This dataset is provided in support of the following publication: "Solar and sensor geometry, not vegetation response, drive satellite NDVI phenology in widespread ecosystems of the western United States". The data and code provided allow users to replicate, test, or further explore results. The dataset includes 2 raster datasets (folder:Rasters): 1) 'cntWinterPks2003_2018DR' provides a count of years with winter peaks from 2003-2018 in an 11-state area in the western United States. 2) 'VegClassGte5_2003_2018' raster, within the zip file 'WinterPeaksVegTypes.zip' identifies the broad vegetation types for locations with common winter peaks (5 or more years out of 16). The dataset also includes Google Earth Engine and R code files used to create the datasets. Additional files/folders provided include 1) Google Earth Engine scripts used to download MODIS data the GEE - javascript interface (folder: 'Code'). 2) Scripts used to manipulate rasters and to calculate and map the occurrence winter NDVI peaks from 2003-2018 using the statistical software package 'R'. 3) Supplemental R-project and packrat files that allow the user to apply the workflow by opening a project that will use the same package versions used in this study, for example the folders 'Rproj.user', and 'packrat', and files '.RData', and 'WinterPeakExtentPR.Rproj'. 4) Empty folders ('GEE_DataAnnPeak', 'GEE_DataLoose', and 'GEE_DataStrict') that should be used to contain the output from the GEE code files as follows: 'GEE_DataAnnPeak' should contain output from the S3 and S4 scripts, 'GEE_DataLoose' should contain output from the S1 script, and 'GEE_DataStrict' should contain output from the S2 script. 5) Graphic file 'Fig_9_MapsOfExtentPortrait2.jpg' shows temporal and ecosystem distribution of winter NDVI peaks in the western continental US, 2003 to 2018, derived from MODIS MCD43A4 product. TOP: Number of years with winter peaks in areas that meet defined thresholds for biomass (median annual peak NDVI >= 0.15) and temperature (mean December minimum daily temperature <= 0°C). BOTTOM: Predominant LANDFIRE Existing Vegetation Type physiognomy (i.e., mode of each 500-m MODIS pixel) in areas with >= 5 years of winter peaks. Present in lesser proportions but not identified on the map for legibility reasons are conifer-hardwood, exotics, riparian, and sparsely vegetated physiognomic categories as well as non-natural/non-terrestrial ecosystem categories. State abbreviations are AZ (Arizona), CA (California), CO (Colorado), ID (Idaho), MT (Montana), NV (Nevada), NM (New Mexico), OR (Oregon), WA (Washington), and WY (Wyoming). The final steps of overlaying common winter peak extent data on the Landfire data were done using ArcGIS and the publicly available Landfire dataset (see source datasets section of metadata and process steps). To successfully use the packrat information to replicate the exact processing steps that were used, the user should refer to packrat documentation available at https://cran.r-project.org/web/packages/packrat/index.html and at https://www.rdocumentation.org/packages/packrat/versions/0.5.0. Alternatively, the user may also use the descriptive documentation within this metadata along with the workflow described in the associated journal article to process the data to achieve the same results using newer packages or other software programs.