A Dendrometer Band Study of the Seasonal Pattern

7/27/2019 A Dendrometer Band Study of the Seasonal Pattern

1/5

New Zealand Jo urnal of Botany, 1983, Vol. 21 : 121-1260028-825X/83/2102-0121 $2.50/0 Crown copyright 1983 121

A dendrometer band study of the seasonal pattern of radialincrement in kaur i {Agathis australis)J. PALMERJ. OGDENDepartment of Botany, Auckland UniversityPrivate Bag, Auckland, New ZealandAbstract Simple dendrometer bands were usedto measure the radial increment of kauri (Agathisaustralis Salisb.) at 3 sites to the south and east ofAuckland during the 1980-81 growing season.Diameter increment cores taken from some of thetrees at the beginning and end of the study showedthat the radial expansion measured by the bandscorrelated significantly (P


2/5

122 New Zealand Journal of Botany, 1983, Vol. 21Table 1 Site descriptions and sample characteristics.Site characteristicsLocationLatitude; LongitudeGrid referenceAltitude (m)Topography (aspect)Main canopy tree*Total basal area (m /ha)Total density (no/ha)Number of banded treesDiameter range of banded trees (cm)

Kon 1Hunua Range3704'; 17508'N48; 708359335Steep slope (NE)Agathis australis1006983318-172

Site nameMoe 1Mount Moehau3632'; 17525'N39; 91498675 0Gentle slope or flat (NE)Weinmannia silvicola4384 11714-71

Moe 2Mount Moehau3631'; 17524'N35; 90001545 0Steep slope (NW)Agathis/Weinmannia7575 8841-139

'Stems of all species greater than 10 cm diam. at 1.2 m above ground level. Density and basal area values derived frompoint-centred quarter sampling (Cottam & Curtis 1956).

from mechanical damage and reduced springcorrosion. The vernier scale enabled measurementto an accurancy of 0.25 mm. Details of the methodand an assessment of some possible sources of e rror,are given in Palmer (1982).At the Hunua Range site, 33 kauris were bandedfrom 16 to 20 June 1980, and on Mt M oehau 25 werebanded from 28 to 30 August 1980 (Table 1).Increment cores, taken at the time of banding,confirmed that the next season's growth had notcommenced at these dates. The bands were read atabout 6-weekly intervals throughout the 1980 and1981 growing season. All banded trees were greaterthan 10 cm diam. at 1.2 m above the ground. At theMt Moehau sites, all erect kauri trees were banded,but at the Hunua Range site a stratified randomsample, based on all of the 10 cm size class intervalsrepresented, was taken; thus a range of tree sizes(and ages) was sampled (Table 1).Two increment cores were taken from each treewhen it was bande d. These were mou nted, surfaced,and used to determine mean radial growth ra tes. Atthe end of the study period, 2 more cores were takenfrom each of a random sample of 12 trees at theHunua Range site. The paired cores from each treewere cross-matched and the width of the 1980-81annual ring measured directly.Climatic records, kept by the Auckland RegionalAuthority at Hunua (10 km from the study site)since 1962, were used to calculate mean monthlyrainfall and temperature for July 1962 to June 1980and for July 1980 to June 1981. The po tential evapo-transpiration (PE) rate was then calculated byfollowing the procedure of Thornthwaite (1948) andtables given by Toebes (1968).

RESULTSMeasurements by dendrometer bands include stemswelling caused by moisture changes in the bark and

Fig. 1 Components of the dendrom eter band.

xylem as well as increases in diameter consequentupon cambial activity and cell expansion (Fritts etal. 1965). The possibility that the measurementsrecorded at the study sites reflect stem shrinkageand swelling rather than changes in growth ratecannot be ruled out. However, the close correlation(r = 0.9163; n = 12; P < 0.001) between the totalradial increment measured by the bands, and themean ring width (measured from cores, up-slopeand down-slope aspects) indicates that total radialgrowth was accurately recorded (Fig. 2). The oneinstance in which the dendrometer reading wasgreater than the direct core measurement probablyresulted from 2 upslope cores being taken as thedownhill side of the tree was inaccessible. As inmost gymnosperms the annual rings of kauri arenormally wider on the down-slope side (Fritts 1976,p. 220).Because of the small sample size, no statisticallysignificant difference in diameter growth could bedemonstrated among trees of different diameters(and ages) on any one site. Consequently theindividual tree data were averaged to obtain themean pattern of growth at each site (Fig. 3). Rapid


