Session 16 ic2011 antony
1 like177 views
This document describes a study that compares three methods for delineating earlywood and latewood within tree rings: Mork's index method, the threshold density method, and the maximum derivative method. The study used density, tracheid diameter, and wall thickness measurements from radial strips of loblolly pine to apply the three demarcation methods and compare how well they agreed. The objective was to evaluate which method most accurately identifies the transition from earlywood to latewood.
More Related Content
Session 16 ic2011 antony
- 1. 8/2/2011 Introduction Density is an important measure of wood quality A comparison of earlywood-latewood demarcation methods within an annual ring – A case study in Large variation in wood density has been observed in loblolly pine many tropical and temperate tree species Within-ring variation attributed to the formation of F. Antony , L.R. Schimleck & R.F. Daniels earlywood (EW) and latewood (LW) within a ring University of Georgia Athens, GA Measure of within-ring variability in density can be used as an indicator of wood uniformity Introduction Introduction Within-ring density is expressed as an average of EW and LW density The threshold method is frequently used to identify the transition from EW to LW in loblolly pine EW and LW density and their corresponding widths within a ring depends on the definition(s) used by researchers i d d h d fi i i ( ) db h How well it agrees with the other methods? Several methods have been used to identify the The objective of this study was to compare the three demarcation between EW and LW (e.g. Mork’s index, different methods for EW-LW demarcation; Mork’s Index, threshold method and the maximum derivative method). the threshold method and the inflection point method 1
- 2. 8/2/2011 Materials and Methods Materials and Methods Basic Density (kg m ) 3 20-radial strips (2 mm tangentially and 7 mm 800 longitudinally) from breast height increment cores 200 10 20 30 40 50 60 70 Distance from pith (mm) Radial strips were analyzed using Silviscan® Radial Diame (m) Air-dry density (AD), eter 35 5 Tracheid wall thickness (WT), 20 Tracheid radial and tangential diameters (RD and TD) 10 20 30 40 50 60 70 Distance from pith (mm) All properties measured at a resolution of 0.05 mm WT= P/8 - ½ (P2/16 – C/d)½ Wall Thickenss (m) • P (perimeter) = 2(RD + TD) 6 • C (coarseness) = RD X TD X AD 4 2 • d = 1500 kg/m3 for all softwoods 10 20 30 40 50 60 70 Distance from pith (mm) Materials and Methods Materials and Methods According to Mork (1928) a tracheid is said to be a Juvenile Wood Ring latewood tracheid if Morks index 1.0 • 2 X WT >= Lumen Diameter (LD) 0.4 • 2 X (2 X WT) >= LD • where LD = RD – 2 WT 10 12 14 16 Distance from pith (mm) Following the second definition of Mork, we computed Mature wood Ring Mork’s Index (MI) Morks index 2 x (2 x WT) 3 MI = LD 1 Transition from EW to LW is defined as the point within a 69.5 70.0 70.5 71.0 71.5 72.0 ring where MI 1 Distance from pith (mm) 2
- 3. 8/2/2011 Materials and Methods Materials and Methods Basic density (BD) at each radial position using the AD data was computed as follows 800 Ring 20 Ring 21 700 BD = -22.0401 + 0.8902 AD nsity (kg m ) 3 600 BD value of 480 kg/m3 was used in the threshold method Basic Den for defining the EW to LW transition 500 3 480 kg m 400 It approximately corresponds to the mean BD of the zone 300 in which earlywood rapidly transitions to latewood in loblolly pine 106 107 108 109 110 111 Distance from pith (mm) Materials and Methods Materials and Methods The derivative method was based on identifying the Juvenile Wood Ring Mature Wood Ring inflection point for the density profile of each individual ring Basic Density (kg m ) 1000 3 700 600 500 A point on the curve is said to be the inflection point where 300 the second derivative of a function fitted to the curve will be 200 equal to zero 10 12 14 16 69.5 70.0 70.5 71.0 71.5 72.0 Smooth splines were fitted to individual ring density profiles (5df as smoothing parameter) 100 Derivative Derivative 0 -100 Point on the derivative plot at which the second derivative -2000 -300 changed its sign from positive to negative was selected as the EW and LW demarcation point 10 12 14 16 69.5 70.0 70.5 71.0 71.5 72.0 3
