Search Results

Lac St. Jean Research Forest Lac St. Jean is a 20,607 square kilometres boreal forest hub site located in Quebec, north of Quebec City. To gain access to more raw data, please contact Patrice Boucher with the Government of Quebec. [Click on any image to magnify] Site Details Climate LiDAR Products Vector Products Forest Monitoring Anchor 1 Site Details Best Available Pixel (2020) Lac St. Jean Hub Site Boundary Best Available Pixel (BAP) composites use Landsat scenes to develop cloud-free, surface reflectance pixel-based image composites capable of large-area production. When incorporated in a time series, they generate land cover, land cover change, and forest structural attributes information products in a dynamic, transparent, systematic, repeatable, and spatially exhaustive manner. This figure displays the 2020 BAP composite within the Haliburton hub site in Ontario. The acquisition of all pixels for this BAP composite were within 30 days of the first of August, 2020. Lac St. Jean Digital Elevation Model (DEM) at 30 m resolution Lac St. Jean Digital Elevation Model (DEM) at 250 m resolution Anchor 2 Climate Projections for change in minimum temperature for the years 2050 and 2090 relative to the reference period (1981-2010) Projections for change in maximum temperature for the years 2050 and 2090 relative to the reference period (1981-2010) Projections for change in seasonal precipitation for the years 2050 and 2090 relative to the reference period (1981-2010) Climate data for historical (1981-2020) and future (2050 and 2090) projections Anchor 3 LiDAR Derived Products Digital Elevation Model (1m) Description - An interpolation of last returns classified as 'ground' points using TIN. Pixel Values - Elevation at a 1 metre resolution. Forestry Application - The Digital Elevation Model is important for topographical information, including slope, aspect, and radiation Radiation (1m) Description - A technique to visualize a shaded relief, illuminating it with a hypothetical light Pixel Values - Raster containing a numeric value representing the solar-radiation aspect index. Values range from 0 (land oriented in a northern direction resulting in less solar radiation exposure) and 1 (land oriented with southern slopes) Forestry Application - Land oriented in northerly (values closer to zero) are typically wetter and cooler compared to land oriented in southerly (values closer to one) Slope (1m) Description - Steepness or the degree of incline of a surface based on the DEM model Pixel Values - Raster containing numeric values representing degrees of incline Forestry Application - Slope influences tree stability, harvesting solutions, productivity of harvesting and collection means, architecture of the road networks, road characteristics, and solutions related to the reclamation of degraded forested land. 20th Height Percentile (20m) Description - Height at which 20% of LiDAR returns fall below from 2m above the ground Pixel Values - Height measurement in metres describing the height at which 20% of LiDAR returns fall below Forestry Application - Assists in determining the height and distribution of the lower section of the canopy Aspect (1m) Description - Orientation of slope, measured clockwise in radians based on the DEM model Pixel Values - Raster containing a numeric value representing the radians of orientation Forestry Application - Land oriented in northerly are typically wetter and cooler compared to land oriented in southerly 95th Height Percentile (20m) Description - Height at which 95% of LiDAR returns fall below from 2m above the ground Pixel Values - Height measurement in metres describing the height at which 95% of LiDAR returns fall below Forestry Application - Assists in determining the height and distribution of the lower section of the canopy Canopy Height Model (1m) Description - Based on an interpolation of the height of the top of trees (using the pitfree algorithm) Pixel Values - Raster containing a numeric value for the distance between the ground and the top of trees Forestry Application - Helpful for determining the distribution of canopy coverage Entropy (20m) Description - Shannon entropy quantifies the diversity and evenness of an elevation distribution of LiDAR points from 2m above the ground Pixel Values - Entropy results range from 0 to 1. Random data has a Shannon entropy value of 1 Forestry Application - Useful for describing and quantifying species diversity in biological systems. Canopy Cover > 2m (20m) Description - Canopy cover at a height greater than 2 metres Pixel Values - Ratio from the sum of first returns > 2 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 2 metres LAIe (20m) Description - A measurement of the gap fraction through the probability of beam penetration of sunlight through the vegetation. Pixel Values - Ratio of one-sided green leaf area per unit ground surface area Forestry Application - Important growth index for the status of crop populations Mean Height (20m) Description - Mean height of first returns above 2m from "ground" (last return data) Pixel Values - Mean height of all point cloud returns greater than 2m above last returns Forestry Application - Determines the mean height of all objects (trees) that are at least 2m tall Mean Standard Deviation (20m) Description - Standard Deviation of height distributions above 2m Pixel Values - Standard deviation height of all point cloud returns greater than 2m above last returns Forestry Application - Determines the standard deviation for the height of all objects (trees) that are at least 2m tall Skewness (20m) Description - A measure of the distribution's symmetry from 2m above the ground Pixel Values - A normal distribution would produce skewness results of zero. Negative values indicate that data is skewed to the left, and positive values indicate that data is skewed to the right. Forestry Application - Skewness is often used with kurtosis to separate ground points and object points from a LiDAR point cloud. It has a variety of applications, including optimizing the DEM, segmentation and classification, and road extraction. Canopy Cover > 5m (20m) Description - Canopy cover at a height greater than 5 metres Pixel Values - Ratio from the sum of first returns > 5 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 5 metres Kurtosis (20m) Description - The size of the tails of a distribution (likelihood that the distribution will produce outliers) from 2m above the ground Pixel Values - A normal distribution would produce kurtosis results of 3. Distributions with kurtosis less than 3 are platykurtic (fewer and less extreme outliers) and distributions with kurtosis greater than 3 are laptokurtic (produce more outliers) Forestry Application - Kurtosis is often used with skewness to separate ground points and object points from a LiDAR point cloud. It has a variety of applications, including optimizing the DEM, segmentation and classification, and road extraction. Canopy Cover > 15m (20m) Description - Canopy cover at a height greater than 15 metres Pixel Values - Ratio from the sum of first returns > 15 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 15 metres Rumple (20m) Description - Crown Surface Roughness from 2m above the ground Pixel Values - A ratio of canopy outer surface area to ground surface area Forestry Application - Higher rumple values indicate more vertical and horizontal heterogeneity Anchor 4 Vector Products

