Chhun-Huor Ung

Canadian national biomass equations: new parameter estimates that include British Columbia data

National allometric equations covering the most common tree species of Canada’s forests were produced based on tree mass data acquired in the early 1980s during the ENergy from the FORest (ENFOR) program. The equations allow us to calculate the mass estimate of four tree components (foliage, branches, stem bark, and stem wood) using either diameter at breast height or a combination of diameter at breast height and height. Missing from that data set, however, were the data from British Columbia. A usable British Columbia data set was finally found and has now been incorporated into the national data set. Here, we present revised allometric equations for six species covered in the previous work and also found in the British Columbia data set as well as for the “hardwoods”, “softwoods”, and “all species” equations. New equations are also provided for eight species specific to the British Columbia data.

Impact of dominant tree dynamics on site index curves

Site index curves were modeled for two species of different shade tolerance, black spruce (Picea mariana (Mill.) B.S.P.) and jack pine (Pinus banksiana Lamb.), from an extended network of permanent sample plots (PSP) that covers periods of time varying from 10 to 30 years, in the province of Quebec. A data set reserved for validation allowed us to compare the site index curves derived from PSPs with published site index curves fitted to temporary sample plots (TSP) and stem analyses (SA). For both species, the site index curves calibrated from PSPs and TSPs behave similarly as they have comparable average bias and accuracy. The major difference is seen with the SA curves that strongly overpredict the dominant height growth of the PSPs. The similar pattern of change of site index curves calibrated from TSP and PSP data reinforces their validity as both types of curves were calibrated with independent data sets and methodologies. The differences observed between SA and PSP curves were likely produced by the dynamics of dominant height related to tree mortality and change in social status. For both species, approximately one tree out of five (22% for black spruce and 16% for jack pine) was replaced every 10 years in the tree group that was used to estimate dominant height. Consequently, the trajectory of dominant height through time for a particular plot is saw-toothed, the size of the “teeth” being, among other things, a function of stand regularity, as measured by an evenness index. Due to this tree replacement dynamic, stand dominant height curves are also more rapidly asymptotic than those of individual trees.

A variance-covariance structure to take into account repeated measurements and heteroscedasticity in growth modeling

This study proposes a within-subject variance-covariance (VC) structure to take into account repeated measurements and heteroscedasticity in a context of growth modeling. The VC structure integrates a variance function and a continuous autoregressive covariance structure. It was tested on a nonlinear growth model parameterized with data from permanent sample plots. Using a stand-level approach, basal area growth was independently modeled for red spruce (Picea rubens Sarg.) and balsam fir [Abies balsamea (L.) Mill.] in mixed stands. For both species, the implementation of the VC structure significantly improved the maximum likelihood of the model. In both cases, it efficiently accounted for heteroscedasticity and autocorrelation, since the normalized residuals no longer exhibited departures from the assumptions of independent error terms with homogeneous variances. Moreover, compared with traditional nonlinear least squares (NLS) models, models parameterized with this VC structure may generate more accurate predictions when prior information is available. This case study demonstrates that the implementation of a VC structure may provide parameter estimates that are consistent with asymptotically unbiased variances in a context of nonlinear growth modeling using a stand-level approach. Since the variances are no longer biased, the hypothesis tests performed on the estimates are valid when the number of observations is large.

Within crown variation in the relationship between foliage biomass and sapwood area in jack pine

The relationship between sapwood area and foliage biomass is the basis for a lot of research on eco-phyisology. In this paper, foliage biomass change between two consecutive whorls is studied, using different variations in the pipe model theory. Linear and non-linear mixed-effect models relating foliage differences to sapwood area increments were tested to take into account whorl location, with the best fit statistics supporting the non-linear formulation. The estimated value of the exponent is 0.5130, which is significantly different from 1, the expected value given by the pipe model theory. When applied to crown stem sapwood taper, the model indicates that foliage biomass distribution influences the foliage biomass to sapwood area at crown base ratio. This result is interpreted as being the consequence of differences in the turnover rates of sapwood and foliage. More importantly, the model explains previously reported trends in jack pine sapwood area at crown base to tree foliage biomass ratio.

Influence of shading on the relationship between leaf area and crown surface area in sugar maple stands

Abstract A functional relationship is proposed between leaf area, fractal dimension of the crown surface, shade tolerance and light availability. This relationship is obtained by setting an arbitrary limit between crown porosity to light and crown convolution. A crown profile is modeled from the total length and horizontal extension of second-order branches to estimate the crown surface area. A submodel simulates total canopy openness from geometrical characteristics of the crown of first-order neighbors to estimate light availability. To allow its insertion into growth and yield models, the model is simplified and validated with data gathered in six stands dominated by sugar maple ( Acer saccharum Marsh.), in southern Quebec.

