We developed a suite of quality metrics that characterize the annual and year-to-year frequency of satellite observations, their year-to-year recurrence, and their within-year distribution. These equate the quality of individual observations as reported by the data providers, and limitations to the usability of these data caused by cloud cover. This dataset includes a zip file for each of the following metrics: nrTiles - Number of Landsat tiles used to compute quality metric. imageFrequency - Number of collected images. monthFrequency - Number of months with collected images. maxQuality - Maximum image quality. totalQuality - Number of collected images, weighted by the quality of each image. lastYear - Closest year with usable data, from the start of the time-series to the reference year. distributionBalance - Average of the maximum monthly image quality. distributionQuality - Ratio between the distribution balance of the first and second half of the year. We calculated these quality metrics for each descending tile as drawn in the World Reference System 2 (WRS-2), and for each year. Then, for each year, we combined the tile-specific metrics by averaging them into global grids with a 1-km resolution using the script "map_landsat_quality.py". This was executed in python 3.10. and the associated module requirements are recorded in the file "requirements.txt". All tile-specific metrics, which is an input for the python script, are provided through the file "LTQA_metadata.csv". The calculation of quality metrics is informed by metadata of all unique acquisitions obtained through the Landsat’s bulk metadata service. This considers images acquired with Landsat 4, 5, 7, 8, and 9, but disregards those from Landsat’s Multispectral Scanner System (MSS). The original metadata is provided through the file "LTQA_metadata.zip", which also includes the R code used in the calculation of quality metrics, and a data structure that can be updated to generate new values. We calculated these metrics are calculated annually discounting those unusable due to 100% cloud cover or advanced image degradation.