Documents

Stand structure and growth patterns of understorey trees in a coniferous forest, Taisetsuzan National Park, northern Japan

Description
Eclgical Research (1994) 9, Stand structure and grwth patterns f understrey trees in a cniferus frest, Taisetsuzan Natinal Park, nrthern Japan YASUHRO KUBOTA, 1~ YASUO KONNO 1 AND TSUTOM HURA 2
Categories
Published
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
Eclgical Research (1994) 9, Stand structure and grwth patterns f understrey trees in a cniferus frest, Taisetsuzan Natinal Park, nrthern Japan YASUHRO KUBOTA, 1~ YASUO KONNO 1 AND TSUTOM HURA 2 *Labratry f Envirnmental Btany, Faculty f Agriculture, Obihir University f Agriculture and Veterinary Medicine, Nishi 2-11, nada-ch, Obihir, Hkkaid 8 and 2 Nakagawa Experiment Frest, Hkkaid University f Agriculture, Otineppu 98-25, Hkkaid, 8Japan Stand structure was studied with special reference t grwth and mrtality patterns f sapling and understrey trees in a cniferus Picea jezensis and Abies sachalinensis frest in Taisetsuzan Natinal Park, Hkkaid, nrthern Japan. Picea jezensis was dminant in the basal area, while A. sachalinensis was abundant in large numbers in the canpy. Estimated mrtalities increased significantly with diameter at breast height (DBH) fr bth P. jezensis and A. sachalinensis in the canpy, but the tendency was different between the tw species. Piceajezensis had a lwer mrtality rate than A. sachalinensis, especially at small DBH classes. The spatial distributin f understrey individuals f A. sachalinensis did nt shw any significant crrelatin with the spatial distributin f canpy gaps, but that f P. jezensis shwed a significant crrelatin. Abies sachalinensis can grw higher than P. jezensis under suppressed cnditins; while P. jezensis requires canpy gaps fr steady height grwth. This grwth pattern leads t a different waiting height in the understrey ( 2 m in height and 1 cm in diameter at breast height). Abies sachalinensis waited fr an imprvement in light cnditins at higher strata (max. 7 m), while P. jezensis waited at lwer strata (max. 3 m). The estimated mrtality f understrey A. sachalinensis increased with size, while that f understrey P. jezensis decreased. Therefre, P. jezensis gives pririty t survival while A. sachalinensis gives pririty t understrey grwth. The difference in the 'waiting pattern' between the tw species in the understrey was cnsidered a significant feature fr the canpy recruitment prcess f P. jezensis and A. sachalinensis. Key wrds: Abies sachalinensis; Piceajezensis; understrey tree; waiting pattern. NTRODUCTON Picea jezensis Carr. and Abies sachalinensis Masters are c-ccurring species in sub-breal frests, in Hkkaid, nrthern Japan. The structure and dynamics f P. jezensis and A. sachalinensis frests have nt been cmpletely investigated. Several hyptheses n the c-existence mechanisms fr Picea and Abies in nrthern cniferus frests have been prpsed with respect t life-histry traits f the tw species. Osting and Reed (1952) and *Present address: Department f Bilgy, Faculty f Science, Tky Metrplitan University, Tky, Japan. Received 18 May Accepted 6 May Veblen (1986) suggested that the tw species cexist by having different life-histry characteristics: P. engelmanii Parry has a lwer recruitment rate and greater lngevity in the canpy layer, while A. lasicarpa Hk has a higher recruitment rate and higher mrtality f canpy trees. Hwever, these life-histry traits, especially canpy recruitment rate, are affected by envirnmental factrs such as natural disturbance. Regeneratin sites frmed by gap disturbance shw varius characteristics that have a wide range in size, and hetergeneus light envirnments in each frmatin pattern (Nakashizuka 1985; Canham 1988b; Pulsn & Platt 1989; Lieberman et al. 1989). The difference in the grwth pattern brings abut different respnses against varius gap disturbance regimes (Hartshrn 1978; Denslw 198; Canham 1988a). Therefre, 334 Y. Kubta et al. it is imprtant t understand the grwth pattern mre clearly, and t investigate the species-specific perfrmance f cmpnent species