< Retour Optimization of the characterization of burning patterns OB 1b Hub: Lac-St-Jean, QC Year: 2021-2022 Gabrielle Thibault, MSc Despite being less frequent in the East, fires remain the main cause of natural disturbances in the boreal forests of Quebec (QC) and Ontario (ON) and regeneration success is highly variable in forests dominated by black spruce. An MSc student will collaborate with the PhD student (OB.1a) and adapt the workflow to the LSJ and RM boreal Hubs. Using metrics of natural stand establishment that characterize failures or successes, the objective will be to improve our understanding the ecological factors that lead to black spruce regeneration after fire disturbance. Outcome (OB.1b): Identification of key stand- and landscape-level characteristics that favour regeneration success and improve forest resilience. Gabrielle Thibault, MSc at Université Laval Main Partner: Ministère des Forêts, de la Faune et des Parcs (Québec) Professor Alexis Achim ​

< Retour Seasonal mosaics of forest cover OB 3a Hub: All Year: 2021-2023 Micheal Burnett, RA There is a critical need for accurate and timely monitoring for near real time forest condition assessment. The increasing availability of moderate and high-resolution imagery, with daily or weekly repeat, is enabling the regular characterization of entire forest areas. Micheal Burnett will apply and adapt approaches to image fusion and compositing to Sentinel-2, Landsat and Planet multispectral data to produce seasonal composites over large forest areas. Outcome (OB.3a): For each Hub site, an operational coding framework on how to derive seasonal mosaics of forest cover and condition and implement a change detection approach. Micheal Burnett, RA at University of British Columbia Main Partner: Kruger Inc. Professor Nicholas Coops Collaborator Alexis Achim ​

