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 size-specific 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 size-asymmetric 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.

Question: What are the composition, structure and extent of contemporary, common woody vegetation communities in New Zealand? How do the woody plant communities we describe, based on representative sampling, compare to those of previous New Zealand classifications?Methods: We used cluster analysis to classify data from 1177 systematically located vegetation plots, calculated spatial extent and ecological statistics for each alliance defined, and combined forest alliances into groups to assess correspondence with earlier mapped classifications. We used plot-based data on vegetation structure to infer potential community changes.Results: Twenty-four alliances were recognized, each of 19 to 105 plots, with estimated extents of 144 000-794 000 ha. Stand structure data suggest that 15 alliances are largely stable whereas nine may undergo compositional change. Among alliances, the proportion of exotic species ranged from 0 to 50%. Our forest and shrubland classification is another stage in progressively revising such classifications (i.e. forest class maps, vegetative cover map, ECOSAT woody classification and land cover database) produced over the last five decades.

We use seed count data from a New Zealand mono-specific mountain beech forest to test for decadal trends in seed production along an elevation gradient in relation to changes in climate. Seedfall was collected (1965 to 2009) from seed trays located on transect lines at fixed elevations along an elevation gradient (1020 to 1370 m). We counted the number of seeds in the catch of each tray, for each year, and determined the number of viable seeds. Climate variables were obtained from a nearby (<2 km) climate station (914-m elevation). Variables were the sum or mean of daily measurements, using periods within each year known to correlate with subsequent interannual variation in seed production. To determine trends in mean seed production, at each elevation, and climate variables, we used generalized least squares (GLS) regression. We demonstrate a trend of increasing total and viable seed production, particularly at higher elevations, which emerged from marked interannual variation. Significant changes in four seasonal climate variables had GLS regression coefficients consistent with predictions of increased seed production. These variables subsumed the effect of year in GLS regressions with a greater influence on seed production with increasing elevation. Regression models enforce a view that the sequence of climate variables was additive in their influence on seed production throughout a reproductive cycle spanning more than 2 years and including three summers. Models with the most support always included summer precipitation as the earliest variable in the sequence followed by summer maximum daily temperatures. We interpret this as reflecting precipitation driven increases in soil nutrient availability enhancing seed production at higher elevations rather than the direct effects of climate, stand development or rising atmospheric CO2 partial pressures. Greater sensitivity of tree seeding at higher elevations to changes in climate reveals how ecosystem responses to climate change will be spatially variable.

Plant functional traits capture important variation in plant strategy and function. Recent literature has revealed that within-species variation in traits is greater than previously supposed. However, we still have a poor understanding of how intraspecific variation is coordinated among different traits, and how it is driven by environment. We quantified intraspecific variation in wood density and five leaf traits underpinning the leaf economics spectrum (leaf dry matter content, leaf mass per unit area, size, thickness and density) within and among four widespread Nothofagus tree species in southern New Zealand. We tested whether intraspecific relationships between wood density and leaf traits followed widely reported interspecific relationships, and whether variation in these traits was coordinated through shared responses to environmental factors. Sample sites varied widely in environmental variables, including soil fertility (25–900 mg kg–1 total P), precipitation (668–4875 mm yr–1), temperature (5.2–12.4 °C mean annual temperature) and latitude (41–46 °S). Leaf traits were strongly correlated with one another within species, but not with wood density. There was some evidence for a positive relationship between wood density and leaf tissue density and dry matter content, but no evidence that leaf mass or leaf size were correlated with wood density; this highlights that leaf mass per unit area cannot be used as a surrogate for component leaf traits such as tissue density. Trait variation was predicted by environmental factors, but not consistently among different traits; e.g., only leaf thickness and leaf density responded to the same environmental cues as wood density. We conclude that although intraspecific variation in wood density and leaf traits is strongly driven by environmental factors, these responses are not strongly coordinated among functional traits even across co-occurring, closely-related plant species.