Martin J. Lechowicz

THE STATISTICAL ANALYSIS OF ECOPHYSIOLOGICAL RESPONSE CURVES OBTAINED FROM EXPERIMENTS INVOLVING REPEATED MEASURES

Physiological ecologists often analyze the responses of physiological or bio- chemical traits to environmental factors such as temperature, irradiance, water potential, or the concentrations of CO2, 0, and inorganic nutrients. The data for such a response curve typically are gathered by sequential sampling of the same plant or animal, and their analysis should explicitly allow for this repeated-measures design. Unfortunately, the sta- tistical analysis of response curves in ecology generally has either been ignored or incorrectly done. In an effort to encourage rigorous analysis of response data, we address statistical treatment of response curves and illustrate the correct alternatives that are available. Four different statistical methods for analyzing response curves are considered: analysis of vari- ance with repeated measures (ANOVAR), multivariate analysis of variance with repeated measures (MANOVAR), a nonparametric split-plot analysis (NP split-plot) and parametric comparison of models fitted to the data by nonlinear regression. Analyses of the CO2 dependence of photosynthesis in the C4 grass Echinochloa crus-galli following chilling are used as examples of these different methods. ANOVAR, potentially the most powerful analysis, makes stringent assumptions about the variance-covariance structure of the data. Within limits these assumptions can be relaxed and a corrected significance level used. When the variance-covariance structure badly violates the ANOVAR assumptions, MANOVAR or NP split-plot are viable alter- natives. In physiological ecology, however, the use of MANOVAR frequently is limited by small sample sizes and the tendency for the number of levels of the treatment factor to exceed the sample size. Greater attention to experimental design can avoid this problem. The NP split-plot is based on simple assumptions and could be widely used. The comparison of curves fitted by nonlinear regression is also distribution free and provides an interesting alternative when the responses are amenable to fitting. For any of these analyses to be viable the thoughtful choice of experimental protocols and design is essential.

The Evolution of Plant Ecophysiological Traits: Recent Advances and Future Directions

lants exhibit enormous ecophysiological and functional diversity, which underlies variation in growth rates, productivity, population and community dynamics, and ecosystem function. The broad congruence of these variations with climatic and environmental conditions on local, regional, and global scales has fostered the concept that plant ecophysiological characteristics are well adapted to their local circumstances. For example, the repeated occurrence of plants with CAM (Crassulacean Acid Metabolism) photosynthesis and succulent leaves or stems in severely water-limited environments, and the independent evolution of these traits in numerous plant lineages, provides compelling evidence of the physiological evolution of these water-conserving traits under the influence of natural selection (Ehleringer and Monson 1993). Similarly, studies of the evolution of heavy metal tolerance confirm that natural selection may cause rapid ecophysiological evolution in just a few generations, leading to local adaptation in populations just a few meters apart (Antonovics et al. 1971). Many ecophysiological traits—considered here as all aspects of resource uptake and utilization, including biochemistry, metabolism, gas exchange, leaf structure and function, nutrient and biomass allocation, canopy structure, and growth—are likely to influence fitness and undergo adaptive evolution. Traits affecting the assimilation and use of resources such as carbon, water, and nutrients directly influence plant growth. Patterns of resource allocation to growth, reproduction, defense, and stress tolerance are also likely to be under strong selection. Phenotypic plasticity, the expression of different phenotypes by

How do traits vary across ecological scales? A case for trait-based ecology

Despite the increasing importance of functional traits for the study of plant ecology, we do not know how variation in a given trait changes across ecological scales, which prevents us from assessing potential scale-dependent aspects of trait variation. To address this deficiency, we partitioned the variance in two key functional traits (leaf mass area and leaf dry matter content) across six nested ecological scales (site, plot, species, tree, strata and leaf) in lowland tropical rainforests. In both traits, the plot level shows virtually no variance despite high species turnover among plots and the size of within-species variation (leaf + strata + tree) is comparable with that of species level variation. The lack of variance at the plot level brings substantial support to the idea that trait-based environmental filtering plays a central role in plant community assembly. These results and the finding that the amount of within-species variation is comparable with interspecific variation support a shift of focus from species-based to trait-based ecology.

