A Sitka spruce height-age model with improved extrapolation properties

1997 ◽  
Vol 73 (3) ◽  
pp. 363-369 ◽  
Author(s):  
Gordon D. Nigh

The height-age model for Sitka spruce (Picea sitchensis (Bong.) Carr.) currently recommended for use in British Columbia has poor extrapolation properties. Therefore, a new height-age model for Sitka spruce using stem analysis data collected from the Queen Charlotte Islands was developed. Care was taken to meet the standard regression assumptions. In particular, accounting for within-plot serial correlation improved the extrapolation abilities of the model by eliminating the crossing-over effect. The new model is being recommended for use in British Columbia because it offers better extrapolated height and site index estimates without sacrificing accuracy at young ages. Key words: Sitka spruce, site index, height-age model, serial correlation, nonlinear regression, extrapolation, model properties

2002 ◽  
Vol 78 (5) ◽  
pp. 690-694 ◽  
Author(s):  
Chris J Cieszewski ◽  
Gordon D Nigh

We present a new Sitka spruce (Picea sitchensis (Bong.) Carr) site-dependent height-age model that is based on a dynamic site equation simulating previously published height-age curves for various productivity sites. The new model is an improvement over the previous model because it uses any arbitrary height-age pair to directly predict a height at another age, instead of using a fixed base-age site index as the older model does. Consequently, it can also be used directly to compute height at any age from site index or site index from any height and age instead of relying on numerical solutions for site index computations. The model predicts the same heights for any site as the original fixed base-age model and has the same desirable properties of polymorphism, inflection point, variable asymptotes, logical behaviour, theoretical basis, parsimony, and improved extrapolation. The model is offered as an algebraic improvement only, and therefore it was calibrated on pseudo-data generated from the old model’s predictions rather than on real data. The proposed equation mimics the old model better than the other dynamic equations tested in this study, which is illustrated using examples with the Chapman-Richards function. Analysis with the real data might offer further improvements to the model predictions. Key words: Base-age invariance, height-age model, model properties, nonlinear regression, Sitka spruce, site index


1989 ◽  
Vol 19 (9) ◽  
pp. 1203-1207 ◽  
Author(s):  
Wilbur A. Farr ◽  
Donald J. DeMars ◽  
J. Edward Dealy

Data from 157 open-grown Sitka spruce (Piceasitchensis (Bong.) Carr.) and 102 western hemlock (Tsugaheterophylla (Raf.) Sarg.) at two locations in coastal Alaska and from 69 Sitka spruce on the Queen Charlotte Islands in British Columbia were used to develop height–diameter and crown width–diameter equations. Height–diameter relations for Sitka spruce were significantly different among all locations. Crown width–diameter relations were not significantly different. Common equations were fit to the spruce data and to the spruce and hemlock data independent of location and site index along the coast.


2019 ◽  
Vol 49 (11) ◽  
pp. 1471-1482
Author(s):  
Woongsoon Jang ◽  
Bianca N.I. Eskelson ◽  
Louise de Montigny ◽  
Catherine A. Bealle Statland ◽  
Derek F. Sattler ◽  
...  

This study was conducted to quantify growth responses of three major commercial conifer species (lodgepole pine (Pinus contorta Douglas ex Loudon var. latifolia Engelm. ex S. Watson), interior Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco var. glauca (Beissn.) Franco), and spruce (white spruce (Picea glauca (Moench) Voss) and hybrid spruce (Picea engelmannii Parry ex. Engelm. × Picea glauca (Moench) Voss × Picea sitchensis (Bong.) Carrière))) to various fertilizer blends in interior British Columbia, Canada. Over 25 years, growth-response data were repeatedly collected across 46 installations. The fertilizer blends were classified into three groups: nitrogen only; nitrogen and sulfur combined; and nitrogen, sulfur, and boron combined. The growth responses for stand volume, basal area, and top height were calculated through absolute and relative growth rate ratios relative to a controlled group. Fertilizer blend, inverse years since fertilization, site index, stand density at fertilization, and their interactions with the fertilizer blend were used as explanatory variables. The magnitude and significance of volume and basal area growth responses to fertilization differed by species, fertilizer-blend groups, and stand-condition variables (i.e., site index and stand density). In contrast, the response in top height growth did not differ among fertilization blends, with the exception of the nitrogen and sulfur fertilizer subgroup for lodgepole pine. The models developed in this study will be incorporated into the current growth and yield fertilization module (i.e., Table Interpolation Program for Stand Yields (TIPSY)), thereby supporting guidance of fertilization applications in interior forests in British Columbia.


Author(s):  
L. Roche ◽  
P. G. Haddock

SynopsisThe natural distribution of Sitka spruce is examined in relation to the species' role in British forestry. Particular attention is therefore given to a description of habitats on the Queen Charlotte Islands, and also on the adjacent mainland of British Columbia where Sitka spruce is sympatric with white spruce (Picea glauca) and hybridisation occurs. Examples of genetic variation are given and it is suggested that a knowledge of the genecologv of the species in its natural habitat would help to elucidate its full ecological and genetic potential in Britain.