Analytical estimation of branchwood volume in sugar maple, linked to branchiness

Abstract An analytical link is proposed between branchwood volume and branchiness. A segmented linear model with one parameter is used to describe the branch basal area density along the tree bole and integrated to find a function describing the cumulative branch basal area. It appears that the bases of insertion of the branches defining the base of the light crown correspond to the maximum branch basal area density along the bole. This function is then used together with an individual branch volume equation to find a model that estimates branchwood volume. This model is calibrated with data gathered in 15 stands dominated by sugar maple (Acer saccharum Marsh.) in southern Quebec. A comparison is made with other models of branchwood volume found in the literature.

Application of Near-Infrared Spectroscopy to Determine the Juvenile–Mature Wood Transition in Black Spruce

Abstract The potential of near-infrared spectroscopy (NIRS) to determine the transition from juvenile to mature wood in black spruce (Picea mariana (Mill.) B.S.P.) was assessed. In total, 127 wood ...

Cover type classification and biomass estimation by spectral analysis

This paper was written within the context of scaling up forest attributes, especially cover type and biomass, from ground plot inventory data to near infrared aerial photos. The material used is represented by ground plots georeferenced on aerial photos. Areas of 150 by 150 m are decomposed into proportions of spectrally distinct land cover elements: shadow and sunlit. Pielous non-randomness index is used as a surrogate of tree spatial distribution. Cover type is predicted with 71% of accuracy by blue, red, and green bands and by Pielous index. Meanwhile, only the blue and green bands are significant explanatory variables of biomass with a poor accuracy. More intensive use of photo texture information should improve the biomass prediction.

Estimating forest biomass using scale linkage from tree to Landsat-TM reflectance data

Estimates of forest biomass are needed to account for carbon at the tree, stand and regional scales. Sample plots of national forest inventories provide the basic database for these estimates. At the tree scale, a common estimation method is the use of an allometric equation that relates a trees predicted compartment biomass yi (i = foliage, branches, stem wood or stem bark) with easily obtained non-destructive measurements, i.e., diameter at breast height (D): yi=bi1Dbi2 or with both D and tree height (H): yi=bi1Dbi2Hbi3, bik being the parameters estimated. A common paradigm observed in biomass literature considers that parameter values vary between stands and regions. At the regional scale, however, when comparing national biomass equations to regional biomass equations, our results showed no significant differences between both types of equation. These results contribute to strengthening the allometric theory as an organizing principle for quantifying the relationship between tree size and biomass across spatial scales. In tandem with the allometry theory, we used a soil-canopy model based on Li-Strahlers approach for up-scaling biomass from the tree to stand scale in a mixed hardwood-coniferous forest. Our results indicated that the shadow fraction of Landsat TM reflectance was correlated with stand biomass. However, this model is indebted with heteroscedasticity, meaning that its error increases appreciably when stand biomass density is high.

Forest productivity assessment using remote sensing and GIS-based tools: test case for eastern Canada

Monitoring net primary productivity (NPP) and assessing site potential at the landscape level are central issues for sustainable landscape management practice. One of the foundations of classical forestry is the computation of forest productivity at the stand level. This computation of growth potential is usually based on a site index derived from measurements of height and age taken on dominant trees. The site index thus integrates the effect of the three main components of forest productivity: vegetation characteristics, climatic environment and site properties. However useful, site index-based methods cannot discern which part of the overall productivity is due to any of these three components. With an increasing demand for ensuring the sustainability of our forest practices, and with the uncertainty engendered by the possibility of climatic change, we now need tools to quantify the intrinsic productivity of a site, and the effect of external factors on this productivity. The ECOLEAP (extended concentration to link ecophysiology and forest productivity) project address these concerns. The overall objective of the ECOLEAP project is to develop a knowledge base from which tools to predict and monitor forest productivity and health will be developed. More specifically, the project will focus on three specific aspects. The first is the measurement of net primary productivity (NPP) in major eastern forest ecosystems, and, within each ecosystem, across a climate or fertility gradient. The second is the determination of functional relationships between specific biophysical factors, related to site, climate and species composition, and ecosystem NPP. The third is to develop tools to predict and monitor the sustainable potential NPP of major forest ecosystems under natural and managed scenarios. The three major forest types under study in ECOLEAP are the sugar maple, balsam fir and black spruce forests.