Skip to Main content Skip to Navigation

Stabilité des relations branche-tige chez l’épicéa commun en régime de futaie régulière : Conséquences sur la modélisation des caractères de la branchaison intéressant la qualité du bois.

Abstract : We modelled Norway Spruce branchiness, in order to estimate wood quality of standing trees grown in monospecific even-aged stands from inventory data. We studied the whorl branches of the preformed growth unit. Their number, basal diameter and insertion angle were estimated. For delimiting crown base, we modelled heights to the first green branch, first green whorl and first dead branch. We started from a model previously established from young trees data representing the average fertility classes found in Vosges mountains. In this model, the averages of branch basal diameter and insertion angle for a whorl were computed from stem height and diameter at breast height. We aimed at: (i) modelling branch basal diameter and insertion angle variance in the whorl; (ii) verifying that the model could be applied to stands grown under conditions not accounted for in the original calibration sample (high site indices, old stands); (iii) determining the geographical area within which the model can be used without requiring any new calibration work. The branch basal diameter and insertion angle studies samples were collected in National Forest Districts Vosges (72 trees) and "Plateau Lorrain" (26 trees). The stands were selected according to their age (from plantation, if this information was available, measured on stumps otherwise) and site index (HØ), defined as the estimated dominant height at stand age 100 years. Northeastern France site classes I and II defined the high fertility modality, while site class III constituted the low fertility modality. Age modalities were: young (35 years), mature (66 and 68 years), old (95 and 133 years). After felling stem data were collected: Total height (Ht), diameter at breast height (D130), year annual height increments. In every second growth unit starting from the current year leader shoot, whorl branches were measured for basal diameter and insertion angle. Branches were declared dead when no green lateral could be found. Crown base was defined by the first green whorl (first whorl from the ground in which at least 3/4 of the branches were living), first green branch, first dead branch (no diameter constraint). Climate was defined by monthly temperature and rainfall averages computed over the 1986-1996 period. For establishing the crown base model, the sample was completed by 24 Belgian trees (4 stands, age = 48 to 82 years, HØ = 29 to 46 m) and 106 Swedish trees from 12 sampling sites distributed along a Northward gradient starting from the Southern coast (age = 61 to 152 years, HØ = 18 to 36 m). A study of the branch axis year annual area increment was carried out on 4 trees collected in a site class I stand of the "Plateau Lorrain" National Forest District. In those trees a stem analysis paralleled the measurement of branch axis year annual area increments in each branch year length increment fom the branch tip to its extremity inserted on the stem. The crown model validation sample consisted in 38 trees collected in Midi-Pyrénées administrative region (12 stands, age = 20 to 60 years, HØ = 29 to 52 m). The branch growth model validation sample was collected in Vosges National Forest District (2 stands aged 36 and 47 years, HØ = 46 and 49 m). In old trees whorls, the high variance of branch basal diameter and insertion angle led us to model separately dead and living branches. For living branches, it was shown that a same model based on Ht, D130 and stand age could estimate branches basal diameter and insertion angle in remote geographical areas. Estimates discrepancies previously observed between geographical areas could be attributed to dead branches and to the branch decay process. This was confirmed by the crown base model, in which accounting for climate was a necessity for describing all the observed variability patterns. Dead braches basal diameter and insertion angle were derived from the living branches model, by simulating branch radial growth. The study of branches year annual area increment effectively demonstrated a good correlation with stem height and radial growth, that can be seen as the baseground of the statistical correlation demonstrated between branch basal diameter and stem dimensions (Ht, D130). However branch year area increment correlated poorly to the contemporaneous stem height and radial increment, but rather to the stem height and radial increments averaged over the preceeding years. This study showed that decreasing the need for model calibration was conditionned to a better understanding of the branch mortality process and to a better explicitation of branch growth. Solutions are proposed to increase brach growth explicitation in future wood quality oriented tree simulation models.
Complete list of metadata

Cited literature [165 references]  Display  Hide  Download
Contributor : Michel Loubère Connect in order to contact the contributor
Submitted on : Friday, March 27, 2015 - 1:09:35 AM
Last modification on : Friday, August 5, 2022 - 2:38:10 PM


  • HAL Id : tel-01136332, version 1



Michel Loubère. Stabilité des relations branche-tige chez l’épicéa commun en régime de futaie régulière : Conséquences sur la modélisation des caractères de la branchaison intéressant la qualité du bois.. Sylviculture, foresterie. ENGREF (AgroParisTech), 2007. Français. ⟨tel-01136332⟩



Record views


Files downloads