights the importance of incorporating ecological knowledge and temporal consistency in land cover classification. The review concludes that the integration of time series data and novel inputs is essential for accurate and consistent land cover characterization.This review article discusses the use of optical remotely sensed time series data for land cover classification. Land cover information is essential for science, monitoring, and reporting, as land cover changes naturally over time and is influenced by human activities. Earth Observation (EO) data, particularly from Landsat and Sentinel-2, enable the production of annual, large-area, gap-free surface reflectance data products. These data support the development of annual land cover products informed by change detection. Time series data provide information on class stability and logical class transitions, aiding in the characterization of land cover. The review highlights the challenges and opportunities in generating and validating time series-informed annual land cover products. It identifies methods suitable for incorporating time series information and other novel inputs for land cover characterization. The availability of EO data, improved computing capacity, and novel image compositing approaches have enabled the production of annual, large-area land cover products. These products are essential for monitoring land cover changes, understanding environmental processes, and supporting policy decisions. The article discusses the development of methods for generating annual land cover products using time series data. It emphasizes the importance of temporal and spatial resolution in land cover mapping and the need for robust classification algorithms. The review also addresses the use of multi-temporal spectral data, which provides rich information for land cover characterization. It highlights the importance of incorporating ecological knowledge and temporal consistency in land cover classification. The review discusses the use of various classification algorithms, including unsupervised and supervised methods, for land cover classification. It emphasizes the need for accurate training samples and the importance of temporal and spatial resolution in land cover mapping. The article also discusses the use of multi-scale and multi-sensor data for improving land cover characterization. It highlights the importance of incorporating ecological knowledge and temporal consistency in land cover classification. The review concludes that the integration of time series data and novel inputs is essential for accurate and consistent land cover characterization.