- 4. 8/2/2011 Results and Discussion Results and Discussion (a) (b) Difference (Threshold-Inflection) Difference (Morks-Inflection) Inflection (mm) Inflection (mm) 80 80 0.5 1.0 -0.5 0.0 40 40 0.0 0 0 5 10 15 5 10 15 0 20 40 60 80 100 0 20 40 60 80 100 D Ring Number from pith Ring Number from pith Threshold (mm) Morks (mm) (c) Difference (Morks-Threshold) 1.5 Threshold (mm) 80 0.5 40 -0.5 0 5 10 15 0 20 40 60 80 100 Ring Number from pith Morks (mm) Results and Discussion Results and Discussion MI consistently overestimated the amount of EW in an MI (based on Silviscan®data) is a biased measure for annual ring compared to the threshold and inflection point defining the EW-LW transition methods Even though loblolly pine has an abrupt transition from The threshold method tends to be positively biased in EW to LW within an annual ring, it might take more time ring juvenile wood rings, i.e. it overestimates EW, compared to for tracheids to reach the defined threshold (2 X 2 X WT) the inflection point method We plan to analyze several of the samples examined in this Despite the differences observed among the methods it is study using conventional microscopy to determine if MI clear that all three methods show close agreement with continues to overestimate the amount of EW each other 4
- 5. 8/2/2011 Is a density of 480 kg/m3 a good threshold for EW-LW demarcation in loblolly p ? y pine 5
- 6. 8/2/2011 Introduction Identification of representative sampling heights for specific gravity and moisture content in plantation grown loblolly pine F. Antony , L.R. Schimleck & R.F. Daniels University of Georgia Athens, GA Loblolly pine is the most important plantation species in the southern US occupying more than half of the standing pine volume Introduction Introduction Quality of wood - Physical, mechanical and anatomical properties Estimates of wood properties - important for tree improvement programs and for the appropriate management of stands Nondestructive estimation - collection of an increment core from a height that best represents whole tree wood properties Loblolly pine is used as a principal source of raw material in the pulp and paper industry and for the production of lumber and composite wood products 1
- 7. 8/2/2011 Introduction Introduction Conventionally, for loblolly pine increment cores collected Raymond and Muneri (2001) at breast height (1.4 m) have been used • Must be accessible from the ground • At a fixed height How well does a breast height core represent whole-tree • Adequately represents whole tree wood property values p p properties? • Does not change from site-to-site • Specific gravity - Breast height SG as explanatory variable • Wahlgren and Fassanacht (1959) - 53% Objectives • Gilmore et al. (1961) - 50% • To examine the efficacy of breast height increment cores to • Moisture content represent whole tree SG and MC • Clark and Daniels (2000) - disks sampled from heights of 1.5 and 4.6 m (86%) • Identify the most representative sampling height for whole tree SG and MC Material and Methods Material and Methods Three trees from each stand Cross sectional disks 3.8cm thick were collected from 0.15, 1.4m and then at 1.5m intervals along the stem g Disk SG was measured using green volume and oven-dry weight 135 stands from six physiographic regions; 20-25 years old; Disk MC was measured using planted at 1250 trees per ha and thinned to 625 trees per ha; green and oven dry weight no fertilization and competition control 2