with respect t the disturbance regime. n this paper, the differences in the grwth pattern f understrey trees f P. jezensis and A. sachalinensis are studied. The 'waiting pattern' f understrey, trees is shwn t be a significant characteristic related t the recruitment int the canpy. Finally, the difference in the 'waiting pattern' is discussed between P. jezensis and A. sachalinensis. STUDY STE This studywas carried ut in a Picea-Abies frest (1 m a.s.1.) n the ESE slpe f Mt Onsen, Taisetsuzan Natinal Park, Hkkaid, nrthern Japan. Precipitatin is apprximately 15 mm per year. Mean daily temperatures during the warmest mnth (August) and the cldest mnth (January) are 17.7 ~ C and ~ C, respectively. The warmth index (Kira 1977) is 44.1, indicating that the regin is lcated between the cl-temperate and sub-breal znes. Snw cver n the frest flr is usually frm Nvember until the fllwing May. The sil is predminantly brwn frest sil, with lcal distributin f blck streams by periglacial frmatin. The study site is dminated by Picea jezensis Carr., Abies sachalinensis Masters, Betula e~anii Cham. and Picea glehnii Masters; Acer ukruenduense Trautv. et Mey. and Srbus cmmixta Hedl. are als fund but nly lcally. The vertical stand structure is multi-layered, where canpy trees f P. jezensis, B. ermanii and P. glehnii vertp the canpy frmed by A. sachalinensis. Tw types f frest flr vegetatin are recgnized; Sasa and Carex n the brwn sil and Vaccinium, Rhddendrn and Menziesia n the blck stream. METHODS A study plt f 1.8 ha (1 m x 18 m) was established in The whle area was cvered with a 1 x 1 m grid system fr field survey. Three life phases were defined fr trees: (i) sapling (3 cm ~ height 2 cm); (ii) understrey tree (height i 2 cm and diameter at breast height 1 cm); and (iii) canpy tree (diameter at breast height t 1 cm). The diameter at breast height (DBH) at 1.3 m was measured fr all living and dead trees i 2 m in height. Height grwth rate and release height (defined as the height where the average stem grwth f the initial 5 years was at least twice the average fr the fllwing 5 years) f understrey trees was determined by detecting scars f branch whrls that frmed annually n the trunk. Fr this purpse, the authrs climbed up t the tree crwns using a ladder. Dead trees i 1 cm in DBH were identified t species level by examining samples f tissues sectined under a micrscpe (Piceajezensis and Picea glehnii were dealt with as the genus Picea) and were classified int fur types accrding t the estimated causes f death; ble snapping, base snapping, uprting and standing dead. Treefall directin was als measured. The damage rati f trees at each DBH class fr each species was defined M i = Di/ (L i + Di), where L i and D i were the numbers f living and dead trees, respectively, at DBH class i. The decay class f fallen lgs was classified int six categries accrding t structural integrity and vegetatin cverage n the lgs (Graham & Crmack 1982; Arthur & Fahey 1989). Class 1 bles were essentially undecayed with mst r all f the bark and branches intact. At the ther extreme, class 6 bles were decayed nearly perfectly and detectable nly by their utline n the frest flr. Class 2, 3, 4 and 5 fitted int a cntinuum between these tw extremes. Subplts 2 m x 2 m in area were regularly settled n the grund at each grid cell intersectin in the plt (18 subplts in ttal). Furthermre, 81 subplts f varius sizes were set up n elevated surfaces such as fallen lgs, munds and stumps frmed by gap disturbances. The number f saplings in each subplt was recrded, and the crwn depth (a), crwn width (1) and stem grwth in the past 5 years were measured fr each sapling. The relative height grwth rate (RHGR = [1gH igh1984]/5; H1989 and H1984 were the tree heights in 1989 and in 1984, respectively f saplings investigated. Furthermre, crwn shape index (CS = lg [a/l]; sensu Nakamura & Obata 1982) fr saplings was cmpared between P. jezensis and A. sachalinensis. Sapling density at each site was estimated as sapling number per subplt area. Stand structure f sub-breal cniferus frest 335 Gaps were defined