Landsat Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) sensors were used to generate the circa 2010 Mosaic of Canada at 30 m spatial resolution. All scenes were processed to Standard Terrain Correction Level 1T by the United States Geological Survey (USGS). Further processing performed by the Canada Centre for Remote Sensing included conversion of sensor measurements to top of atmosphere reflectance, cloud and cloud shadow detection, re-projection, selection of best measurements, mosaic generation ,noise removal and quality control. To provide a clear sky measurement for each location in Canada, data from the years 2009, 2010, and 2011 were used, but 2010 was preferentially selected. Bands 3 (0.63-0.69 µm), 4 (0.76-0.90 µm), 5 (1.55-1.75 µm), and 7 (2.08-2.35 µm) are provided in this version as significant atmosphere effects strongly limit the quality of the blue (0.45-0.52 µm) and green (0.52-0.60 µm) bands. Multi-criteria compositing was used for the selection of the most representative pixel. For ETM+ onboard Landsat 7 a scan line malfunction caused missing lines of data in all scenes collected after May 2003. Atmosphere and target variability between scenes cause these lines to have significant radiometric differences in some cases. A Fourier transformation approach was applied to correct this occurrence. This mosaic was developed for land cover and biophysical mapping applications across Canada. Other applications of these data are also possible, but should consider the temporal and spectral limitations of the product. Research to enhance the spatial, spectral and temporal aspects are in development for future versions of moderate resolution products from historical Landsat sensors, Landsat 8, and Sentinel 2 data.

High-resolution false-color Landsat image composite of Canada's forested ecosystems (2022). This national image product represents the Composite to Change (C2C) proxy composite image derived from thousands of Landsat images acquired between July 1 and August 30, 2022. It is developed within the framework of Canada’s National Terrestrial Ecosystem Monitoring System (NTEMS). The overall process followed is described in (Hermosilla et al. 2016 ) with details on the generation of gap-free surface reflectance composites in ( Hermosilla et al. 2015). Following the motivation and rationale presented in White et al. (White et al. 2014), Landsat imagery is subjected to a series of processing steps to remove clouds and shadows as well as haze and other unwanted atmospheric effects. Year-on-year time series of Landsat imagery are interrogated to avoid missing values, and to ensure exhaustive spatial coverage of the national surface reflectance composites. False-colour 3-channel image (bands: shortwave infrared, SWIR1; near infrared; red)When using these data, please cite as: Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G.W. Hobart, L.B. Campbell, 2016. Mass data processing of time series Landsat imagery: pixels to data products for forest monitoring. International Journal of Digital Earth 9(11), 1035-1054 (Hermosilla et al. 2016 ).

These products are derived from RGB (red/green/blue) images, a satellite processing technique that uses a combination of satellite sensor bands (also called channels) and applies a red/green/blue (RGB) filter to each of them. The result is a false-color image, i.e. an image that does not correspond to what the human eye would see, but offers high contrast between different cloud types and surface features. The on-board sensor of a weather satellite obtains two basic types of information: visible light data (reflected light) reflecting off clouds and different surface types, also known as "reflectance", and infrared data (emitted radiation) which are short-wave and long-wave radiation emitted by clouds and surface features. RGBs are specially designed to combine this type of satellite data, resulting in an information-rich final product.Other products are based on the enhancement of channel data for a single wavelength, also aimed at highlighting meteorological features of the observed surface or clouds, but in a simpler way since only a single wavelength is involved. This older approach is still useful today, as its simplicity makes image interpretation easier in some cases.

These products are derived from RGB (red/green/blue) images, a satellite processing technique that uses a combination of satellite sensor bands (also called channels) and applies a red/green/blue (RGB) filter to each of them. The result is a false-color image, i.e. an image that does not correspond to what the human eye would see, but offers high contrast between different cloud types and surface features. The on-board sensor of a weather satellite obtains two basic types of information: visible light data (reflected light) reflecting off clouds and different surface types, also known as "reflectance", and infrared data (emitted radiation) which are long-wave radiations emitted by clouds and surface features. RGBs are specially designed to combine this type of satellite data, resulting in an information-rich final product. Four types of products are currently generated from the GOES-West and GOES-East satellites: "NightIR" and "NightMicrophysics", at 2km resolution, are generated 24 hours a day with infrared channels, so are visible both night and day, and "NaturalColour" and "DayCloudConvection", at 1km resolution, which combine visible light channels with infrared channels; their higher resolution makes the latter two products more popular, but they are not available during most of the night (between 02UTC and 07UTC for GOES-Est, and between 06UTC and 11UTC for GOES-Ouest) given the absence of reflected sunlight. Other RGB products should be added gradually in the future to meet different needs.