< Retour Silvicultural practices at the pace of global changes: a public policy challenge AD 8a Hub: All Year: 2022-2023 Anne Bernard, PDF In the face of global changes, the scientific community is actively developing solutions to respond to the pressures on forest ecosystems and, more broadly, on territories. In order to be implemented, the innovative silvicultural practices that result from these research initiatives will have to integrate the various legislative frameworks that govern forestry practices. In Canada, this involves the provincial and federal levels. In this research project, we will attempt to answer the following question: How can new forestry practices be linked to global changes? More specifically, the research objective is to understand the regulatory and social context of forestry practices in Canada. To achieve this, an analysis of public policies in six provinces will be conducted. We will also look at the influence of international organizations on national and regional forest policies. This research is part of the Silva21 partnership, which includes 11 study areas in six provinces (British Columbia, Ontario, Quebec, New Brunswick, Nova Scotia and Newfoundland-Labrador). For these areas, we will develop different socio-economic portraits for each forest case study. The methodologies envisioned to meet the project's objectives are part of public policy analysis. The expected results will take various forms. First, it is expected to produce scientific articles for Canadian forestry professionals (approximately one per province or per region: West, Ontario-Quebec and Maritimes). The case studies will document local realities for use by all Silva 21 members to present the social and political context of the study areas. Finally, one or two scientific articles will be developed for the scientific community specialized in public policy and/or forest governance. Anne Bernard, PDF at Université Laval Main Partner: Ministère des Forêts, de la Faune et des Parcs (Québec) Professor Alexis Achim Collaborator Maude Flamand-Hubert Canuel, C-M., Bernard, A., Thiffault, N., et al. (2022). Analysis of a wood production strategy from expert perspectives. The Forestry Chronicle, 98(1): 1-9. https://pubs.cif-ifc.org/doi/pdf/10.5558/tfc2022-004 Bernard, A., Gélinas, N. (2022) La multifonctionnalité du territoire forestier dans tous ses états. In: Une économie écologique pour le Québec: Comment opérationnaliser une nécessaire transition, PUQ.

< Retour Identifying and assessing the impact of non-stand-replacing disturbances within a near- real-time context OB 5c Hub: Quesnel Year: 2023-2024 Madison Brown, M.Sc. Once a robust continuous forest inventory framework has been developed (OB 5a), we will implement the proposed methodologies through two MSc projects at the Quesnel hub site. This third MSc student will have the objective to apply the framework and develop tools to assess forest stand stresses, such as infestation, wind damage or disease. The MSc student will use a series of indicators of forest health derived at the 20 x 20 m resolution and assess which predicted metrics are sufficiently reliable to be considered into silvicultural prescriptions for a range of stand ages and composition. Outcome (OB.5c): Recommendations of a series of metrics that can be derived at a 30 x 30 m resolution to guide silvicultural interventions dependent on stand condition. Madison Brown, MSc at University of British Columbia Main Partner: Department of Energy and Resource Development (New Brunswick) Professor Nicholas Coops ​

PUBLICATIONS Infolettres Extension Notes Publications

< Retour From theory to action at the Montmorency Forest OB 3c Hub: Montmorency Forest Year: 2023-2024 Recruiting, RA Recruiting (RA) at Université Laval Main Partner: Ministère des Forêts, de la Faune et des Parcs (Québec) Professor Alexis Achim ​

