Views
5 years ago

Get PDF (786K) - Wiley Online Library

Get PDF (786K) - Wiley Online Library

Roeder, M. et al. Traits

Roeder, M. et al. Traits and growth of liana regeneration in primary and secondary forests Axis 2 (27%) Axis 2 (27%) 1.0 0.5 0.0 –0.5 –1.0 –1.0 –0.5 0.0 Axis 1 (29%) 0.5 1.0 3 2 1 0 –1 –2 a1 a2 species plot (mean) plot (SD) 1.0 b1 1.0 c1 –3 –3 –2 –1 0 1 2 3 4 5 Axis 1 (29%) length. The second PCA axis explained 27% of variation and was mainly determined by WLR and plant length. Most primary forest species occurred in one cluster that was characterized by round leaves and short length. Secondary forest species were more scattered in the ordination plot than primary forest species. One group of secondary forest species (including one primary forest species, Cheiloclinium hippocrateoides) could be separated and described as tall plants with abundant, large, elongated leaves. Three species from primary and secondary forest (Dicranostyles scandens, Maripa glabra, Aristolochia rumicifolia) formed a distinct, isolated group: they all had long, delicate, winding stems (low ratio of slenderness) with scarce foliation. Traits per plot Influence of forest type and canopy cover The mean values of traits per plot differed between forest types in four cases: liana regeneration had longer shoots and larger leaves in Cecropia forest than in primary forest; plants had more leaves per unit length in the Vismia forest than in primary forest; and herbivory was higher in Cecropia than in Vismia forest (all F 4 3.7, P o 0.05, Table 3). The variability of traits within a plot (SD) was significantly smaller for plant length and leaf size in the primary forest than in the two secondary forest types. SLA Axis 2 (24%) Axis 2 (24%) 0.5 0.0 –0.5 –1.0 –1.0 –0.5 0.0 Axis 1 (33%) 0.5 1.0 4 2 0 –2 –4 b2 –6 –3 –2 –1 0 1 2 3 4 5 Axis 1 (33%) Axis 2 (18%) Axis 2 (18%) 0.5 0.0 –0.5 –1.0 –1.0 –0.5 0.0 Axis 1 (37%) 0.5 1.0 3 2 1 0 –1 –2 c2 –3 –4 –3 –2 –1 0 1 2 3 Axis 1 (37%) Fig. 2. Three principal components analyses of seven traits in 26 liana species for trait per species (a1, a2), trait per plot (b1, b2) and variability of traits per plot (standard deviation, SD) (c1, c2). Upper graphs (a1-c1) show loading of traits for the first and second PCA axes; lower graphs (a2-c2) show scores of species or plots for the first and second PCA axes. Species abbreviations are found in Appendix S1. WLR = width–length ratio of a leaf, SLA = specific leaf area. RSS = ratio of stem slenderness. Main habitat of species is indicated by different symbols. Plots in the same forest type are marked with the same symbol. Locations of the plots are indicated by letters: c (Closso), d (Dimona), g (Gaviao), pa (Porto Alegre). was greater in Vismia than in primary forest (all F 4 5.2, P o 0.05, Table 3). Canopy cover and trait variables correlated in two cases: herbivory increased with higher canopy cover (r = 0.42, P = 0.041) and the variability of leaf size was higher when canopy cover decreased (r = 0.48, P = 0.017, SD leaf size log-transformed) (Appendix S2). Ordination of plot traits The ordination of traits at plot level (mean per plot) gave a result similar to the ordination of species traits, although axes were rotated by 90% (Fig. 2b). The first axis (accounting for 33%) was determined mainly by plant length, leaf size and WLR, while the second axis (accounting for 24%) was determined by leaves per unit length. Most primary forest plots formed one cluster, characterized by a trait combination of short length, small and round leaves and few leaves per unit length. Five of six Cecropia forest plots had intermediate values for most traits and therefore clustered in the middle of the ordination, whereas most Vismia forest plots were widely scattered in the ordination. Ordination of plot trait variability The first axis explained 37% of variation, the second axis 18% (Fig. 2c). Standard deviation of plant length, SLA Applied Vegetation Science Doi: 10.1111/j.1654-109X.2011.01152.x © 2011 International Association for Vegetation Science 113

