Accès libre

Mean height or dominant height – what to prefer for modelling the site index of Estonian forests?

Toomas Tarmu
Toomas Tarmu
, 
Diana Laarmann
Diana Laarmann
 et 
Andres Kiviste
Andres Kiviste
   | 18 sept. 2020
Forestry Studies's Cover Image
À propos de cet article
Article précédent
Article suivant
Citez
Article Category: Research paper
Publié en ligne: 18 sept. 2020
Pages: 121 - 138
Reçu: 29 mai 2020
Accepté: 30 juin 2020
DOI: https://doi.org/10.2478/fsmu-2020-0010
Mots clés
© 2020 Toomas Tarmu et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1

Distribution of the ENFRP plot measurements by forest type and main tree species.
Distribution of the ENFRP plot measurements by forest type and main tree species.

Figure 2

Distributions of the main species share in the dominant tree cohort for main species groups.
Distributions of the main species share in the dominant tree cohort for main species groups.

Figure 3

Distribution of residuals and threshold values for excluding diameter annual increment (id) outliers on a scatterplot of id mixed model residuals and relative diameter (d/dgdom).
Distribution of residuals and threshold values for excluding diameter annual increment (id) outliers on a scatterplot of id mixed model residuals and relative diameter (d/dgdom).

Figure 4

The difference between the estimated dominant height hgdom and the empirical arithmetic mean height of the 100 thickest trees per hectare.
The difference between the estimated dominant height hgdom and the empirical arithmetic mean height of the 100 thickest trees per hectare.

Figure 5

Share of the thinned basal area in the basal area of the main species between two consecutive plot measurements depending on stand age. The solid horizontal line shows a 5% threshold value for thinning.
Share of the thinned basal area in the basal area of the main species between two consecutive plot measurements depending on stand age. The solid horizontal line shows a 5% threshold value for thinning.

Figure 6

The share of the dead tree basal area in the basal area of the main species between two consecutive plot measurements depending on stand age.
The share of the dead tree basal area in the basal area of the main species between two consecutive plot measurements depending on stand age.

Figure 7

Scatterplots of the mean height hg and the dominant height hgdom of cohorts by tree species.
Scatterplots of the mean height hg and the dominant height hgdom of cohorts by tree species.

Figure 8

The average change in the mean height and dominant height with a 95% confidence interval. The sample plots were measured at a 5-year interval. Red dots represent thinned sample plots and blue dots represent unthinned sample plots.
The average change in the mean height and dominant height with a 95% confidence interval. The sample plots were measured at a 5-year interval. Red dots represent thinned sample plots and blue dots represent unthinned sample plots.

Figure 9

The difference in the dominant height between the beginning and the end of the five-year period. Red dots represent thinned sample plots and blue dots represent unthinned sample plots.
The difference in the dominant height between the beginning and the end of the five-year period. Red dots represent thinned sample plots and blue dots represent unthinned sample plots.

Figure 10

Growth error of the mean height model (7) and the dominant height model (8). Red dots represent thinned sample plots and blue dots represent unthinned sample plots.
Growth error of the mean height model (7) and the dominant height model (8). Red dots represent thinned sample plots and blue dots represent unthinned sample plots.

Height curve statistics specifications for Table 2.

Height curve statistics
n – number on dominant trees;
me – arithmetic mean residual hmeasured – hpredicted (systematic error);
sde – quadratic mean residual hmeasured – hpredicted (random error);
p-value – t-test for hypothesis that mean residual is different from zero;
desc – despite excluding outliers and a moderate number of observations (n ≥ 16) in each cohort, the height curve was descending (parameter a < 0 for models 2 and 3, parameter b < 0 for model 4) in the case of some cohorts.

Values of species-specific coefficients for height curve models 4, 5 and 6.

ModelCoefficientScots pineNorway spruceSilver birchOther deciduous
5asp0.302010.443250.279360.27936
6asp0.320440.441030.395540.25803
6bsp0.004340.006430.006290.00582
4, 6csp1.480761.372551.329982.04430

Species-specific coefficients for regression model (9).

CoefficientScots pineNorway spruceSilver birchOther deciduous
a02.04943.36613.04741.7910
a1−0.0964−0.1231−0.0824−0.1108
a2−15.686−14.070−34.854−15.215

Fit statistics of the height curve models (2–6) for dominant trees based on experimental data. (n – number of trees; me – mean residual; sde – mean square residual; p-value – p-value of t-test; desc – number cohorts with a descending height curve).

StatisticScots pineNorway spruceSilver BirchOther deciduousAll species
n5165914426766581
Model (2) me−0.08−0.17−0.18−0.23−0.10
Model (2) sde1.371.181.381.411.35
Model (2) p-value<0.001<0.0010.0080.165<0.001
Model (2) desc50106
Model (3) me−0.07−0.13−0.15−0.21−0.09
Model (3) sde1.371.181.381.401.35
Model (3) p-value<0.0010.0010.0250.188<0.001
Model (3) desc50106
Model (4) me−0.01−0.08−0.110.01−0.03
Model (4) sde1.371.171.371.351.34
Model (4) p-value0.6020.0440.1130.9370.127
Model (4) desc50106
Model (5) me−0.09−0.61−0.14−0.44−0.17
Model (5) sde1.471.451.511.491.47
Model (5) p-value<0.001<0.0010,0600,009<0.001
Model (6) me−0.03−0.05−0.230.02−0.04
Model (6) sde1.471.321.521.371.45
Model (6) p-value0.1420.3010.0020.8960.013
eISSN:
1736-8723
Langue:
Anglais
Périodicité:
2 fois par an
Sujets de la revue:
Life Sciences, Plant Science, Ecology, other
RSS Feed de la revue