Romeo Malette Research Forest Romeo Malette Forest is a 6,897 square kilometres boreal forest hub site located in northern Ontario, north of Lake Huron and near Timmins. To gain access to more raw data, please contact Dave Morris, and Chris Stratton with the Government of Ontario. [Click on any image to magnify] Site Details Climate LiDAR Products Vector Products Forest Monitoring Anchor 1 Site Details Best Available Pixel (2020) Romeo Malette Hub Site Boundary Best Available Pixel (BAP) composites use Landsat scenes to develop cloud-free, surface reflectance pixel-based image composites capable of large-area production. When incorporated in a time series, they generate land cover, land cover change, and forest structural attributes information products in a dynamic, transparent, systematic, repeatable, and spatially exhaustive manner. This figure displays the 2020 BAP composite within the Haliburton hub site in Ontario. The acquisition of all pixels for this BAP composite were within 30 days of the first of August, 2020. Romeo Malette Digital Elevation Model (DEM) at 30 m resolution Romeo Malette Digital Elevation Model (DEM) at 250 m resolution Anchor 2 Climate Projections for change in minimum temperature for the years 2050 and 2090 relative to the reference period (1981-2010) Projections for change in maximum temperature for the years 2050 and 2090 relative to the reference period (1981-2010) Projections for change in seasonal precipitation for the years 2050 and 2090 relative to the reference period (1981-2010) Climate data for historical (1981-2010) and future (2050 and 2090) projections Anchor 3 LiDAR Derived Products Digital Elevation Model (1m) Description - An interpolation of last returns classified as 'ground' points using TIN. Pixel Values - Elevation at a 1 metre resolution. Forestry Application - The Digital Elevation Model is important for topographical information, including slope, aspect, and radiation Radiation (1m) Description - A technique to visualize a shaded relief, illuminating it with a hypothetical light Pixel Values - Raster containing a numeric value representing the solar-radiation aspect index. Values range from 0 (land oriented in a northern direction resulting in less solar radiation exposure) and 1 (land oriented with southern slopes) Forestry Application - Land oriented in northerly (values closer to zero) are typically wetter and cooler compared to land oriented in southerly (values closer to one) Slope (1m) Description - Steepness or the degree of incline of a surface based on the DEM model Pixel Values - Raster containing numeric values representing degrees of incline Forestry Application - Slope influences tree stability, harvesting solutions, productivity of harvesting and collection means, architecture of the road networks, road characteristics, and solutions related to the reclamation of degraded forested land. 20th Height Percentile (20m) Description - Height at which 20% of LiDAR returns fall below from 2m above the ground Pixel Values - Height measurement in metres describing the height at which 20% of LiDAR returns fall below Forestry Application - Assists in determining the height and distribution of the lower section of the canopy Aspect (1m) Description - Orientation of slope, measured clockwise in radians based on the DEM model Pixel Values - Raster containing a numeric value representing the radians of orientation Forestry Application - Land oriented in northerly are typically wetter and cooler compared to land oriented in southerly 95th Height Percentile (20m) Description - Height at which 95% of LiDAR returns fall below from 2m above the ground Pixel Values - Height measurement in metres describing the height at which 95% of LiDAR returns fall below Forestry Application - Assists in determining the height and distribution of the lower section of the canopy Canopy Height Model (1m) Description - Based on an interpolation of the height of the top of trees (using the pitfree algorithm) Pixel Values - Raster containing a numeric value for the distance between the ground and the top of trees Forestry Application - Helpful for determining the distribution of canopy coverage Entropy (20m) Description - Shannon entropy quantifies the diversity and evenness of an elevation distribution of LiDAR points from 2m above the ground Pixel Values - Entropy results range from 0 to 1. Random data has a Shannon entropy value of 1 Forestry Application - Useful for describing and quantifying species diversity in biological systems. Canopy Cover > 2m (20m) Description - Canopy cover at a height greater than 2 metres Pixel Values - Ratio from the sum of first returns > 2 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 2 metres Mean Height (20m) Description - Mean height of first returns above 2m from "ground" (last return data) Pixel Values - Mean height of all point cloud returns greater than 2m above last returns Forestry Application - Determines the mean height of all objects (trees) that are at least 2m tall Mean Standard Deviation (20m) Description - Standard Deviation of height distributions above 2m Pixel Values - Standard deviation height of all point cloud returns greater than 2m above last returns Forestry Application - Determines the standard deviation for the height of all objects (trees) that are at least 2m tall Skewness (20m) Description - A measure of the distribution's symmetry from 2m above the ground Pixel Values - A normal distribution would produce skewness results of zero. Negative values indicate that data is skewed to the left, and positive values indicate that data is skewed to the right. Forestry Application - Skewness is often used with kurtosis to separate ground points and object points from a LiDAR point cloud. It has a variety of applications, including optimizing the DEM, segmentation and classification, and road extraction. Canopy Cover > 5m (20m) Description - Canopy cover at a height greater than 5 metres Pixel Values - Ratio from the sum of first returns > 5 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 5 metres Kurtosis (20m) Description - The size of the tails of a distribution (likelihood that the distribution will produce outliers) from 2m above the ground Pixel Values - A normal distribution would produce kurtosis results of 3. Distributions with kurtosis less than 3 are platykurtic (fewer and less extreme outliers) and distributions with kurtosis greater than 3 are laptokurtic (produce more outliers) Forestry Application - Kurtosis is often used with skewness to separate ground points and object points from a LiDAR point cloud. It has a variety of applications, including optimizing the DEM, segmentation and classification, and road extraction. Canopy Cover > 15m (20m) Description - Canopy cover at a height greater than 15 metres Pixel Values - Ratio from the sum of first returns > 15 metres divided by the total first returns Forestry Application - Important for determining the area occupied by the vertical projection of tree crowns greater than 15 metres LAIe (20m) Description - A measurement of the gap fraction through the probability of beam penetration of sunlight through the vegetation. Pixel Values - Ratio of one-sided green leaf area per unit ground surface area Forestry Application - Important growth index for the status of crop populations Rumple (20m) Description - Crown Surface Roughness from 2m above the ground Pixel Values - A ratio of canopy outer surface area to ground surface area Forestry Application - Higher rumple values indicate more vertical and horizontal heterogeneity Anchor 4 Vector Products