3/5

Palmer & OgdenRadial increment in kauri 123

3 0 -

53 2 0-

1 0 -

Q

5 0 100Tree diameter (cm)

15 0

Fig. 2 Comparison of radial increment from dendrometerbands (open circles) with measurements from 2 cores pertree (vertical bars). The core measurements were thoughtto represent the maximum and minimum ring widthincrement.

0 0 4 -

' I

0 0 3 -

0 0 2-

0 01-

Date

Fig. 3 Change in growth rate between sample periods.Kon 1 (circles); Moe 1 (triangles); Moe 2 (diamonds).Standard error bars omitted for clarity; see text.

Table 2 Radial growth of kauri (Agathis austmlis) at the 3 sites.Kon 1 Site namesMoe 1 Moe 2

Number of cores analysedMaximum period analysed (maximum core span)Mean period analysed (mean core span)Mean annual ring width (mm)Standard error (mm)

441 6 1 9 - 1 9 8 01 8 1 5 - 1 9 8 0

1 . 9 3 7 *0.098

211 7 3 3 - 1 9 8 01 8 5 7 - 1 9 8 0

1.0810.080

61 8 1 0 - 1 9 7 91 8 5 5 - 1 9 7 9

1.2290.097

*This mean value is significantly (P


4/5

124 New Zealand Journal of Botany, 1983, Vol. 2147% below normal*. A slight excess of evapo-transpiration over precipitation may be frequent inthe Hunua Ranges in January but soil moisturedeficits extended for 3 months during this studyperiod. Recharge only occurred after exceptionallyheavy rainfall in April (Fig. 4).

DISCUSSIONThe results obtained by using simple dendrometersaccurately reflect the radial increment associatedwith the formation of an annual ring. However, thedendrometer band measurements were generallyslightly smaller than the mean of the ring widthestimated from the 2 increment cores (Fig. 2). Thismay indicate an initial period of 'taking up the slack'before the band s registered growth. Growth was notuniform at all sites throughout the season, but themid-summer growth check was not associated withthe formation of a 'false ring' or wider than averagelate wood.Previous studies on kauri diameter growth havesuggested that growth occurs through most of theyear, ceasing only in June, July, and August (I.Barton pers. com m.). A second, late season leaf andheight growth flush has been recorded for kauriseedlings (R. C. Lloyd, in Ecroyd 1982) andoccasionally for mature trees (I. Barton pers.comm.). The similar bimodal diameter growthpatterns imply that some common variable affectedgrowth at the lower altitude sites. This effect did nethowever extend to the higher altitude site wherethere was no indication of a second growth peak.Water shortage during summer is known to beone of the most important environmental factorsinfluencing cambial activity (Studhalter et al. 1963).Furthermore, if water stress is alleviated by late-season rainfall radial growth may resume untillimited again by declining temperatures (Zahner1968). The dry summer and associated soil moisturedeficit in the Hunua Ranges seems a plausiblereason for growth cessation. Both of the lower sitesare at medium altitudes on steep, dry slopes. Incontrast, the higher altitude site is mainly flat,boggy, and is situated on the leeward side of thesummit of Mt Moehau (888 m). This summit isfrequently enveloped in cloud (Cranwell & Moore1936); condensation and 'fog drip' may contributesubstantially to the total moisture available even inyears of below average rainfall.

16-? 12-LUQ.=8 8 -Q.

4 -

A S O N D J FTime (months)

M A M J

20 -

16 -

8 -

4 -

A S O N D J FTime (months)

M A M J

*Analysis of records from Station A64871 (Albert Park)given in Climatological Tables in Supplements to the NewZealand Gazette for the relevant months.

Fig. 4 Average precipitation (P; triangles) and potentialevapo-transpiration (PE; circles) for: top, the period July1962 to June 1980 and bottom , the period July 1980 toJune 1981. When PE is greater than P (shaded) soilmoisture is not being recharged sufficiently and moisturestress may occur.

The exceptionally good growth of kauri at theupper site in the warm sunny summer of 1980-81may imply that, in common with other tree speciesgrowing near their altitudinal limits, low tempera-tures may limit growth more frequently than doessoil moisture at high altitude sites.

ACKNOWLEDGMENTSWe thank Ian Barton and Andrew Dakin of the AucklandRegional Authority at Hunua for advice and fieldassistance. Richard Serra helped with technical aspects.Permission to work in the Hunua Ranges and on MountMoehau was given respectively by the Auckland RegionalAuthority and the Auckland Conservancy of the NewZealand Forest Service. Various friends and colleaguesaccompanied J. P. on trips to the top of Mount Moehau.


5/5

Palmer & OgdenRadial increment in kauri 125REFERENCESBenecke, U. ; Hav ranek, W. M. 1980: Phenologicalgrowth characteristics of trees with increasingaltitude, Craigieburn Range, New Zealand. In:Benecke, U.; Davis, M. R. ed . Mountainenvironments and subalpine tree growth. NewZealand Forest Service Technical Paper No. 70.Wellington, Government Printer, pp. 155-174.Borman, F. H.; Kozlowski, T. T. 1962: Measurements oftree ring growth with dial gauge dendrometers andvernier tree ring bands. Ecology 43 : 289 294.Cottam , G.; Curtis, J. T. 1956: The use of distancemeasures in phytosociological sampling. Ecology37 : 451-460.Cranwell, L. M.; Moore, L. B. 1936: The occurrence ofkarui in montane forest on Te Moehau. NewZealand journal of science and technology IS :531-643.Dunw iddie, P. W. 1979: Dendrochronological studies ofindigenous New Zealand trees. New Zealand

journal of botany 17 : 251266.Ecroyd, C. E. 1982: Biological flora of New Zealand 8.Agathis australis (D. Don) Lindl. (Araucariaceae)kauri. New Zealand journal of botany 20 : 17 36.Fritts, H. C. 1976: Tree rings and climate. London,Academic Press.Fritts, H. C ; Smith, D. G.; Stokes, M. A. 1965: Thebiological model for paleoclimatic interpretation ofMesa Verde tree-ring series. American antiquity31 : (2), (part 2) 101-121.

Hall, R. C. 1944: A vernier tree-growth band. Journal offorestry 44 : 742-743.Lamarche, V. C. Jr; Holmes, R. L.; Dunwiddie, P. W.;Drew, L. G. 1980: Tree ring chronologies of theSouthern Hemisphere. Chronology Series 5, Vol. 3.Tucson Laboratory of Tree-Ring Research, Uni-versity of Arizona, p. 77.Liming, F. G . 1957: Homemade dend rometers. Journal offorestry 55 : 575-577.Mason, G. W.; Cham bers, T. C. 1950: V egetation ofnorth-east Coromandel (Cape Colville Peninsula)with special reference to Mt Ts Moehau. Tane 3 :6 9 - 7 1 .Palmer, J. G. 1982: A dendrochronological study of kauri(Agathis australis Salisb.) M.Sc Thesis. AucklandUniversity.Studha lter, R. A.; Glock, W. S.; Ga erter, S. R. 1963:Tree growth some historical chapters in the studyof diameter growth. Botanical review 29 : 245 365.Tho rnthwa ite, C. W. 1948: An approach towards anational classification of climate. Geographicalreview 38 : 5 5 - 9 4 .Toebes, C. 1968: Com putation of the potential evapo-transpiration by the Thornthw aite method (Proce-dure No. 14) Handbook of hydrological procedures.Zahner, R. 1968: Water deficits and the growth of trees.In: Kozlowski, T. T. ed . Water deficits and plantgrowth Vol. 2 New York, Academic Press, pp.191-254.

Documents

A Dendrometer Band Study of the Seasonal Pattern