Each pixel value corresponds to the best quality maximum NDVI recorded within that pixel over the week specified. Poor quality pixel observations are removed from this product. Observations whose quality is degraded by snow cover, shadow, cloud, aerosols, and/or low sensor zenith angles are removed (and are assigned a value of “missing data”). In addition, negative Max-NDVI values, occurring where R reflectance > NIR reflectance, are considered non-vegetated and assigned a value of 0. This results in a Max-NDVI product that should (mostly) contain vegetation-covered pixels. Max-NDVI values are considered high quality and span a biomass gradient ranging from 0 (no/low biomass) to 1 (high biomass).

This dataset corresponds to daily snow cover percentage at 1km resolution grid over land areas of Canada from 2006-2010. The data are subsampled by 4km to reduce data volumes and considering the geolocation uncertainty of the input satellite imagery. The daily maps are generated by assimilation of daily cloud screened NOAA AVHRR satellite imagery and Canadian Meteorological Centre (CMC) snow depth analysis snow depth and density fields within an off-line version of the CMC daily snow depth model. The snow depth model is modified to include snowpack reflectance model and a surface radiative transfer scheme that relates vegetation and snowpack reflectance to top-of-canopy bi-directional reflectance. A logistic vegetation phenology model is used to parameterize temporal dynamics of canopy leaf area index. A per-pixel particle filter with a 30 day moving window is applied to assimilation observations corresponding to 1km resolution visible band directional reflectance and normalized difference vegetation index and 24km CMC daily snow depth and monthly snow density fields. The assimilation is forced using daily air temperature and precipitation fields. Validation of the datasets has been performed by comparison to MODIS snow cover maps and in-situ snow depth stations across Canada. Validation suggests similar accuracy to MODIS snow cover products over relatively flat terrain. Validation over mountainous regions is ongoing.

Wildfire change magnitude 1985-2022. Spectral change magnitude for wildfires that occurred from 1985 to 2022. It is developed within the framework of Canada’s National Terrestrial Ecosystem Monitoring System (NTEMS). The wildfire change magnitude included in this product is expressed via differenced Normalized Burn Ratio (dNBR), computed as the variation between the spectral values before and after a given change event. Higher dNBR values are related to higher burn severity. Time series of Landsat data with 30-m spatial resolution were used to characterize national trends in stand replacing forest disturbances caused by wildfire for the period 1985-2022 for Canada's 650 million-hectare forested ecosystems.When using this data, please cite as: Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G.W. Hobart, L.B. Campbell, 2016. Mass data processing of time series Landsat imagery: pixels to data products for forest monitoring. International Journal of Digital Earth 9(11), 1035-1054. (Hermosilla et al. 2016).See references below for an overview on the data processing, metric calculation, change attribution and time series change detection methods applied, as well as information on independent accuracy assessment of the data.Hermosilla, T., Wulder, M. A., White, J. C., Coops, N.C., Hobart, G.W., 2015. An integrated Landsat time series protocol for change detection and generation of annual gap-free surface reflectance composites. Remote Sensing of Environment 158, 220-234. (Hermosilla et al. 2015a).Hermosilla, T., Wulder, M.A., White, J.C., Coops, N.C., Hobart, G.W., 2015. Regional detection, characterization, and attribution of annual forest change from 1984 to 2012 using Landsat-derived time-series metrics. Remote Sensing of Environment 170, 121-132. (Hermosilla et al. 2015b).

Harvest changes occurred from 1985 to 2022 displaying the year of greatest harvest disturbance. It is developed within the framework of Canada’s National Terrestrial Ecosystem Monitoring System (NTEMS). The information outcomes represent 38 years of harvest activity in Canada's forests, derived from a single, consistent, spatially explicit data source in a fully automated manner. Time series of Landsat data with 30 m spatial resolution were used to characterize national trends in stand replacing forest disturbances caused by harvest for the period 1985-2022 for Canada's 650-million-hectare forested ecosystems.When using this data, please cite as: Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G.W. Hobart, L.B. Campbell, 2016. Mass data processing of time series Landsat imagery: pixels to data products for forest monitoring. International Journal of Digital Earth 9(11), 1035-1054. https://doi.org/10.1080/17538947.2016.1187673 ( Hermosilla et al. 2016).See references below for an overview on the data processing, metric calculation, change attribution, and time series change detection methods applied, as well as information on independent accuracy assessment of the data.Hermosilla, T., Wulder, M. A., White, J. C., Coops, N.C., Hobart, G.W., (2015). An integrated Landsat time series protocol for change detection and generation of annual gap-free surface reflectance composites. Remote Sensing of Environment 158, 220-234. ( Hermosilla et al. 2015a).Hermosilla, T., Wulder, M.A., White, J.C., Coops, N.C., Hobart, G.W., (2015). Regional detection, characterization, and attribution of annual forest change from 1984 to 2012 using Landsat-derived time-series metrics. Remote Sensing of Environment 170, 121-132. ( Hermosilla et al. 2015b).Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G. W. Hobart, (2017). Updating Landsat time series of surface-reflectance composites and forest change products with new observations. International Journal of Applied Earth Observation and Geoinformation. 63,104-111. https://doi.org/10.1016/j.jag.2017.07.013 (Hermosilla et al. 2017)