by the vertical prjectin area where n stems reached 1 m in height r 1 cm in DBH. The size f gaps was calculated as an ellipse by measuring the lng and shrt axes (Runkle 1982). Understrey trees beneath each gap, designated gap successrs, were recrded (133 trees in ttal). Gap ages were determined frm the release dates based n the increment f annual rings in cres btained frm the tw t fur tallest stems in each gap. lluminance at,.5, 1. and 2. m abve grund was measured regularly at the centers f 9 grid cells in summer. All the measurements were carried ut under vercast cludy cnditins using phtelectric illuminmeters (Tshiba SP-5, Tky). These illuminance values were divided by the illuminance in the pen site t btain relative illuminance. RESULTS Stand structure and gap frmatin pattern The basal area and density f canpy trees f the main canpy species (P, jezensis, P, glehnii, A. sachalinensis and B, ermanii) were m 2 and 355 per hectare, respectively. The density f understrey trees f canpy species was 247 per hectare. The basal area f Picea spp. exceeded that f A. sachalinensis, but A. sachalinensis ccurred in greater number. The DBH frequency distributins f canpy trees were inversely J-shaped. A. sachalinensis shwed a peak in stem number at 1 cm DBH class and declined sharply in larger size classes, while Picea spp. and B. ermanii shwed a gradual decrease in stern number (Table 1). Table 2. The cause f death fr Picea spp., A. sachalinensis and B. ermanii. Cause f death Picea spp. A. sachalinensis B. ermanii Base snapping Ble snapping Standing dead Uprting Frty-five per cent f fallen stems fell frm east t suth, which was the aspect f the slpe. The ble snapping, base snapping and uprting were the majrity (85 %) f the causes f death. The causes f death were nt significantly different amng species (Table 2, X 2 test, P .1). The DBH frequency distributin f ble snapping stems f Picea spp. differed significantly frm that f A. sachalinensis (Mann-Whitney U-test, P .1). The damage rati increased with DBH class fr Picea spp. and A. sachalinensis (Kendall rank crrelatin test, P .5), but B. ermanii did nt shw any significant crrelatins between mrtality and size. The increasing tendencies f damage rati differed between Picea spp. and A. sachalinensis (Table 3). Therefre, the estimated mrtality f Picea spp. was lwer, especially in small DBH classes, than that f A. sachalinensis. The size f gaps ranged frm 5 t 1245 m 2. The ttal gap area was 3467 m 2, r 19.3% f the whle study plt, shwing a lgnrmal frequency distributin (Fig. 1). This agreed with the results reprted by Fster and Reiners (1986), White et al. (1985) and Taylr (199). Relative illuminance increased with height and shwed an inverse J-.shaped fre- Table 1. DBH class distributin f living and dead stems. DBH class Number f stems (per 1.8 ha) (cm) P. jezensis P. glehnii A. sachalinensis B. ermanii (6) (47) 72 ( - ) (13) (56) 36 (1) (13) (58) 27 (3) (19) (45) 25 (9) (25) (48) 18 (9) (18) 6 2 (16) 15 (2) (1) () 5 () (2) 1 (2) 2 (1) (2) (1) 1 (2) Dead stems are shwn in parentheses. 336 Y. Kubta et al. Table 3. Changes in a damage rati with DBH class. S DBH dass Damage rati (%) 2 (cm) Picea spp. A. sachalinensis B. ermanii - i quency distributin at each height (Fig. 2). The lwer relative itluminance at -.5 m height may be ascribed t the fliage layer f Sasa kuriensis. ~9 5 d Z 3 2-,, i i i i,, ~, l l ~ l l J 1 ~OOr]n nnn n Om 5m 1.m Establishment site f seedlings and saplings The density f saplings n different frest flr substrata is shwn in Table 4. Saplings f A. sachalinensis were mst abundant, which was twice the density f P. jezensis. Occurrences f P. jezensis, A. sachalinensis and B. ermanii were mainly restricted t fallen lgs. As was reprted by Christy and Mack (1984) and Stewart (1989), establishment ccurred primarily n fallen lgs and munds. Establishment sites were mstly restricted t fallen OOOnOnn~ nnn ~ nf1 r] n ~ i Relative illuminance (%) 2.m g. g [] Others 9 B.ermanii [] Picea spp. 9 A.sachalinensis Fig. 2. Frequency distributins f relative illuminance at different heights f,.5, 1. and 2. m. lgs, especially fr P. jezensis. Mst fallen lgs were in decay class 2 and decay class 3. The highest sapling density fr all the species was fund in decay class 2 (Table 5). 