Better late than early - earlywood-latewood demarcation methods

Wood Focus magazine
,
30 Dec 2011
rings in tree trunk

Wood’s properties are key to its potential applications, with density variation often a limiting factor. Finto Antony, Assistant Research Scientist from The University of Georgia, USA, outlines a comparison study in lobolly pine explaining earlywood-latewood demarcation methods within an annual ring.

Download/view a PDF of this article with all supporting diagrams (356k/opens in new window)

The density of wood, which is defined as the weight of fibre per unit volume, is considered to be one of its most important characteristics. This is due to density’s strong relationship with most physical and mechanical properties, and its correlation with pulp yield. Density varies between locations, from forest to forest, tree to tree and even within trees, both with height and from pith to bark. But the greatest variability in density is observed within annual rings that arise from the formation of different types of wood during the year – earlywood (EW), a light-coloured band of wood formed early in the growing season and latewood (LW), dark-coloured wood formed later in the year. For example, in loblolly pine average earlywood (EW) density is 330kg/m3, while in the latewood (LW) it is approximately 650kg/m3, with maximum density approaching 1,000kg/ m3. The differences between the two wood types arise from dissimilarity in the structure of cells formed in EW and LW. Measurements of within-ring variability are important, and can be used as an indicator of wood uniformity and to provide information relating to the physiology of wood formation.

Get to the point

Variation of density within rings is measured using X-ray densitometry and can be used to estimate annual ring, EW and LW width, and maximum and minimum densities within an annual ring. The determination of EW and LW width relies on the identification of the point where EW turns into LW. The criteria used by researchers to demarcate the transition has a significant influence on measurements of within-ring components such as the width and density of EW and LW, and average ring density. Some of the techniques used to identify the transition within an annual ring are Mork’s Index (MI), threshold density, and the maximum derivative method.

Mork’s Index is an anatomical-based definition proposed to distinguish between EW and LW within an annual ring. Two commonly used interpretations of MI are:

  • A tracheid (tubular cell in the xylem of vascular plants that transports collected water and mineral salts to other parts of the plant) is said to be latewood if the common wall thickness between adjacent cells (double the wall thickness of a tracheid) is greater than or equal to lumen diameter.
  • A tracheid is a latewood if twice the common wall thickness (twice the double wall thickness of a tracheid) between adjacent cells is greater than or equal to lumen diameter.


The threshold method uses a pre-defined density value to demarcate between EW and LW within an annual ring, for example a threshold density of 480kg/m3 has been used in loblolly pine (Pinus taeda L.). An alternative technique is known as the maximum derivative, or inflection point, method. A function is fitted to the density profile of each ring and its derivative is determined, the EW-LW transition is then defined as the point on the ring where the derivative is at its maximum.

Crossing the threshold

In loblolly pine, the threshold method is frequently used to identify the transition from EW to LW but it is not known how well it agrees with MI and the inflection point method. A study has been conducted to answer this question, where 20 radial strips (2mm tangentially and 7mm longitudinally, radial length was determined by the pith-to-bark length of the sample) prepared from 12mm increment cores sampled at a height of 1.37m (breast height) were used. The radial strips had been examined using property measurement tool Silviscan. The air-dry density (AD), tracheid wall thickness (WT) and tracheid radial and tangential diameters (RD and TD, respectively) were measured at high spatial resolution. The measured AD, WT and RD radial profiles for a single strip are shown (p15, top, Fig. 1).

The point of Earlywood-Latewood transition was determined for each ring in the 20 strips using MI, the threshold method and the inflection point method. Based on the calculated MI, the transition from EW to LW is defined as the point within a ring where MI is greater than or equal to one. To measure the transition based on the threshold and inflection point methods, the basic density at each radial position was computed from air-dry density using an appropriate conversion equation. A basic density value of 480kg/ m3 was used as a threshold to demarcate EW to LW transition in this study.

The third technique, called the inflection point method, was based on identifying the inflection point for the density profile of each individual ring. An inflection point is the point on a curve where the curvature changes – where the second derivative is equal to zero. Smooth spline models were fitted to individual ring density profiles and the point where the second derivative of the fitted function was equal to zero was selected as the demarcation point for the EW to LW transition within a ring.

An example of how the Earlywood- Latewood transition was identified for individual growth rings using the three different methods is shown (p15, middle, Fig. 2). The data used was from the three radial profiles shown (p15, top, Fig. 1) and contrast a juvenile wood (ring 2) and mature wood (ring 13) ring. Based on the threshold and inflection point methods, the Earlywood-Latewood transition occurred at 22.2mm (the distance from the pith) for the selected juvenile wood ring in Fig 2. However, the transition point based on MI occurred at 23.4mm, leading to an overestimate of the amount of earlywood in the ring. For the mature wood ring the transition point determined using the threshold and inflection point methods was 68mm and the transition point based on MI was 68.1mm.

Result variation

Examination of individual ring transition points for the 20 radial strips in this study showed that MI consistently overestimated the amount of EW in an annual ring compared to the threshold and inflection point methods. This is shown (p15, bottom, Fig. 3) where the difference between estimated transition points such as threshold-inflection, MIinflection and MI-threshold is plotted for each ring and each radial strip from pith to bark. The overestimation of EW based on MI was more pronounced for rings in the juvenile wood zone (primarily for the first five rings close to the pith. See p15, bottom, Fig. 3b and 3c). Similarly, compared to the inflection point method, the threshold method tends to overestimate EW in identifying the transition point for rings in the juvenile wood zone (Fig. 3a). This indicates that the use of a single threshold value (480kg/m3) to delineate Earlywood from Latewood in both juvenile wood and mature wood might be misleading in loblolly pine.

It was found that MI (based on Silviscan data) is a biased measure for defining the EW-LW transition, since it is based on the anatomical features of tracheids. Even though loblolly pine has an abrupt transition from Earlywood to Latewood within an annual ring, it might take more time for the tracheids to reach the defined threshold of double the common wall thickness being greater than lumen diameter, especially for rings from the juvenile wood zone.

Further information

Finto Antony, Email: fintoa@warnell.uga.edu

Thanks to co-authors Laurence Schimleck and Richard Daniels.