|Forest Investment Account (FIA) - Forest Science Program|
|FIA Project Y092171|
|Equivalent clear cut area thresholds in large-scale disturbed forests|
|Project lead: Weiler, Markus (University of British Columbia)|
|Contributing Authors: Weiler, Markus; Coops, Nicholas C.; Teti, Patrick; Boon, Sarah; Varhola, Andres; Bater, Christopher W.|
|Subject: Forest Investment Account (FIA), British Columbia|
|Series: Forest Investment Account (FIA) - Forest Science Program|
|Forest canopy change can be caused by large-scale disturbance such as insect infestation, fire and forest harvesting. These changes alter the vegetation cover and ecological integrity of watersheds, and have serious impacts on environmental values, as forested vegetation is critical for terrain stability, riparian areas, water quality and quantity. Changes in forest structure following disturbance are poorly understood and quantified. Each type of disturbance leaves different canopy complexity; for example, forest structure following infestation is markedly different than that remaining after harvesting, as the remaining canopy structure is a complex mixture of standing and fallen dead and live trees.|
Many hydrologic processes are affected by these changes in forest structure. These include precipitation interception and ground shading. Interception describes the process of temporal storage of precipitation in the tree canopy, from where it either evaporates/sublimates or falls to the ground. Since interception is an important component of the water balance, deteriorating stand structure and timber harvesting are expected to increase annual and storm runoff. In watersheds with strong snowmelt regimes, the seasonal effects of large scale canopy disturbance are enhanced by the influence of canopies on wind- and radiation-related ablation losses. The loss of canopy cover and changes in structure also alter the energy balance during snow accumulation and melt, thus affecting the snow melt runoff regime. Initial results from research in this area indicates that while changes in stand structure are a major determinant of ground snow accumulation and melt rates in healthy and disturbed stands, these complex canopy responses have yet to be adequately quantified as noted in proposals by Boon and Teti.
The concept of equivalent clear-cut area (ECA) is frequently used as a measure of cumulative disturbance in a watershed. However, the ECA method has not yet been adapted for areas affected by large-scale canopy disturbance. Several stand level studies are underway (see project linkages) that compare stands with different canopy structure; however, none of these studies provides enough information to determine new guidelines for ECA calculation over larger areas. Preliminary results suggest that the percentage of disturbed trees, dead tree clustering, terrain attributes, tree size and density are the greatest factors in altering interception and snow melt, and hence the calculation of ECA in regions with large-scale forest disturbance.
A number of detailed stand-level models have been developed which attempt to quantify snow or radiation interception expected over a range of forest canopies or canopy clumping (e.g. Hedstrom and Pomeroy 1998, Chen et al, 1993, Spittlehouse et al. 2004). However, model parameterization is a major issue and is not feasible over large areas. These models also have focused on undisturbed forest stands, hence limiting our ability to predict these processes in a partly or completely disturbed forest stand. In addition, as interception and snow melt studies have historically been undertaken only in small plots, extrapolation to larger areas remains a limiting factor. As a result, the forest industry and forest hydrologists have neither the empirical data, nor the modeling output, to provide a detailed understanding of the direct linkages between canopy structure and interception/shading with respect to large-scale forest disturbance.
Snow interception strongly depends on canopy density (the ratio of canopy-covered area per unit area of ground), which is related to LAI, but also on the larger topographical settings (e.g. aspect, slope), tree species or patchiness of the forest (Pomeroy et al, 2002). Simple two-stream canopy radiation models have been successfully used to predict direct and diffuse radiation below a relative homogenous forest canopy (e.g. Chen et al. 1993), however, in sparse or discontinuous forests, which are common in disturbed forests, the forest structure information has to be taken into account to predict radiative fluxes (Essery et al, 2005). These tree and stand level parameters are difficult and very time consuming to measure on site, but remote sensing technology provides a mean to characterize these attributes over a large area.
Therefore, an alternative approach to individual tree- or stand-based modeling is one that utilizes remote sensing technology to predict the structural attributes of healthy and disturbed stands, and combines this data with field measurements of snow accumulation and melt (continuous stand-level measurements are currently collected by Boon and Teti in the geographical region of interest). Applying this approach along transects provides a basis for new datasets and models for watershed assessment in BC following large-scale disturbance. Sampling along a several hundred kilometer long transect has the advantage over a watershed study, by permitting a much wider range and a distribution of stand ages, species compositions, disturbance regimes, and terrain attributes that are more typical of the whole area of interest (see also Figure 2). Using this approach, the remotely derived parameter that statistically explains differences in snow accumulation and melt can finally be related to data from forest cover maps to derive new guidelines for ECA calculation in large-scale disturbed areas.
Our proposed approach will utilize two key remote sensing technologies: LiDAR and high resolution aerial photography. The application of airborne LiDAR data offers the potential to significantly enhance the timeliness, scope, and rigor of forest measurement information. LiDAR technology offers an innovative method to assess forest structure by measuring the height of the canopy, underlying terrain morphology, distribution of foliage, tree clumping and other metrics of crown architecture (Lim et al, 2003). Conceptually, the interaction of the laser pulse with the forest canopy is similar to the transmission of precipitation or a sun beam through the canopy, thus linking LiDAR with interception and snow melt is a logical step. In addition, we will use very high spatial resolution digital aerial photography under cloud to provide non-shadowed imagery in which crowns, logs, gaps between crowns, and crown density can be discerned. This imagery is currently being used by Teti as part of his ongoing interception studies in the interior of BC.
|Related projects:  FSP_Y081171|
|Executive summary (81Kb)|
To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.
Updated August 16, 2010
Please direct questions or comments regarding publications to For.Prodres@gov.bc.ca