Traits and growth of liana regeneration in primary and secondary forests Roeder, M. et al. Table 3. Differences in traits of liana regeneration for three forest types in the Central Amazon, Brazil. Mean trait values per plot and the variability per plot (SD) were tested using ANOVAs and Tukey post-hoc HSD tests. WLR = width–length ratio of a leaf, SLA = specific leaf area. RSS = ratio of stem slenderness. SLA and leaf size were log-transformed prior to analysis. Sample sizes were: n = 9 for primary forest (PF), n = 7 for Cecropia forest (C), n =8 for Vismia forest (V). Significant results (P o 0.05) are in bold. Trait Mean per plot Variation per plot (SD) F2, 21 P Differences in post-hoc test F 2, 21 P Differences in post-hoc test Plant length 4.720 0.020 C 4 PF 16.031 4 0.001 C,V 4 PF Leaves per length 3.658 0.043 V 4 PF 2.635 0.094 – Leaf size (log) 3.941 0.035 C 4 PF 6.428 0.006 C,V 4 PF WLR 3.244 0.059 – 2.151 0.141 – SLA (log) 0.907 0.419 – 5.200 0.015 V 4 PF Herbivory 4.938 0.017 C 4 V 2.147 0.142 – RSS 0.188 0.830 – 1.745 0.199 – and ratio of stem slenderness correlated negatively with the first axis; standard deviation of herbivory and WLR mainly determined the second axis. Primary forest plots were grouped on account of their low variability of traits, whereas standard deviation of traits in secondary forest varied widely. Plots of the two secondary forest types were not separated along the first two axes. Plots of the same site and forest type were usually not clustered, but this could be observed for Cecropia forest plots at the Colosso site. Relative growth rate Relative growth rate per species The relative growth rate (RGR) of the 22 studied species was negatively correlated with the initial plant length (r = 0.512, P = 0.015). No other correlation involving RGR were uncovered. RGR for most primary forest species was about zero, and RGR did not differ between secondary and primary forests species (F = 0.007, P = 0.934). Davilla kunthii had the highest relative growth rate (0.4 cm cm 1 yr 1 ). One secondary forest species (Adenocalymna subicanum) and one primary forest species (Abuta imene) had negative relative growth rates. When ordered according to standard deviation, most primary forest species were on the side of low variability and secondary forest species showed higher variability (Fig. 3). Four of the five species with the highest variability were secondary forest species. This separation of secondary and primary forest species regarding variability of RGR was confirmed by an ANOVA (F1,18 = 5.41, P = 0.032). Smilax syphilitica and Memora moringiifolia could not be assigned to secondary or primary forest and therefore were excluded from this ANOVA. Relative growth rate per plot Relative growth rate per plot was different among the three forest types. The Vismia forest plots had higher relative growth rate (yr –1 ) 1.5 1.0 0.5 0.0 –0.5 –1.0 –1.5 growth rates than plots in primary and Cecropia forest (F = 20.33, P o 0.001). The variability of RGR per plot was lower in the primary forest than Vismia forest (F = 7.06, P = 0.005). Canopy cover was negatively correlated with RGR (r = 0.507, P = 0.014) and but not with the standard deviation of RGR (r = 0.396, P = 0.061) (Fig. 4). When tested in an ANCOVA, both the covariate canopy cover and the forest types had a significant influence on relative growth rates (Fcovariate = 17.49, P o 0.001; F = 15.82, P o 0,001). Discussion Mem mor Abu san Mar gla Amp ama Che hip Rou cus Pau SpA Mem ade Abu ruf Mac sp1 Dav spH Sen tap Der flo Mac hoe Ade sub Smi syp Dol den Dav kun Dic sca Abu ime Odo spA Mik sp. Fig. 3. Relative growth rate (mean and SD per species) of 22 Amazonian liana species, ordered according to increasing standard deviation. The main habitat of each species is indicated by symbols. See Appendix S1 for species names and sample size. Distinctness of primary forest Traits of lianas in primary forest could be separated from traits in the two secondary forest types at the species and community levels. Primary forest communities had typical shade-tolerant traits and very low variability of traits, i.e. regeneration in all such plots appeared very similar regarding their traits. The similar pattern of Applied Vegetation Science 114 Doi: 10.1111/j.1654-109X.2011.01152.x © 2011 International Association for Vegetation Science

Get PDF (72K) - Wiley Online Library
Get PDF (9212K) - Wiley Online Library
PDF(711K) - Wiley Online Library
Get PDF (205K) - Wiley Online Library
PDF(1418K) - Wiley Online Library
PDF(887K) - Wiley Online Library
Michael Hensel; Frei Otto; - Wiley Online Library
Saving Mothers Lives - Wiley Online Library
Information behavior realities in organizations - Wiley Online Library
The nth-order stellar hydrodynamic equation - Wiley Online Library
PDF Download Richard Wentworth: Making Do and Getting By Ebook | READ ONLINE
The issue in PDF - Scandinavian Library Quarterly
Malaria Guidelines April 07 (for online).pdf - NT Health Digital Library
PDF Download Colors of Loss and Healing: An Adult Coloring Book for Getting Through Tough Times Ebook | READ ONLINE
PDF Download Poems About Sculpture (Everyman's Library Pocket Poets Series) Ebook | READ ONLINE
Online [PDF] CFS Unravelled: Get Well By Treating The Cause Not Just The Symptoms Of CFS, Fibromyalgia, POTS Related Syndromes - All Ebook Downloads
Read Online (PDF) Osler s Bedside Library: Great Writers Who Inspired a Great Physician - All Ebook Downloads
PDF Download The Death and Life of Great American Cities: 50th Anniversary Edition (Modern Library) Ebook | READ ONLINE
Get PDF - Wiley Online Library
Get PDF - Wiley Online Library
Get PDF - Wiley Online Library