- 8. 8/2/2011 Material and Methods Material and Methods Material and Methods Material and Methods Whole-tree volume weighted SG and MC was computed Whole tree volume weighted SG and MC = weighted using the inside bark diameter, SG and MC measured from average SG and MC values for the disks that were sampled each disk • the weight used was a ratio between the volume of each segment and the total volume of the tree Volume A Au • Each bolt segment - frustum of a paraboloid V l L Correlation between the observed disk SG and MC value 2 • Bottom segment – cylinder V d 40000 L2 of each sampling height with the whole tree average SG • Top segment - cone V 3 d 40000 L 1 2 and MC was used to identify the best sampling height Total volume of the whole tree = sum of volume of each A linear regression model was proposed to predict whole section tree SG and MC values from different sampling heights 3
- 9. 8/2/2011 Results Results High correlations exist between disk Sampled Specific A nonlinear decreasing trend was observed with height for and whole tree SG at several heights Height (m) Gravity SG, an increasing trend with height was observed for MC between 1.4 and 7.6 m 0.3 0.88 1.4 0.91 3.0 0.92 The strength of the correlations 4.6 0.93 quickly decreased at heights greater SG of trees from the south Atlantic and Gulf Coastal than 7.6 m h 76 6.1 0.92 7.6 0.90 Plain’s were higher than the SG of trees from other regions 9.1 0.87 Across all regions, samples 10.7 0.82 collected at 4.6 m best represent 12.2 0.81 whole tree SG MC was significantly higher for trees from the north 13.7 0.76 15.2 0.70 Atlantic Coastal Plain compared to other regions 16.8 0.66 18.3 0.69 19.8 0.73 Results Results Specific gravity High correlations were observed for Sampled Moisture south north Height (m) Content Atlantic Atlantic Upper Gulf Hilly heights ranging from 4.6 to 9.1 m Height Coastal Piedmont Coastal Coastal 0.3 0.80 Coastal Coastal Plain Plain Plain 1.4 0.82 Plain Plain For heights greater than 9.1 m, 3.0 0.88 0.3 0.88 0.87 0.88 0.87 0.80 0.80 correlations between disk and whole 4.6 0.92 1.4 0.91 0.92 0.91 0.92 0.84 0.84 tree MC fell rapidly 3.0 30 0.92 0 92 0.92 0 92 0.88 0 88 0.93 0 93 0.87 0 87 0.86 0 86 6.1 0.93 4.6 0.94 0.95 0.86 0.96 0.93 0.83 7.6 0.89 6.1 0.95 0.92 0.80 0.94 0.94 0.84 7.6 0.90 0.93 0.88 0.90 0.92 0.78 Across all regions, samples 9.1 0.90 9.1 0.84 0.92 0.81 0.82 0.88 0.82 collected at 6.1 m best represent 10.7 0.83 10.7 0.80 0.62 0.68 0.83 0.82 0.80 whole tree MC 12.2 0.79 12.2 0.80 0.83 0.58 0.78 0.83 0.74 13.7 0.75 13.7 0.71 0.83 0.60 0.70 0.78 0.70 15.2 0.64 0.83 0.55 0.65 0.72 0.66 15.2 0.67 16.8 0.62 0.85 0.55 0.68 0.65 0.61 16.8 0.68 18.3 0.67 0.86 0.69 0.70 0.60 0.72 19.8 0.72 0.85 0.45 0.41 0.72 0.84 18.3 0.61 19.8 0.65 4
- 10. 8/2/2011 Results Discussion Moisture Content High correlations were observed between disk SG and MC south north Atlantic Atlantic Upper Gulf Hilly with volume weighted whole tree SG and MC for many of Coastal Piedmont Coastal Coastal Coastal Coastal Plain Plain Plain the sampled heights in the lower third of the tree Height Plain Plain 0.3 0.83 0.78 0.68 0.76 0.85 0.81 1.4 0.83 0.65 0.76 0.80 0.87 0.86 • SG - samples collected between 1.4-6.1 m 3.0 0.86 0.86 0.75 0.90 0.89 0.89 • MC - samples collected between 4.6-6.1m 4.6 0.90 0.94 0.80 0.91 0.94 0.90 6.1 0.92 0.90 0.86 0.96 0.92 0.90 7.6 9.1 0.83 0.86 0.84 0.91 0.88 0.79 0.90 0.88 0.94 0.92 0.85 0.89 For SG, breast height cores represented whole tree SG very 10.7 0.80 0.51 0.67 0.82 0.90 0.86 well with a marginal difference compared to the most 12.2 0.78 0.73 0.48 0.76 0.84 0.83 representative height 13.7 0.73 0.74 0.51 0.73 0.78 0.78 15.2 0.64 0.56 0.60 0.68 0.66 0.67 16.8 18.3 0.63 0.71 0.78 0.77 0.56 0.59 0.72 0.76 0.73 0.47 0.61 0.47 A linear relationship was observed between whole disk SG 19.8 0.74 0.93 0.42 0.38 0.76 0.74 and MC with volume weighted whole tree SG and MC irrespective of region Discussion Discussion Specific gravity Moisture Content 5