Effects of experimental drought on hydraulic properties and leaf traits of upper canopy and understory tree species in a perhumid tropical forest in Central Sulawesi, Indonesia [Elektronische Ressource] / vorgelegt von Bernhard Schuldt

Bernhard Schuldt

Sujets

• Biologie

Informations

• Publié par: georg-august-universitat_gottingen
• Publié le: 01 janvier 2010
• Nombre de lectures: 6
• Langue: English
• Poids de l'ouvrage: 3 Mo

Extrait

EFFECTS OF EXPERIMENTAL DROUGHT ON HYDRAULIC
PROPERTIES AND LEAF TRAITS OF UPPER CANOPY AND
UNDERSTORY TREE SPECIES IN A PERHUMID TROPICAL
FOREST IN CENTRAL SULAWESI, INDONESIA



DISSERTATION
zur Erlangung des Doktorgrades
der Mathematisch-Naturwissenschaftlichen Fakultäten
der Georg-August-Universität zu Göttingen





vorgelegt von
Bernhard Schuldt
aus Vejle, Dänemark




Göttingen, 21.09.2010






























Referent: Prof. Dr. Christoph Leuschner
Korreferent: Markus Hauck

Tag der mündlichen Prüfung: 27. - 28.11.2010
SUMMARY

The remaining tropical moist forests may be threatened in future by more frequent and more
severe droughts that come along with the predicted climate change in South-East Asia and
South America, though ecosystem consequences of strong drought events are hardly
predictable. Therefore, manipulative field experiments are needed to identify gradual
ecosystem responses and threshold values of ecosystem functions under a changing climate.
Tropical drought experiments have so far only been conducted in seasonal dry forests in East
Amazonia, where the biota most likely possess specific adaptations to regular dry spells.
Experiments on the drought response of perhumid tropical forests with continuously high soil
moistures and air humidity do not yet exist.
Both observational studies on natural drought events and the Amazonian throughfall
displacement experiments showed that under prolonged drought especially large and tall
canopy trees experienced higher mortalities than trees with smaller size.
We carried out a replicated throughfall displacement experiment in a perhumid premontane
old-growth forest stand in Central Sulawesi, Indonesia, with annual precipitation rates of
more than 2500 mm and constantly high relative air humidity close to saturation. We assumed
that tree species of this forest do not possess adaptations to severe drought (e.g. deep-reaching
roots) compared to the Amazonian experiments.
The purpose of this study was twofold. First, we aimed at explaining why tall tropical trees
may possess higher mortalities after extended droughts than smaller ones. Secondly, we
analyzed the morphological and physiological responses of an abundant tall-growing upper
canopy tree species to 24 months of throughfall displacement, which resulted in a reduction of
soil moisture content in the upper soil layers below the conventional wilting point.
Three hypotheses were formulated that concerned tree physiological and morphological
adaptations to large tree size, and the response of trees from tropical moist forests to soil
water shortage. The study aims were to test whether (i) the environmental control of sap flux
density is directly related to tree height, (ii) tropical trees adapt their hydraulic architecture
when growing tall to counteract the effect of growing hydraulic resistance with increasing
flow path length, and (iii) tall trees of the premontane forest in Central Sulawesi are adapted
to the prevailing perhumid conditions and thus are more vulnerable to prolonged soil water
deficits than species from tropical humid or semihumid forests.
To achieve these goals, a wide range of ecophysiological, morphological and anatomical traits
were investigated in mature trees. Key parameters measured were several hydraulic properties
of the xylem of twigs and trunks, wood anatomy, leaf morphology and foliar nutrient
contents, stable isotope ratios of C, N and O, sap flux density, litter fall and stand
microclimatic variables.
We found evidence that co-occurring tropical tree species differ strongly in measured xylem
sap flux densities in the trunk, which is largely dependent on the canopy position within the
forest stand. Despite the perhumid climate, vapor pressure deficit (VPD) was the most
important environmental factor controlling sap flow. Mean VPD increased linearly with
height in the canopy. The close relation between sap flux density and tree height in this
perhumid forest, irrespective of systematic position, may be interpreted as convergent pattern
in the water use of tropical trees.
We found several important changes in the hydraulic architecture with tree height in the eight
studied species. Vessels were tapering acropetally from the stem base towards the upper
canopy in a tall-growing tree species, and the smallest vessels were found in all species in the
distal twigs. Tall trees generally possessed the largest vessels along the whole flow path. The
vessel diameter showed an optimum curve with maximal diameters found in the trunk and not
in the roots. Leaf-specific and sapwood-area specific conductivity increased with tree height;
both conductivities were linked closely to the increase in vessel diameter.
The most abundant upper canopy tree species of this forest (C. acuminatissima) did not show
signs of critical damage after 24 months of soil desiccation, despite the fact, that the hydraulic
conductivity of twigs and trunks decreased due to smaller vessel diameters in the most recent
xylem, the number of leaves on distal twigs was lowered, and stem diameter growth was
reduced (non-significant tendency) in the trees exposed to soil desiccation. We assume that
the prevailing low evaporative demand throughout the experiment in this perhumid climate
prevented critical damage to occur, despite soil desiccation beyond the conventional wilting
point. Nevertheless, the reduction in sap flux densities in the desiccation period was more
pronounced in taller trees than in smaller ones, indicating that drought-induced physiological
effects should appear earlier in tall than in smaller trees. Stem diameter growths, the diameter
growth of xylem vessels, and leaf bud formation were found to be particularly sensitive
growth processes in C. acuminatissima, while pre-senescent leaf shedding or canopy dieback
were not observed.
We conclude that tall trees in this forest stand possess a number of morphological and
physiological traits that distinguish them clearly from trees in the lower strata. We assume
that (a) the exposure to a higher evaporative demand in the upper canopy, (b) the inevitable
increase in hydraulic resistance in a longer flow path, and (c) the wider vessels at the base of
the trunk, that result from the longer flow path, are the most important causes of the reported
higher mortality rates of tall tropical trees after prolonged drought.




TABLE OF CONTENTS

1 INTRODUCTION .................................................................................................................. 1
1.1 Impact of Climate change on tropical rainforests ...................................................... 2
1.2 Influence of drought on trees ..................................................................................... 4
1.2.1 Carbon starvation hypothesis.............................................................................. 4
1.2.2 Xylem embolism................................................................................................. 5
1.2.3 Conduit anatomy and cavitation risk.................................................................. 7
1.2.4 Wood density...................................................................................................... 8
1.2.5 Rooting depth ..................................................................................................... 9
1.2.6 Tree size............................................................................................................ 10
1.3 Tropical Throughfall Displacement Experiments.................................................... 12
1.4 Project objectives 15
1.5 References ................................................................................................................ 16

2 METHODOLOGY............................................................................................................... 25
2.1 Characterization of the study area............................................................................ 26
2.1.1 Study site .......................................................................................................... 26
2.1.2 Climate.............................................................................................................. 26
2.1.3 Soils and root distribution................................................................................. 27
2.2 Experimental design................................................................................................. 28
2.3 Field setup, instrumentation and methods................................................................ 30
2.3.1 General components of the field setup ............................................................. 30
2.3.2 Summary of the measured parameters.............................................................. 31
2.3.3 Sap flow measurements.................................................................................... 31
2.3.4 Derivation of the empirical sap flux velocity c