5- Fig. 1. :!,,, Lg (gap area, m2) Size distributin f gaps in the plt. Table 4. strata. Occurrences f saplings in frest flr sub- Sapling density (per m 2) P. jezensis A. sachalinensis B. ermanii Fallen lg Mund Grund..2.1 Stand structure f sub-breal cniferus frest 337 Table 5. The frequency f fallen lgs and density (per m 2) f established saplings fr all the species n each decay class. Decay class Frequency Sapling density Grwth pattern and crwn shape f saplings The relative height grwth rate (RHGR) f sapling s shwed n significant difference between P. jezensis and A. sachalinensis, except saplings at 1-2 cm (t-test, P .5; Fig. 3)i The RHGR f saplings in gaps was significantly higher than that beneath canpies (P .1). Mst saplings f B. emlanii were distributed beneath gaps, and RHGR was higher than that f the ther species (t-test, P .1). This wuld indicate that B, ermanii depends n gaps as a regeneratin site at an earlier life stage than cnifers. The relatinship between RHGR and the crwn shape index (CS) fr saplings was analyzed (Fig. 4). fa tree has a slender crwn shape, CS . f the crwn is an umbrellashape, which is the mst effective shape fr matter prductin under shaded cnditins (Hrn 1971; Khyama 198; Ktippers 1989), CS . The CS decreased mre rapidly fr P, jezensis with decreasing RHGR than A, sachalinensis (analysis f cvariance, P .1), indicating that P, jezensis tended t have an umbrella-shape crwn under pr cnditins fr grwth. 6 3 A.sachalinensis Lii 5 i P.jezensis 3-5 cm nnd UU J t t i B. ermanii 3-5 cm _il,,jl cm 5-1 cm 5-1 cm / e- 3 Z d i i i i i i cm 1-2 cm 1-2 cm 3 nhndw i :1 6-3 :2,1 6 Lg (relative height grwth rate, cm) Fig. 3. Frequency distributins f relative height grwth rate (RHGR) f saplings during the last 5 years. Upper, middle and lwer rws shw RHGR fr saplings f 3-5, 5-1 and 1-2 cm in height, respectively. Slid parts indicate saplings beneath canpy gaps and pen parts shw thse under the clsed canpy. 338 Y. Kubta et al. -~- - 8 X CD A.sachalinensis 9 P.jezensis %q~ c ~ - A.sachalinensis { 9 ) c- O e'~ ell t- ~ /~ 'O 9 / / / / 9 O 2 1 O -3 -t -i Lg (relative height grwth rate, cm) Fig. 4. Relatinship between relative height grwth rate (RHGR) and crwn shape index (CS) fr saplings P.jezensis Waiting pattern f understrey trees 2- { ) ndividuals f A. sachalinensis were released at a significantly higher layer (max. 7 m) than thse f P. jezensis (max. 3 m; Mann-Whitney U-test, P .1; Fig. 5). This suggests that understrey trees f A. sachalinensis wait fr gap frmatin at higher layers, while thse f P. jezensis wait at lwer layers. Percentages f gap successrs (GSER after Leemans [1991]) fr each DBH dass were examined, and an assciatin with gaps was tested by the Z 2 test f gdness f fit (the null hypthesis is that the rati f the gap successrs t waiting trees under the clsed canpy is 19.3 : 8.7, accrding t the u} 8 {D') 13._ P.gle {} 5 4 E 6.5.~, ~ 3.5 g 2.5- f,, B 1.5 n-.5 [] P.jezensis 9 A.sachalinensis ' ' ' ' ',, ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' ' N. f understrey trees Fig. 5. Frequency distributin f release heights fr understrey trees. Abbreviatins as fr thse in Fig D.B.H. (cm) { } Fig. 6. Changes in the rati f understrey individuals ccurring in canpy gaps t that f ttal understrey individuals with DBH class. Slid symbls shw significant ccurrence (P .5) in the canpy gaps. The data pints in parentheses are mitted because f uncertainty f appearance in canpy gaps r beneath the clsed canpy. Stand structure f sub-breal cniferus frest 339 actual canpy gap rati; Fig. 6). The density f understrey trees f P. jezensis and A. sachalinensis was 26 and 153 per hectare, respectively. Understrey trees f A. sachalinensis demnstrated n significant assciatin with gaps, even in large size classes (DBH = 8 cm), while P. jezensis shwed a significant gap assciatin (P .5) with increasing DBH class. Understrey trees f P. jezensis