The Department of Natural Resources Canada (NRCan) has a requirement to characterize the composition and structure of Canada’s forests to support science and reporting activities. Over time the capacity for mapping forests has increased via the use of satellite imagery and high performance computing. NRCan requires technical investigations and prototyping to determine and demonstrate the emerging capacities for forest mapping. As such, it is required that a study of how time series satellite data can be used to improve national scale species mapping and forest age characterization is conducted. The work will involve the following two key investigations, one related to forest age, the other related to forest species. Forest age is understood will be mapped at the national level, with forest age denoted via time since disturbance. Prior to any mapping, scoping activity is required to determine what spatial data can be used to inform a time since disturbance (TSD) mapping activity. Once a review is complete, it will be required to determine how forest structure can be used to inform on possible forest age, time since last disturbance – as we expect that for trees of a given height or volume, using knowledge of growth and yield, it can be determined how long would it take (years) for the demonstrated values to be obtained. To enable the TSD mapping, remotely sensed data can be used for relating disturbances that occurred over the last ~40 years (at a spatial resolution of 30 m). From this data a logic can be developed and applied to TSD product that incorporates differing data inputs (available by time period). It is also expected that for some locations there will be no evidence of disturbance, therefore no TSD, but an age or maturity level can be related. It is anticipated that data and information use cases will be assessed to determine what is the appropriate product attribution to relay for locations that are old, but difficult to estimate with a high accuracy. The concepts of age and time since disturbance in forest monitoring require overview and summarization, from a forest monitoring context. All treed areas in Canada’s forest dominated ecosystems can be provided with an estimate of age via TSD. Based upon the review and product development, it will be possible to make recommendations towards possible improvements to age (TSD) mapping for forest ecosystems.National monitoring, science, and reporting activities also require improve spatial mapping of forest species. For instance, species is an important element of both jurisdictional strategic forest inventories as well as Canada’s National Forest Inventory (NFI). It is increasingly understood that cloud computing and modern machine learning algorithms provide new opportunities for spatial mapping of forest species. In support of national aims, NFI data is available for calibration and validation. Making the distinction between strategic and operational forest inventories will be important element of quality assessment. Strategic inventory use cases to be consulted to ensure proper targeting on the information content of remote sensing derived species maps. Understanding the quality and uses of current strategic forest inventory species information will assist in understanding utility of actual inventory outcomes, realized not specified outcomes. Further, the capacity to use species class likelihoods to create assemblages will be investigated. Elements of spatial representation, temporal latency in product development, capture of local detail, integration time series information to refine species outcomes also to be investigated. Ultimately, this activity has two key elements: 1. reality of species in strategic forest inventories, setting content for remote sensing mapping of species; 2. prototypically mapping of species at national level (forested ecosystems). The context setting activity (critical review) will inform the mapping activity to be implemented.
