This thesis investigates landscape disturbance history in Westland since 1350 AD. Specifically, I test the hypothesis that large-magnitude regional episodes of natural disturbance have periodically devastated portions of the landscape and forest, and that these were caused by infrequent earthquakes along the Alpine Fault. Forest stand history reconstruction was used to determine the timing and extent of erosion and sedimentation events that initiated new forest cohorts in a 1412 ha study area in the Karangarua River catchment, south Westland. Over 85 % of the study area was disturbed sufficiently by erosion/sedimentation since 1350 AD to initiate new forest cohorts. During this time four episodes of catchment-wide disturbance impacted the study area, and these took place about 1825 AD ± 5 years (Ruera episode), 1715 AD ± 5 years (Sparkling episode), 1615 AD ± 5 years (McTaggart episode), and 1445 AD ± 15 years (Junction episode). The three most recent episodes disturbed 10 %, 35-40 % and 32-50 % respectively of the study area. The Junction episode disturbed at least 6 % of the study area, but elimination of evidence by more recent disturbances prevented an upper limit being defined. The three earliest episodes correspond to the date-ranges for three Alpine Fault earthquakes from geological data, and are the only episodes of disturbance within each date-range. An earthquake cause is also consistent with features of the disturbance record: large portions of the study area were disturbed, disturbance occurred on all types 'of landforms, and terrace surfaces were abandoned upstream of the Alpine Fault. On this basis erosion/sedimentation induced by Alpine Fault earthquakes has disturbed 14-20 % of the land surface in the study area per century. Storms and other non-seismic erosional processes have disturbed 3-4 % per century. To examine the importance of the Alpine Fault earthquakes to forest disturbance throughout Westland, I collated all available data on conifer stand age structures in the region and identified dates of disturbance events from 55 even-aged cohorts of trees. Three region-wide episodes of forest disturbance since 1350 AD were found in this sample, and these matched the three Alpine Fault earthquake-caused episodes found in the Karangarua. Forest disturbance at these times was widespread across Westland over at least 200 km from Paringa to Hokitika, and originated from both tree fall and erosion processes. This disturbance history can explain the long-observed regional conifer forest pattern in Westland, of a predominance of similar-sized stands of trees and a relative lack of small-sized (young) stands. The many similar-sized stands are a consequence of synchronous forest disturbance and re-establishment accompanying the infrequent Alpine Fault earthquakes, while the dominance of mature stands of trees and relative lack of young small-sized trees in stands is explained by the long lapsed time since the last Alpine Fault earthquake (c. 280 years). I applied the landscape disturbance history information to the existing geological data to reconstruct the paleoseismicity of the Alpine Fault since 1350 AD. Best estimates for the timing of the most recent three rupture events from these data are 1715 AD ± 5 years, 1615 AD ± 5 years and 1445 AD ± 15 years. Earthquake recurrence intervals were variable, ranging from about 100 years to at least 280 years (the lapsed time since the last event). All three events caused forest and geomorphic disturbance over at least a 200 km section of Fault between the Karangarua and Hokitika Rivers, and were probably single rupture events. Suppressions in cross dated tree-ring chronologies in the western South Island suggest that the last rupture occurred in 1717 AD, and extended as a single rupture from Haupiri to Fiordland, a distance along the Fault of 375 km.
This report reviews the literature on regeneration requirements of main canopy tree species in Westland. Forests managed for production purposes have to be harvested in an ecologically sustainable way; to maintain their natural character, harvesting should facilitate regeneration of target species and ensure that their recruitment is in proportion to the extent of extraction. The reasons for species establishing at any point in time are unclear; however, they are probably related to the availability of suitable microsites for establishment, the size of the canopy openings formed by disturbance, and whether or not seeds are available at or around the time of the disturbance. Age structures from throughout Westland show that extensive, similar-aged, post-earthquake cohorts of trees are a feature of the region. This suggests that infrequent, massive earthquakes are the dominant coarse-scale disturbance agent, triggering episodes of major erosion and sedimentation and leaving a strong imprint in the forest structure. In other forests, flooding and catastrophic windthrow are major forms of disturbance. The findings suggest that, in general, large disturbances are required for conifer regeneration. This has implications for any sustained yield management of these forests if conifers are to remain an important component. Any harvesting should recognise the importance for tree establishment of: forest floor microsites, such as fallen logs and tree tip-up mounds; and the variable way in which canopy gaps are formed. Harvesting should maintain the 'patchy' nature of the natural forest—large patches of dense conifers interspersed with more heterogeneous patches of mixed species.
Mixed conifer, beech and hardwood forests are relatively common in Aotearoa/New Zealand, but are not well studied. This thesis investigates the coexistence, regeneration dynamics and disturbance history of a mixed species forest across an environmental gradient of drainage and soil development in north Westland. The aim was to investigate whether conifers, beech and non-beech hardwood species were able to coexist on surfaces that differed in their underlying edaphic conditions, and if so to understand the mechanisms that influenced their regeneration on both poorly drained and well drained soils. The site selected was an area of high tree species diversity on a lowland 0.8 km² post-glacial terrace at the base of Mount Harata in the Grey River Valley. My approach was to use forest stand history reconstruction at two spatial scales: an intensive within-plot study of stand dynamics (chapter 1) and a whole-landform approach (chapter 2) that examined whether the dynamics identified at the smaller within-plot scale reflected larger patterns across the terrace. In chapter 1, three large permanent plots (0.3-0.7 ha) were placed at different points along the drainage gradient, one plot situated in each of the mainly well-drained, poorly drained and very poorly drained areas along the terrace. Information was gathered on species age and size structures, spatial distributions of tree ages, species interactions, microsite establishment preferences, patterns of stand mortality, and disturbance history in each plot. There were differences in stand structure, composition and relative abundance of species found between the well drained plot and the two poorer drained plots. On the well drained site conifers were scarce, the beeches Nothofagus fusca and N. menziesii dominated the canopy, and in the subcanopy the hardwood species Weinmannia racemosa and Quintinia acutifolia were abundant. As drainage became progressively poorer, the conifers Dacrydium cupressinum and Dacrycarpus dacrydioides became more abundant and occupied the emergent tier over a beech canopy. The hardwoods W. racemosa and Q. acutifolia became gradually less abundant in the subcanopy, whereas the hardwood Elaeocarpus hookerianus became more so. In the well drained plot, gap partitioning for light between beeches and hardwoods enabled coexistence in response to a range of different sized openings resulting from disturbances of different extent. In the two more poorly drained plots, species also coexisted by partitioning microsite establishment sites according to drainage. There were several distinct periods where synchronous establishment of different species occurred in different plots, suggesting there were large disturbances: c. 100yrs, 190-200 yrs, 275-300 yrs and 375-425 yrs ago. Generally after the same disturbance, different species regenerated in different plots reflecting the underlying drainage gradient. However, at the same site after different disturbances, different sets of species regenerated, suggesting the type and extent of disturbances and the conditions left behind influenced species regeneration at some times but not others. The regeneration of some species (e.g., N. fusca in the well-drained plot, and Dacrydium in the poorer drained plots) was periodic and appeared to be closely linked to these events. In the intervals between these disturbances, less extensive disturbances resulted in the more frequent N. menziesii and especially hardwood regeneration. The type of tree death caused by different disturbances favoured different species, with dead standing tree death favouring the more shade-tolerant N. menziesii and hardwoods, whereas uprooting created a mosaic of microsite conditions and larger gap sizes that enabled Dacrycarpus, N. fusca and E. hookerianus to maintain themselves in the poorly drained areas. In chapter 2, 10 circular plots (c. 0.12 ha) were placed in well drained areas and 10 circular plots (c. 0.2 ha) in poorly drained plots to collect information on species population structures and microsite preferences. The aims were to reconstruct species' regeneration responses to a range of disturbances of different type and extent across the whole terrace, and to examine whether there were important differences in the effects of these disturbances. At this landform scale, the composition and relative abundances of species across the drainage gradient reflected those found in chapter 1. There were few scattered conifers in well drained areas, despite many potential regeneration opportunities created from a range of different stand destroying and smaller scale disturbances. Three of the four periods identified in chapter 1 reflected distinct terrace-wide periods of regeneration 75-100 yrs, 200-275 yrs and 350-450 yrs ago, providing strong evidence of periodic large, infrequent disturbances that occurred at intervals of 100-200 yrs. These large, infrequent disturbances have had a substantial influence in determining forest history, and have had long term effects on forest structure and successional processes. Different large, infrequent disturbances had different effects across the terrace, with the variability in conditions that resulted enabling different species to regenerate at different times. For example, the regeneration of distinct even-aged Dacrydium cohorts in poorly drained areas was linked to historical Alpine Fault earthquakes, but not to more recent storms. The variation in the intensity of different large, infrequent disturbances at different points along the environmental drainage gradient, was a key factor influencing the scale of impacts. In effect, the underlying edaphic conditions influenced species composition along the drainage gradient and disturbance history regulated the relative abundances of species. The results presented here further emphasise the importance of large scale disturbances as a mechanism that allows coexistence of different tree species in mixed forest, in particular for the conifers Dacrydium, Dacrycarpus and the beech N. fusca, by creating much of the environmental variation to which these species responded. This study adds to our understanding of the effects of historical earthquakes in the relatively complex forests of north Westland, and further illustrates their importance in the Westland forest landscape as the major influential disturbance on forest pattern and history. These results also further develop the 'two-component' model used to describe conifer/angiosperm dynamics, by identifying qualitative differences in the impacts of different large, infrequent disturbances across an environmental gradient that allowed for coexistence of different species. In poorer drained areas, these forests may even be thought of as 'three-component' systems with conifers, beeches and hardwoods exhibiting key differences in their regeneration patterns after disturbances of different type and extent, and in their microsite preferences.
Saltwater Forest is a Dacrydium cupressinum-dominated lowland forest covering 9000 ha in south Westland, South Island, New Zealand. Four thousand hectares is managed for sustainable production of indigenous timber. The aim of this study was to provide an integrated analysis of soils, soil-landform relationships, and soil-vegetation relationships at broad and detailed scales. The broad scale understandings provide a framework in which existing or future studies can be placed and the detailed studies elucidate sources of soil and forest variability. Glacial landforms dominate. They include late Pleistocene lateral, terminal and ablation moraines, and outwash aggradation and degradation terraces. Deposits and landforms from six glacial advances have been recognised ranging from latest Last (Otira) Glaciation to Penultimate (Waimea) Glaciation. The absolute ages of landforms were established by analysis of the thickness and soil stratigraphy of loess coverbeds, augmented with radiocarbon dating and phytolith and pollen analysis. In the prevailing high rainfall of Westland soil formation is rapid. The rate of loess accretion in Saltwater Forest (ca. 30 mm ka⁻¹) has been low enough that soil formation and loess accretion took place contemporaneously. Soils formed in this manner are known as upbuilding soils. The significant difference between upbuilding pedogenesis and pedogenesis in a topdown sense into an existing sediment body is that each subsoil increment of an upbuilding soil has experienced processes of all horizons above. In Saltwater Forest subsoils of upbuilding soils are strongly altered because they have experienced the extremely acid environment of the soil surface at some earlier time. Some soil chronosequence studies in Westland have included upbuilding soils formed in loess as the older members of the sequence. Rates and types of processes inferred from these soils should be reviewed because upbuilding is a different pedogenic pathway to topdown pedogenesis. Landform age and morphology were used as a primary stratification for a study of the soil pattern and nature of soil variability in the 4000 ha production area of Saltwater Forest. The age of landforms (> 14 ka) and rapid soil formation mean that soils are uniformly strongly weathered and leached. Soils include Humic Organic Soils, Perch-gley Podzols, Acid Gley Soils, Allophanic Brown Soils, and Orthic or Pan Podzols. The major influence on the nature of soils is site hydrology which is determined by macroscale features of landforms (slope, relief, drainage density), mesoscale effects related to position on landforms, and microscale influences determined by microtopography and individual tree effects. Much of the soil variability arises at microscales so that it is not possible to map areas of uniform soils at practical map scales. The distribution of soil variability across spatial scales, in relation to the intensity of forest management, dictates that it is most appropriate to map soil complexes with boundaries coinciding with landforms. Disturbance of canopy trees is an important agent in forest dynamics. The frequency of forest disturbance in the production area of Saltwater Forest varies in a systematic way among landforms in accord with changes in abundance of different soils. The frequency of forest turnover is highest on landforms with the greatest abundance of extremely poorly-drained Organic Soils. As the abundance of better-drained soils increases the frequency of forest turnover declines. Changes in turnover frequency are reflected in the mean size and density of canopy trees (Dacrydium cupressinum) among landforms. Terrace and ablation moraine landforms with the greatest abundance of extremely poorly-drained soils have on average the smallest trees growing most densely. The steep lateral moraines, characterised by well drained soils, have fewer, larger trees. The changes manifested at the landform scale are an integration of processes operating over much shorter range as a result of short-range soil variability. The systematic changes in forest structure and turnover frequency among landforms and soils have important implications for sustainable forest management.