shwed greater height grwth with increasing DBH dass, while A. sachalinensis had n such tendencies (Fig. 7a). The cefficients f variatin (CV) f height grwth rate f A. sachalinensis increased with DBH class, but the CV f P. jezensis, which was smaller than that f A. sachalinensis, had n relatin t DBH class (Fig. 7b). Skewness f height grwth rate f A. sachalinensis shwed cnstantly negative values in all DBH classes (Fig. 7c). Hwever, that f P. jezensis changed frm psitive t negative at 2-6 cm DBH, which was the size significantly assciated with gaps (Fig. 6). The dead stem rati f A. sachalinensis, which can be regarded as estimated mrtality, increased in 2-6 cm DBH classes, then declined. Hwever, that f P. jezensis decreased cnstantly with DBH class (Fig. 7d). DSCUSSON (a) 7 ~ 6.m 5 g 4 ~ 3 Z- ~ lo 1 8O 6 d 4 () 2 2 r- ~ O 3 E N -1-2 (b) (c) (d) 5 4 A. sachalinensis P.jezensis ii i l r S q./%_ [ Veblen (1986) explained the cexistence mechanism f Picea engelmanii and Abies lasicarpa based n percentage frequencies f fallen trees and ccupants f treefall sites, which represent mrtality and recruitment rate, respectively. The present results were similar: the damage rati f canpy trees f A. sachalinensis exceeded that f P. jezensis, and A. sachalinensis ccurred in a greater number f understreys. Hwever, the abundance f understrey trees prir t gap creatin des nt necessarily lead t a high prbability f filling canpy gaps (Lertzman 1992). T accept the canpy recruitment rate as a parameter fr cexistence is essential in elucidating the grwth and mrtality patterns f understrey trees. These patterns f understrey trees indicate that A. sachalinensis can grw int higher layers even under suppressed cnditins while P. jezensis requires canpy gaps fr height grwth (Figs 5-7a). These different patterns between P. jezensis and A. sachalinensis understrey may be regarded as the 9 3 O 6 2 ~ lo ~ (D _ i l D.B.H. (cm) D.B.H. (cm) Fig. 7. (a) Relatinships beteen DBH class and height grwth in a 2 year perid fr understrey trees. Symbls shw means + s.d. (with bar) f height grwth. (b) The relatinships between DBH class and cefficient f variance f the distributin f height grwth rate in each DBH class fr understrey trees. (c) The relatinships between DBH class and skewness f the distributin f height grwth rate in each DBH class fr understrey trees. (d) Changes in dead stem rati f understrey trees with DBH class. The dead stem rati was calculated by dividing the number f dead stems by the sum f living and dead stems. tw extremes f life-histry traits against shade (Canham 1989). Piceajezensis has shade tlerance and individuals can persist in the understrey, but 34 Y. Kubta et al, appredable net grwth ccurs nly after frmatin f gaps, while A. sachalinensis can grw slwly and cnsistently even beneath a dsed canpy (Canham 1988c). Light cnditins fr understrey trees are variable crrespnding t the vertical distributin f light. n the canpy layer, light intensity decreases expnentially frm the tp t the lwest branches with increasing cumulative fliage mass (Mnsi & Saeki 1953). Hwever, Hrn (1971) and Kikuzawa el al. (1988) reprt that the light cnditins in the understrey are prest beneath the umbra due t the lwest branches f canpy trees, then becme better away frm the canpy layer. This implies that understrey trees f P. jezensis suffer less frm suppressin by canpy trees than A. sachalinensis and had a lwer mrtality because f waiting at a lwer layer. Hwever thse f A. sachalinensis suffer higher mrtality because f shading by lwer branches f canpy trees (Fig. 7d). Picea jezensis gives pririty t surviving and A. sachalinensis t grwing in the understrey. Picea jeznensis and A. sachalinensis exhibit different life-histry traits. Piceajezensis has a greater life span ( 3 years) and begins t prduce cnes at larger sizes (DBH 4 cm), and A. sachalinensis has a shrt life span ( 1-2 years) and begins t prduce cnes at smaller sizes (DBH 2 cm; Hiura etal. 1992; Kubta unpubl, data). Cnsidering the grwth and mrtality patterns
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks