http://hdl.handle.net/10125/495 Fri, 24 May 2013 17:00:50 GMT 2013-05-24T17:00:50Z http://scholarspace.manoa.hawaii.edu:80/bitstream/id/1744/PacSci.jpg http://hdl.handle.net/10125/495 http://hdl.handle.net/10125/12587 Sat, 01 Jan 1983 00:00:00 GMT http://hdl.handle.net/10125/12587 1983-01-01T00:00:00Z http://hdl.handle.net/10125/750 Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/750 1983-10-01T00:00:00Z Mueller-Dombois, Dieter http://hdl.handle.net/10125/749 A conceptual model that provides a means of viewing changes in vegetation as responses to weathering of soil mantles over thousands of years has been developed from investigations of vegetation communities on a soil chronosequence on sand dunes extending back in time to at least the last interglacial. Progressive and retrogressive phases in natural plant succession are indicated by this sequence. A similar model is used to postulate that the various canopy dieback symptoms and agents observed in the New England tablelands of New South Wales, Australia, represent the mechanisms by which changes are occurring in retrogressive ecosystems that have been disturbed by human land use. We suggest that investigations of the etiology of diebacks need to be made in the context of temporal changes in soil fertility associated with the degree of soil weathering. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/749 1983-10-01T00:00:00Z Walker, J; Thompson, CH; Jehne, W http://hdl.handle.net/10125/748 Symptoms in the crowns of Eucalyptus obliqua and E. regnans trees affected by "regrowth dieback" are reevaluated and amplified using tree climbing, tree felling, and optical study methods. A distinction is made between the dying upper branches and the lower crown of dieback-affected trees. Declining branches showed an increase in mortality or absence of growing shoots, naked buds, and accessory buds with increase in dieback severity. Mean distance between leaf scars indicated that growth rate of shoots was slowed before death. With increase in dieback severity of a branch, leaf size and leaf area index were reduced. Leaves on dieback-affected branches showed increased chlorosis and reddish colors. There are common features linking dieback-affected shoots with shoots showing mild "witches brooming."Twiginhabiting insects are considered of only secondary importance. There was no evidence of any culturable microorganism from the vascular system. The symptoms could be the result of water deficits and high leaf temperatures. Alternatively, the symptoms are suggestive that microbes such as a virus or more especially a mycoplasma are involved. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/748 1983-10-01T00:00:00Z Plazer, C http://hdl.handle.net/10125/747 Species of the woody root rot fungus Armillaria are indigenous in cool temperate rain forest, mixed forest, and wet and dry sclerophyll eucalypt forests in Australia. Four species have been described or identified from southeastern Australia: A. luteobubalina Watling and Kile, A. fumosa Kile and Watling, A. hinnulea Kile and Watling, and A. novae-zelandiae (Stevenson) Boesewinkel. The latter species was first described from New Zealand, and A. hinnulea also occurs in that country. Armillaria novae-zelandiae and A. hinnulea occur in wet forests (rain forest, mixed forest, and wet sclerophyll communities), while A. luteobubalina and A. fumosa are found mainly in dry sclerophyll forests. Armillaria luteobubalina is so far the only species known to behave as a primary pathogen in native forests. While the fungus has an extensive geographical distribution in southeastern Australia, damage is most severe in selectively logged forests in the central highlands of Victoria, where it is estimated that approximately 3-5% of the forest area is moderately to severely affect~. The fungus kills all species of eucalypts and a wide range of the under- story trees and shrubs present in the forests. Most infections occur in small (0.1-1.0 ha), well-defined patches, but larger (up to 20-30 ha), more diffuse infections also occur. Evidence of primary pathogenicity includes (a) constant association of the fungus with disease; (b) the pattern of disease development within stands (the fungus spreads by root contact from infected food bases); (c) correlation between root infection and symptom development in large trees; (d) evidence of host resistance to infection; and (e) pathogenicity in pot and field trials. There is no evidence that climatic stress or other pests or pathogens initiate disease. Within the forest, the fungus has a discontinuous distribution. Studies of genotypes of the fungus (identified by analyses of mating alleles, since Armillaria sp. are bifactorial heterothallic, or intraspecific antagonism), suggest that A. luteobubalina consists of a community of genetically distinct mycelia. Individual genotypes may contract, expand, or coalesce, depending on circumstance. The development and status of the community depends on the two processes of new basidiospore infection and local spread by vegetative growth through root systems. Similar patterns of genotype distribution and clonal development were evident in logged and unlogged forest. It is concluded that root rot caused by the fungus is endemic in these forests but that logging has aggravated the disease. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/747 1983-10-01T00:00:00Z Kile, GA http://hdl.handle.net/10125/746 The phenomenon of patch dieback in stands of Nothofagus on Mt. Giluwe, in the southern highlands of Papua New Guinea, is described. Ecological studies that have been carried out are reviewed. Based upon the results of surveys on Mt. Giluwe, the role of Phytophthora cinnamomi in contributing to the dieback is examined. A comparison is made of Nothofagus dieback on Mt. Giluwe with Metrosideros dieback in Hawaii. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/746 1983-10-01T00:00:00Z Arentz, Frans http://hdl.handle.net/10125/745 Old and young trees of southern rata (Metrosideros umbel/ata) were artificially defoliated to examine their response to foliage loss. Partial (50%) or total defoliation before budbreak depressed shoot growth more markedly in old than in young trees. Fifty percent leaf loss shortly after budbreak or at the end of the growing season did not significantly impair shoot growth in either age class. By contrast, total postbudbreak defoliation resulted in the death of most shoots in both old and young trees. While the terminal buds of both old and young trees totally defoliated in late autumn showed good overwinter survival, only in the young trees were there sufficient energy reserves to enable some of these shoots to flush the following season. Young trees proved to be more resilient to foliage loss, suggesting a more positive carbon balance (energy surplus) in young trees than in older trees. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/745 1983-10-01T00:00:00Z Payton, IJ http://hdl.handle.net/10125/744 During the 1950s, dieback of southern rata (Metrosideros umbel/ata) and kamahi (Weinmannia racemosa) was documented in the conifer/ hardwood forests of the Kokatahi and Fox catchments in Westland, New Zealand. Thirty years later, southern rata was usually absent as a live tree in dieback stands. Kamahi, although absent from the canopy in the Kokatahi, had partially recovered at Fox, where it was still a dominant canopy species. Regeneration studies on a range of sites indicate that in the short term, kamahi and Quintinia acutifolia will become the structurally dominant canopy species in many of the dieback stands, and southern rata will at best be a minor component. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/744 1983-10-01T00:00:00Z Allen, RB; Rose, AB http://hdl.handle.net/10125/743 Many researchers in New Zealand have accepted equilibrium models of vegetation change that assume within-stand self-replacement of the dominant tree species as the norm. Consequently, many discontinuous stand structures have been used as evidence of forest instability. For example, the patterns of regeneration and mortality in the rata-kamahi forests of Westland have led many to believe that the present canopy tree mortality is excessive. As a result, there has been considerable research on browsing by the introduced brush-tailed possum as the primary cause of the mortality. We suggest that any interpretation of this forest pattern must include a consideration of the influences on the vegetation of natural disturbances. Abundant evidence suggests that at least some of the mortality is due to senescence of cohorts of trees that originated at approximately the same time after events such as windthrow and mass movements. It may be that browsing by possums hastens the death of trees already susceptible as a result of natural stand development processes. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/743 1983-10-01T00:00:00Z Stewart, Glenn H; Veblen, Thomas T http://hdl.handle.net/10125/742 Browsing by the introduced Australian brush-tailed possum (Trichosurus vulpecula) has been generally accepted in New Zealand during the past 30 yr as the principal cause of damage to hardwood forests dominated by rata (Metrosideros spp.) and kamahi (Weinmannia racemosa). Recently, this view has been challenged, and assertions have been made that the forests in a North Island tract were in poor order before possums invaded, that their natural collapse was only a matter of time, and that there is ample evidence relating the more dramatic changes in South Island forests to geological and meteorological events. In this paper, the evidence for repeated coincidence of increase of possum numbers and spectacular modification of the forests is reexamined. Such coincidences, the continued good health of montane hardwood forests where possums have not attained high numbers, and some experimental data, lead to the conclusion that the possum is responsible for dieback in rata-kamahi forests. Control of possums is therefore vital if the forests are to be maintained. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/742 1983-10-01T00:00:00Z Batcheler, CL http://hdl.handle.net/10125/741 Tropical cyclones usually form between 10° and 20° latitude but frequently move as far south as New Zealand. Cyclone Bernie, in April 1982, caused extensive damage in central North Island forests. Four other severe tropical cyclone s since 1936, are known to have caused damage to indigenous forests throughout the North Island and in parts of the South Island. Severe storms of extratropical origin also affect New Zealand, and many also result in significant forest damage. The storm regime to which New Zealand is subject is severe enough so that storms themselves could be a major factor in molding stand composition and structure in many, or even most , parts of the country. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/741 1983-10-01T00:00:00Z Shaw, WB http://hdl.handle.net/10125/737 Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/737 1983-10-01T00:00:00Z http://hdl.handle.net/10125/736 Dieback has been observed in New Zealand Nothofagus forests for some time, and a number of causal factors have been recognized. Some understanding of the effects of dieback on forest structure has been gained in a study of events after snowfalls had caused partial damage to an area of mountain beech forest. The results of this study are used to interpret the structure of beech forests elsewhere in New Zealand. Thu, 01 Sep 1983 00:00:00 GMT http://hdl.handle.net/10125/736 1983-09-01T00:00:00Z Wardle, JA; Allen, RB http://hdl.handle.net/10125/735 Accelerated mortality is attributed to an unusually high percentage of old trees, an abundance of pathogenic fungi, and a putative lowering of water tables in the 1960s. There is some evidence to suggest that this may be a cyclical phenomenon. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/735 1983-10-01T00:00:00Z Skipworth, JP http://hdl.handle.net/10125/734 Areas of vegetation mortality have been recognized in the upland forests of the Kaimai Ranges, North Island, New Zealand. The mortality is not confined to any particular vegetation type, but occurs above a critical altitude that differs at different points along the ranges. The mortality is episodic and has occurred at intervals over a period of several hundred years, with severe droughts implicated as the causative factor. The numerous mortality events have resulted in forests of many age classes where trees do not reach potential maturity. The coincidence between the occurrence of cloud and the mortality is suggested to make the forests drought-sensitive through poorly developed root systems caused by prolonged soil waterlogging. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/734 1983-10-01T00:00:00Z Jane, GT; Green, TGA http://hdl.handle.net/10125/733 Studies of climatic fluctuations on the island of Hawai'i have been undertaken using data from 113 climatological stations on the windward slopes of Mauna Kea, Mauna Loa, and Kilauea. Both monthly rainfall and mean monthly air temperature data have been analyzed for periods ranging up to 91 yr. In addition, an estimation scheme for mean monthly air temperatures on the island of Hawai'i has been developed. The climate diagram technique of Walter (1971) has been used to relate rainfall and temperature. Annual climate diagrams have been compared to median climate diagrams as an index of periods of relative drought or wetness. Highly significant spatial uniformity in climate fluctuation patterns is found over the study area. The resulting patterns are discussed in relation to observed patterns of '6hi'a (Metrosideros polymorpha) dieback in this area of Hawai'i. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/733 1983-10-01T00:00:00Z Evenson, William E http://hdl.handle.net/10125/732 'Ohi'a (Metrosideros polymorpha) is the dominant rain forest tree species in Hawai'i. It has long impressed botanists for its morphological variability and ecological amplitude. The present study has documented differences in the distribution of Metrosideros polymorpha varieties in populations of 'ohi'a growing on adjacent young and old flows. Pubescent varieties are present on young volcanic substrates, but tend to be absent from older soils in the rain forest zone. The genetic basis for these varieties has been confirmed in a common garden experiment. The water relations of pioneer and older forest '6hi'a have been examined and shown to differ, so that the pioneer 'ohi'a tend to be able to maintain turgor at lower relative water contents than the glabrous varieties. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/732 1983-10-01T00:00:00Z Stemmermann, Lani http://hdl.handle.net/10125/731 A soil and foliar nutrient analysis was carried out in the Hawaiian Metrosideros rain forest for the purposes of elucidating a previously published physical habitat classification and for finding an explanation for the widespread canopy dieback, which is not caused by a biotic agent in this ecosystem. Soil elements analyzed were C, N, P, Ca, AI, Mn, Fe, and other parameters such as pH. Foliar analysis was restricted to N, P, Ca, Mn, and Fe and assessed only for the canopy M. polymorpha and its major associate, the tree ferns, in the undergrowth. It was found that the earlier recognized two-way breakdown into oligotrophic and eutrophic nutrient regimes had to be enlarged for the habitat classification to a mesotrophic category. Young, deep-ash soils were recognized as mesotrophic on account of their moderate pH and high Ca levels, but they showed severe limitations in available nitrogen and phosphorus that were reflected also in the foliage. Intermediate-aged (1000 yr) ash soils were found to be nutritionally well supplied and balanced and thus considered eutrophic, while older ash soils were designated as oligotrophic on account oftheir extreme acidity and associated higher and potentially toxic levels of soluble AI, Mn, and Fe. Except for the mesotrophic 'a'a lava site, other shallow rock outcrop habitats on pahoehoe were recognized as oligotrophic, because of very low available N and high acidity and, in the poorly drained sites, also high levels of potentially toxic metals. The bog habitats had the highest levels of potentially toxic metals, but surprisingly moderate levels of available N and P. The canopy dieback phenomenon cannot be entirely explained from the nutrient imbalances found in most of the substrate types, but these imbalances are considered contributory in the sense that they become increasingly stressful with stand development, thereby decreasing tree vigor and predisposing stands to dieback. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/731 1983-10-01T00:00:00Z Balakrishnan, Nadarajah; Mueller-Dombois, Dieter http://hdl.handle.net/10125/730 Approximately 50,000 ha of native '6hi'a (Metrosideros polymorpha Gaud.) forest on the island of Hawai'i experienced a drastic reduction (dieback) of the tree canopy between 1954 and 1977. Concern for the management of this important native ecosystem has stimulated a great deal of research on the Hawaiian dieback phenomenon. In this paper we address the question of changes in the '6hi'a population after an area has experienced dieback. Since 1976, we have established 62 400-m2vegetation sampling plots throughout the dieback and adjacent nondieback forest areas on the island of Hawai'i. The tall tree vigor and '6hi'a seedling and sapling growth were resamp1ed in 26 of these study plots in 1982. The results of the reassessment of the '6hi'a populations indicate that the forest dieback has not spread appreciably since 1977. However, nearly all the plots located in areas that originally experienced a drastic reduction of the tree canopy cover were found to have a large number of '6hi'a seedlings and saplings. Based on this apparent high level of regeneration following the initial canopy loss, we speculate that most of the forest dieback areas will again develop a closed, tall-statured '6hi'a tree canopy. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/730 1983-10-01T00:00:00Z Jacobi, James D; Gerrish, Grant; Mueller-Dombois, Dieter http://hdl.handle.net/10125/729 Massive tree dieback has occurred periodically in the Hawaiian montane rain forest. The species mainly involved is Metrosideros polymorpha, which is the prevailing upper canopy tree species throughout most of this forest on all high Hawaiian islands. The canopy dieback occurs in stands over the entire spectrum of sites, from well-drained lava flows over nutritionally rich volcanic ash to permanently wet bogs with toxic soils. A biotic agent could not be found to cause this dieback. Five main dieback patterns have been recognized, and all are site-specific. These patterns suggest certain causal mechanisms, but they explain only a fraction of the dieback syndrome. A number of additional facts were established which have led to a new dieback theory involving a chain-reaction process: (1) cohort senescing as a predisposing factor; (2) a dieback trigger, which can be either internal (a species characteristic) or external (a fluctuating and recurring site-specific perturbation), and (3) a dieback hastening (biotic agents) or stalling mechanism. It is believed that the dieback phenomenon is not restricted to Hawaii but occurs also in other, functionally similar Pacific forests. A corollary to the Hawaiian dieback etiology is a new succession theory, which explains the temporally recurring dieback as a pattern and process sequence in primary succession. The larger dieback patterns are considered a consequence of catastrophic disturbances in the past, such as lava flows, ash blanket deposits, or landslides, which gave rise to large cohorts. Canopy dieback of these large cohorts during their senescing stage then gives rise to new cohorts. However, these become successively smaller and more patchy with each dieback cycle. Concomitantly with the dieback cycles a turnover of successional races or ecotypes appears to occur within the Metrosideros polymorpha species complex. This may allow us to determine pioneer, seral, and near-climax races within this species complex. Sat, 01 Oct 1983 00:00:00 GMT http://hdl.handle.net/10125/729 1983-10-01T00:00:00Z Mueller-Dombois, Dieter