Six stands located on different land forms in mixed old-growth Nothofagus forests in the Matiri Valley (northwest of South Island, New Zealand) were sampled to examine the effects of two recent large earthquakes on tree establishment and tree-ring growth, and how these varied across land forms. 50 trees were cored in each stand to determine age structure and the cores were cross-dated to precisely date unusual periods of radial growth. The 1968 earthquake (M = 7.1, epicentre 35 km from the study area) had no discernible impact on the sampled stands. The impact of the 1929 earthquake (M = 7.7, epicentre 20 km from the study area) varied between stands, depending on whether or not they had been damaged by soil or rock movement. In all stands, the age structures showed a pulse of N. fusca establishment following the 1929 earthquake, with this species dominating establishment in large gaps created by landslides. Smaller gaps, created by branch or tree death, were closed by both N. fusca and N. menziesii. The long period of releases (1929-1945) indicates that direct earthquake damage was not the only cause of tree death, and that many trees died subsequently most likely of pathogen attack or a drought in the early 1930s. The impacts of the 1929 earthquake are compared to a storm in 1905 and a drought in 1974-1978 which also affected forests in the region. Our results confirm that earthquakes are an important factor driving forest dynamics in this tectonically active region, and that the diversity of earthquake impacts is a major source of heterogeneity in forest structure and regeneration.
Question: Does canopy tree regeneration response to different large disturbances vary with soil drainage? Location: Old-growth conifer (Dacrydium and Dacrycarpus), angiosperm (Nothofagus and Weinmannia) rain forest, Mount Harata, South Island, New Zealand. Methods: Trees were aged (1056 cores) to reconstruct stand history in 20 (0.12 - 0.2 ha) plots with different underlying drainage. Spatial analyses of an additional 805 tree ages collected from two (0.3 - 0.7 ha) plots were conducted to detect patchiness for five canopy tree species. Microsite preferences for trees and saplings were determined. Results: There were clear differences in species regeneration patterns on soils with different drainage. Conifer recruitment occurred infrequently in even-aged patches (> 1000 m²) and only on poorly drained soils. Periodic Nothofagus fusca and N. menziesii recruitment occurred more frequently in different sized canopy openings on all soils. Weinmannia recruitment was more continuous on all soils reflecting their greater relative shade-tolerance. Distinct periods of recruitment that occurred in the last 400 years matched known large disturbances in the region. These events affected species differently as soil drainage varied. Following earthquakes, both conifers and N. menziesii regenerated on poorly drained soils, while Nothofagus species and Weinmannia regenerated on well-drained soils. However, Dacrydium failed to regenerate after patchy storm damage in the wetter forest interior; instead faster-growing N. fusca captured elevated microsites caused by uprooting. Conclusions: Underlying drainage influenced species composition, while variation in the impacts of large disturbance regulated relative species abundances on different soils.
Tree mortality is a fundamental process governing forest dynamics, but understanding tree mortality patterns is challenging because large, long-term datasets are required. Describing size-specific mortality patterns can be especially difficult, due to few trees in larger size classes. We used permanent plot data from Nothofagus solandri var. cliffortioides (mountain beech) forest on the eastern slopes of the Southern Alps, New Zealand, where the fates of trees on 250 plots of 0.04 ha were followed, to examine: (1) patterns of size-specific mortality over three consecutive periods spanning 30 years, each characterised by different disturbance, and (2) the strength and direction of neighbourhood crowding effects on sizespecific mortality rates. We found that the size-specific mortality function was U-shaped over the 30-year period as well as within two shorter periods characterised by small-scale pinhole beetle and windthrow disturbance. During a third period, characterised by earthquake disturbance, tree mortality was less size dependent. Small trees (,20 cm in diameter) were more likely to die, in all three periods, if surrounded by a high basal area of larger neighbours, suggesting that sizeasymmetric competition for light was a major cause of mortality. In contrast, large trees ($20 cm in diameter) were more likely to die in the first period if they had few neighbours, indicating that positive crowding effects were sometimes important for survival of large trees. Overall our results suggest that temporal variability in size-specific mortality patterns, and positive interactions between large trees, may sometimes need to be incorporated into models of forest dynamics.