1995 ◽  
Vol 12 (2) ◽  
pp. 57-63 ◽  
Author(s):  
Bijan Payandeh ◽  
Yonghe Wang

Abstract Stem analysis data from plantations of black spruce, white spruce, and jack pine from northern Ontario were fitted to base-age specific and base-age invariant site index models. The resulting equations and their respective parameters were compared via nonlinear analysis of covariance. The base-age specific models produced a somewhat better fit to the data than their base-age invariant counterparts, although the latter are considered theoretically more elegant. Graphical comparison of plantation and natural stand site index curves for the three species showed both similarities and differences. North. J. Appl. For. 12(2):57-63.


2000 ◽  
Vol 15 (3) ◽  
pp. 117-121 ◽  
Author(s):  
Gordon D. Nigh ◽  
Bobby A. Love

Abstract Good estimates of juvenile heights are important for silviculture and timber supply decision-making. These height estimates are particularly important for estimating the number of years it takes a stand to reach green-up and breast height. Thirty-nine 0.04 ha plots were established in managed juvenile stands of interior spruce in the Prince Rupert Forest Region. Four sample trees in each plot were split longitudinally and height-age data were obtained from the exposed pith nodes or branch whorls. The height-age data were fit to a combined exponential and power function to create a juvenile height model for interior spruce. This model is a function of age and site index. Years to breast height and green-up age models were developed with data generated from the juvenile height model. Forecasts from these models indicate that the time to reach breast height and green-up age is less than previously expected. The juvenile height model, years to breast height model, and green-up age model should be used for site indices above 14.2 m. In addition, the height model should only be used up to total age 20. West. J. Appl. For. 15(3):117-121.


1989 ◽  
Vol 4 (3) ◽  
pp. 85-88
Author(s):  
James L. Vander Ploeg ◽  
James A. Moore

Abstract Stem analysis data from Douglas-fir (Pseudotsuga menziesii) collected throughout the inland Northwest were used for testing height growth and site index equations. The equations performed well in northern and central Idaho, northeast Oregon, and northeast Washington on vegetative types similar to those sampled in model development. However, if the equations were applied on drier sites outside the original geographic study area, overestimates of height growth and under-estimates of site index could result. Therefore, revised height growth and site index equations are presented for western Montana and central Washington. West. J. Appl. For. 4(3):85-88, July 1989.


1985 ◽  
Vol 9 (3) ◽  
pp. 166-169 ◽  
Author(s):  
Ralph L. Amateis ◽  
Harold E. Burkhart

Abstract Stem analysis data collected from dominant and codominant loblolly pine (Pinus taeda) trees in cutover, site-prepared plantations were used to develop site index curves. The data were collected over much of the natural range of loblolly pine. A separable differential equation which expresses height growth as a function of both height and age was used to develop the site index curves. These site index curves should be applicable to loblolly pine plantations on cutover, site-prepared lands through much of the South.1


2000 ◽  
Vol 132 (6) ◽  
pp. 811-823 ◽  
Author(s):  
Thomas W. Phillips ◽  
Gerald N. Lanier

AbstractHost specificity of Pissodes strobi (Peck) from different geographic regions and genetic divergence of local host-associated weevil populations were studied in a series of experiments. Pacific coast P. strobi reared from Sitka spruce, Picea sitchensis (Bong.) Carr (Pinaceae), were unable to successfully colonize either eastern white pine, Pinus strobus L. (Pinaceae), or western white pine, Pinus monticola Dougl. ex D. Don., in a forced-infestation study on interplanted trees in New York. Reproductively mature field-collected P. strobi from British Columbia did not oviposit on eastern white pine in New York, but field-collected New York weevils successfully reproduced in Sitka spruce leaders in British Columbia. Unacceptability of eastern white pine for western P. strobi was shown to be under genetic control, rather than influenced by prior host experience on Sitka spruce. Pissodes strobi originating from Sitka spruce but reared one generation in the laboratory on the exotic Norway spruce, Picea abies (L.) Karst., were also unable to utilize eastern white pine as a host in a forced-infestation experiment in the field. Population genetic studies using allozyme electrophoresis found that P. strobi populations occurring on different host species within 2 km of each other had significant differences in allele frequencies in three out of four cases. These results suggest that P. strobi can exist as small breeding populations that can facilitate host specialization. Applied research on host resistance against P. strobi could target mechanisms that prevent western P. strobi from utilizing nonhosts such as eastern and western white pines.


1998 ◽  
Vol 74 (4) ◽  
pp. 588-596 ◽  
Author(s):  
Gordon D. Nigh

The focus of recent site productivity research in British Columbia has been to develop height-breast height age, growth intercept, and ecosystem-site index models. These models, together with a years-to-breast-height model, form a system for estimating site index and height. This system is described for western hemlock (Tsuga heterophylla (Raf.) Sarg.) in the interior of British Columbia. Forty-four western hemlock stem analysis plots were used in the construction of this system. As there are three models for estimating site index, the appropriate model for a given stand depends largely on the stand condition and the precision of the models. A graph of model precision against breast height age gives some assistance in deciding which model should be used to estimate site index. Key words: western hemlock, site index, height-age model, growth intercept model, years-to-breast-height model, ecosystem-site index model, model precision


Sign in / Sign up

Export Citation Format

Share Document