< Retour Advanced RS: free-to-grow to thinning stage OB 2 Hub: Romeo Malette, ON; Quesnel, BC Year: 2021-2025 Liam Irwin, PhD Recent advances in high spatial resolution imaging with drone based lidar can provide accurate height and structural estimates of regenerating vegetation in a repeatable way. For these methodologies to be implemented in commercial operations, solutions need to be developed for the complex conditions observed in Canadian stands past the ‘free-to-grow’ stage. The objective of this PhD will be to determine how to best use these newly available data sources to monitor stand development between the free-to-grow stage and the time of commercial thinning operations at the operational stand levels. Focussing on spruce and pine stands in the RMF, and QN Hub sites, drone surveys will be conducted across age cohorts of varying productivity and silvicultural origins. Outcome: New workflows to process drone based lidar for mid-rotation stand assessment, assessment of detection capacity for species, height, and density, identification of areas where silvicultural interventions could help achieve growth potential on a site. Liam Irwin, PhD at University of British Columbia Main Partner: Ministry of Natural Resources and Forestry (Ontario) Professor Nicholas Coops Collaborators: Doug Reid (MNRF-ON) Irwin, K., Coops, N.C., Riofrio, J., Barbeito, I., Gubringer, S., Achim, A., Roeser, D. (2023) Quantifying tree-level drivers of tree growth in high density managed stands with drone lidar. SilviLaser Conference Presentation. 6 - 8 September, UCL, London, UK. Irwin, L., Coops, N.C. , Roussel, J.R., Achim, A. (2022) Evaluating fine-scale forest competition and structural assessments with drone based lidar; A case study in dense Canadian operational conditions before and after thinning treatments. ForestSAT Conference Presentation, 30 august, Berlin, Germany Irwin, L., Coops, C.C., Queinnec, M., McCartney, G., White, J. (2021) Single photon lidar signal attenuation under boreal forest conditions. Remote Sensing Letters, 12(10). https://doi.org/10.1080/2150704X.2021.1962575