Knowledge of past climate variability is essential for understanding present and future climate trends. This study used Halocarpus biformis (pink pine) ring-width chronologies to investigate palaeotemperature history in Westland, New Zealand. The ensuing reconstruction is among the longest palaeoseries produced for New Zealand to date. It is in good agreement with other tree-ring-based records, and with instrumental (both local and hemispheric) data. Thirteen pink pine chronologies were developed. Ring-width measurements were detrended using the Regional Curve Standardisation method to retain as much low-frequency variance as possible. Crossdating revealed the existence of a strong common signal among trees. Inter-site comparison indicated that a common control mechanism affected tree growth not only within sites, but also across sites. To determine whether climate was the main factor that controlled the growth of pink pine in Westland, correlation and response function analyses were employed. Temperature, precipitation and the Southern Oscillation Index were tested for their relationship with tree growth. Mean monthly temperature was identified as the primary growth-limiting factor. Chronologies were positively correlated with temperature over an extended period (5-17 months), and climate response modelling showed that temperature explained 11-60% variance in the tree-ring data. The highest and most stable correlations occurred between tree growth and summer (January-March) temperatures. Tree-ring data from the six sites that contained the strongest temperature signal were combined, and the Westland Regional Chronology (WRC) was developed. The WRC was then used to reconstruct January-March temperatures back to A.D. 1480. The calibration model explained 43% of the variance in temperature, and all calibration and verification tests were passed at high levels of significance. The reconstruction showed that temperatures in Westland have been following a positive trend over the last 520 years. The coolest 25-year period was 1542-1566, while temperatures reached their maximum in 1966-1990. Spectral analysis of the Westland palaeotemperature record revealed cycles at periods of about 3, 5-6, 11, 14, 22, 45 and 125 years. This study also confirmed that climate response is species-dependent. A separate exercise, which compared two species from the same site, demonstrated that while pink pine's growth was mainly influenced by summer temperatures, Libocedrus bidwillii was affected by conditions at the beginning of the growing season. However, the temperature signal in Westland's Libocedrus bidwillii was insufficient to produce a reliable reconstruction. It might be because the climate signal in this species was obscured by disturbances, as was shown in the final section of this project. Frequent growth releases and suppressions implied that Libocedrus bidwillii integrated both major (Alpine Fault earthquakes) and minor (windthrow) disturbances in its ring widths. Pink pine, on the other hand, was not sensitive to disturbance, and was therefore a better indicator of palaeotemperatures in Westland. This research has strengthened the New Zealand network of chronology sites, and confirmed that pink pine has great dendroclimatic value. The last 520 years of temperature fluctuations were reconstructed with a high degree of fidelity - the model developed in this thesis is currently the most accurate estimate of a temperature-growth relationship in the country.
Liquefaction affects late Holocene, loose packed and water saturated sediment subjected to cyclical shear stress. Liquefaction features in the geological record are important off-fault markers that inform about the occurrence of moderate to large earthquakes (> 5 Mw). The study of contemporary liquefaction features provides a better understanding of where to find past (paleo) liquefaction features, which, if identified and dated, can provide information on the occurrence, magnitude and timing of past earthquakes. This is particularly important in areas with blind active faults. The extensive liquefaction caused by the 2010-2011 Canterbury Earthquake Sequence (CES) gave the geoscience community the opportunity to study the liquefaction process in different settings (alluvial, coastal and estuarine), investigating different aspects (e.g. geospatial correlation with landforms, thresholds for peak ground acceleration, resilience of infrastructures), and to collect a wealth geospatial dataset in the broad region of the Canterbury Plains. The research presented in this dissertation examines the sedimentary architecture of two environments, the alluvial and coastal settings, affected by liquefaction during the CES. The novel aim of this study is to investigate how landform and subsurface sedimentary architecture influence liquefaction and its surface manifestation, to provide knowledge for locating studies of paleoliquefaction in future. Two study cases documented in the alluvial setting showed that liquefaction features affected a crevasse splay and point bar ridges. However, the liquefaction source layer was linked to paleochannel floor deposits below the crevasse splay in the first case, and to the point bar deposits themselves in the second case. This research documents liquefaction features in the coastal dune system of the Canterbury Plains in detail for the first time. In the coastal dune setting the liquefiable layer is near the surface. The pore water pressure is vented easily because the coastal dune soil profile is entirely composed of non-cohesive, very well sorted sandy sediment that weakly resists disturbance from fluidised sediment under pressure. As a consequence, the liquefied flow does not need to find a specific crack through which the sediment is vented at the surface; instead, the liquefied sand finds many closely spaced conduits to vent its excess of pore water pressure. Therefore, in the coastal dune setting it is rare to observe discrete dikes (as they are defined in the alluvial setting), instead A horizon delamination (splitting) and blistering (near surface sills) are more common. The differences in styles of surface venting lead to contrasts in patterns of ejecta in the two environments. Whereas the alluvial environment is characterised by coalesced sand blows forming lineations, the coastal dune environment hosts apparently randomly distributed isolated sand blows often associated with collapse features. Amongst the techniques tested for the first time to investigate liquefaction features are: 3D GPR, which improved the accuracy of the trenching even six years after the liquefaction events; thin section analysis to investigate sediment fabric, which helped to discriminate liquefied sediment from its host sediment, and modern from paleoliquefaction features; a Random Forest classification based on the CES liquefaction map, which was used to test relationships between surface manifestation of liquefaction and topographic parameters. The results from this research will be used to target new study sites for future paleoliquefaction research and thus will improve the earthquake hazard assessment across New Zealand.