The Canadian long term satellite data record (LTDR) derived from 1-km resolution Advanced Very High Resolution Radiometer (AVHRR) data was produced by the Canada Center for Remote Sensing (CCRS). Processing included: geolocation, calibration, and compositing using Earth Observation Data Manager (Latifovic et al. 2005), cloud screening (Khlopenkov and Trishchenko, 2006), BRDF correction (Latifovic et. al., 2003), atmosphere and other corrections as described in Cihlar et. al. (2004). For temporal analysis of vegetation cross-sensor correction of Latifovic et al. (2012) is advised. Data collected by the AVHRR instrument on board the National Oceanic and Atmospheric Administration (NOAA) 9,11,14,16,17,18 and 19 satellites were used to generate Canada-wide 1-km 10-day AVHRR composites. Data are available starting in 1985. It is important to note that there are three types of AVHRR sensors: (i) AVHRR-1 flown onboard TIROS-N, NOAA-6, NOAA-8, and NOAA-10; (ii) AVHRR-2 flown onboard NOAA-7, NOAA-9, NOAA-11, NOAA-12, and NOAA-14; and (iii) AVHRR-3 currently operational onboard NOAA-15, NOAA-16, NOAA-17, NOAA-18 and NOAA-19. The AVHRR-1 has four channels, AVHRR-2 has five channels and the AVHRR-3 has six channels, although only five channels of AVHRR-3 can be operational at any one time. As such, channels 3A (1.6 m) and 3B (3.7 m) work interchangeably. The processing procedure was designed to minimize artefacts in AVHRR composite images. There are thirty six 10-day image composites per year. The following three processing levels are provided: P1) top of atmosphere reflectance and brightness temperature, P2) reflectance at surface and surface temperature and P3) reflectance at surface normalized to a common viewing geometry (BRDF normalization). The processing level P1 and P2 are provided for all 36 composites while level P3 is provided for 21 composites from April – October.

Wildfire change year 1985-2022.Wildfire changes occurred from 1985 to 2022 displaying the year of greatest wildfire disturbance. It is developed within the framework of Canada’s National Terrestrial Ecosystem Monitoring System (NTEMS). The information outcomes represent 38 years of wildfires in Canada's forests, derived from a single, consistent, spatially explicit data source in a fully automated manner. Time series of Landsat data with 30 m spatial resolution were used to characterize national trends in stand replacing forest disturbances caused by wildfire for the period 1985-2022 for Canada's 650-million-hectare forested ecosystems.When using this data, please cite as: Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G.W. Hobart, L.B. Campbell, 2016. Mass data processing of time series Landsat imagery: pixels to data products for forest monitoring. International Journal of Digital Earth 9(11), 1035-1054. https://doi.org/10.1080/17538947.2016.1187673 (Hermosilla et al. 2016).See references below for an overview on the data processing, metric calculation, change attribution, and time series change detection methods applied, as well as information on independent accuracy assessment of the data..Hermosilla, T., Wulder, M. A., White, J. C., Coops, N.C., Hobart, G.W., (2015). An integrated Landsat time series protocol for change detection and generation of annual gap-free surface reflectance composites. Remote Sensing of Environment 158, 220-234. https://doi.org/10.1016/j.rse.2014.11.005 (Hermosilla et al. 2015a).Hermosilla, T., Wulder, M.A., White, J.C., Coops, N.C., Hobart, G.W., (2015). Regional detection, characterization, and attribution of annual forest change from 1984 to 2012 using Landsat-derived time-series metrics. Remote Sensing of Environment 170, 121-132. https://doi.org/10.1016/j.rse.2015.09.004 (Hermosilla et al. 2015b).Hermosilla, T., M.A. Wulder, J.C. White, N.C. Coops, G. W. Hobart, (2017). Updating Landsat time series of surface-reflectance composites and forest change products with new observations. International Journal of Applied Earth Observation and Geoinformation. 63,104-111. https://doi.org/10.1016/j.jag.2017.07.013 (Hermosilla et al. 2017).