< Retour Climatic drivers of tree growth AN 1a Hub: All Year: 2021 Catherine Chagnon, RA Understanding future climate dynamics is critically important to understand how adaptive management approaches can be developed and applied. However, adjusting growth projections to the new climate reality first requires identification of the key climatic variables that alter tree growth. Using a dendrochronological approach, Catherine Chagnon will utilise existing tree core databases, collected across Canadian forests, as well as new core information from Hub sites with the objective to identify the climatic events and conditions most susceptible to alter tree growth. The analysis includes the effects of both short, acute climatic events like drought, and more monotonic increases in average temperatures. Outcome (AN.1a): Identification and definition of key climate variables that affect tree growth across Canada. Catherine Chagnon, RA at Université Laval Main Partner: Ministère des Forêts, de la Faune et des Parcs (Québec) Professor Alexis Achim Leduc, F., Chagnon, C., Moreau, G., Dumont, S., St-Jean, É., Achim, A. (2023) American beech outgrows sugar maple at the sapling stage regardless of partial harvest intensity in northern hardwood forests. Forest Ecology and Management, 533: 121630, https://doi.org/10.1016/j.foreco.2023.121630 Chagnon, C., Wotherspoon, A.R., Achim, A. (2022) Deciphering the black spruce response to climate variation across eastern Canada using a meta-analysis approach. Forest Ecology and Management, 520: 120375. https://doi.org/10.1016/j.foreco.2022.120375 Moreau, G., Chagnon, C., Achim, A., et al. (2022). Opportunities and limitations of thinning to increase resistance and resilience of trees and forests to global change. Forestry, 1-21. https://doi.org/10.1093/forestry/cpac010. Chagnon, C., Moreau, G., Bombardier-Cauffopé, C., Barrette, J., Havreljuk, F., Achim, A. (2022). Broad-scale wood degradation dynamics in the face of climate change: A meta-analysis. GCB-Bioenergy, 14(8): 941-958. https://doi.org/10.1111/gcbb.12951 Chagnon C, Moreau G, D’Orangeville L, Caspersen J, Labrecque-Foy J-P and Achim A (2023) Strong latitudinal gradient in temperature-growth coupling near the treeline of the Canadian subarctic forest. Front. For. Glob. Change 6:1181653. doi: 10.3389/ffgc.2023.1181653 Moreau, G., Chagnon, C., Cecil-Cockwell, MFL, Pothier, D., Achim, A., Bédard, S., Guillemette, F., Caspersen, J. (2023) Simplified tree marking guidelines enhance value recovery as well as stand vigour in northern hardwood forests under selection management. Forestry: An International Journal of Forest Research; cpad045. https://doi.org/10.1093/forestry/cpad045

< Retour Integration of climate drivers into growth modelling (AN3a) AN 3a Hub: Eastern Townships, QC Year: 2022-2025 Christina Howard, PhD Diversifying the species composition and structure of forest stands is often proposed as a solution to increase the resilience and resistance of forests to stressors. Yet, most existing models to predict forest growth and stand dynamics in Canada have been developed for even-aged, single-species stands. Christina Howard, a PhD student, will focus on the Eastern Townships hardwood site with the objective to integrate climate drivers into the model framework of existing forest growth and yield models from the Prognosis/FVS family. This will improve our capacity to anticipate the effects of interactions between climate, stand composition, and structure on growth. Model predictions will be verified using stand structure and growth rate combinations highlighted through the big data analysis undertaken in OB.7. Outcome: New model formulations for mixed-species, multi-cohort stands available via open source platforms to allow use and testing at additional sites. Christina Howard, PhD at University of British Columbia Main Partner: Ministère des Forêts, de la Faune et des Parcs (Québec) Professor Bianca Eskelson, UBC ​

< Retour Using state-of-the-art technology to achieve multiple forest management objectives (AD3b) AD 3b Hub: Quesnel, BC Year: 2023-2024 Mario Stolz, MSc As we actively manage recently disturbed sites within the QN Hub site and surrounds for the next generation, our objective will be to determine how can species selection and modified regeneration/stocking levels be used to effectively attenuate wildfire risk, and assess potential impacts on timber and non-timber objectives. A PhD student will assess a number of existing trials already underway examining stand establishment following the 2017 fires and assess their future value through both data collection and modelling, the future uses of the regenerating species mix with respect to commercial and more innovative non-timber products and values. Outcome (AD.3b): Identification of which species, and at what density stands should be re-established to ensure the regeneration of resistant and valuable stands from both timber and non timber perspectives. Mario Stolz, MSc at University of British Columbia Main Partners: Future of Forestry Think Tank (British Columbia) Professor Dominik Roeser ​