Why Do Temperate Deciduous Trees Leaf Out at Different Times? Adaptation and Ecology of Forest Communities

With the onset of spring in the deciduous forests of eastern North America, tree leaves do not all emerge in perfect synchrony. Geographic variation in the timing of bud break within species (Lamb 1915; Altman and Dittmer 1962, table 104) can reasonably be attributed to latitudinal climatic trends (Schnelle 1955; Lieth 1974). It is less clear what causes phonological differences among tree species in a particular region (Smith 1915; Ahlgren 1957). Even within single forests, leaf emergence in spring varies over several weeks among coexisting native trees (fig. 1), and species produce leaves at quite different rates during the rest of the growing season. For example, species of Popillds (Critchfield 1960) and Betdla (Kozlowski and Clausen 1966) produce a flush of early leaves in the spring followed by a series of individual late leaves through the summer, but species of Fraxinus (Gill 1971) and Carya (Foster 1931) normally produce only a spring flush of foliage unless defoliated. Querciis has a primary flush of spring leaves, which in young trees can be followed by secondary flushes in the summer (Johannestriebe, lammas shoots) even without defoliation (Lavarenne-Allary 1965; Reich et al. 1978; Kriebel et al. 1976). The proximate control of initial leaf emergence in most temperate deciduous trees is usually the cumulative thermal sum (degree-hours, degree-days) to which buds are exposed after a prerequisite cold period (Lyr et al. 1967; Kramer and Kozlowski 1979; Valentine 1983). Occasionally, as in Fagus, which lacks a chilling requirement, bud break is under photoperiodic control (Wareing 1953; Romberger 1963). Although leaf emergence can be advanced a few days by defoliation the preceding year (Heichel and Turner 1976), in general year-to-year variation in the timing of tree phenological stages is low when related to thermal sums rather than calendar days (fig. 2; Lindsey 1963; Taylor 1974). Populations originating from different localities within a species range do differ in their times of bud break when grown in provenance trials. For example, when grown together in Ohio more northern populations of Acer saccharin leafed out earlier than populations originating more to the south (Kriebel 1957); conversely, in Juglans nigra, populations of more southern origin began growth earlier even in plantations near the northern edge of the species range (Bey 1979). While such intraspecific variations

Predicting the Timing of Budburst in Temperate Trees

1. Four models for predicting budburst in northern hardwood trees, based on response to spring warming alone, or with the response to spring warming modified by winter chilling and photoperiod, were compared. An historical, 18-year budburst record, and artificial datasets with budburst dates generated according to each of four conceptual models, were used to analyse the abilities of the models to predict budburst dates. 2. The four models all gave better predictions than could be obtained by taking the average date of budburst of a species. The historical budburst dates were most accurately predicted by models based only on spring warming from a fixed start date, or from a start date determined by the satisfaction of a chilling requirement

The Ecology and Genetics of Fitness in Forest Plants. II. Microspatial Heterogeneity of the Edaphic Environment

Variation in the physical environment was investigated at scales of 0.1-50 m in the understorey of a southern Quebec forest, dominated by Acer saccharum and Fagus grandifolia, to determine if the spatial heterogeneity necessary to account for the evolution and maintenance of genetically diverse populations of forest plants exists in this old-growth forest. To characterize spatial heterogeneity in the understorey during the late summer, soil pH and the availability of K + and NO 3 − ions in the soil solution were measured (...)

INVASIBILITY AND ABIOTIC GRADIENTS: THE POSITIVE CORRELATION BETWEEN NATIVE AND EXOTIC PLANT DIVERSITY

We sampled the understory community in an old-growth, temperate forest to test alternative hypotheses explaining the establishment of exotic plants. We quantified the individual and net importance of distance from areas of human disturbance, native plant diversity, and environmental gradients in determining exotic plant establishment. Distance from disturbed areas, both within and around the reserve, was not correlated to exotic species richness. Numbers of native and exotic species were positively correlated at large (50 m 2 ) and small (10 m 2 ) plot sizes, a trend that persisted when relationships to environ- mental gradients were controlled statistically. Both native and exotic species richness in- creased with soil pH and decreased along a gradient of increasing nitrate availability. Exotic species were restricted to the upper portion of the pH gradient and had individualistic responses to the availability of soil resources. These results are inconsistent with both the diversity-resistance and resource-enrichment hypotheses for invasibility. Environmental conditions favoring native species richness also favor exotic species richness, and com- petitive interactions with the native flora do not appear to limit the entry of additional species into the understory community at this site. It appears that exotic species with niche requirements poorly represented in the regional flora of native species may establish with relatively little resistance or consequence for native species richness.

The spatial structure of the physical environment

There is substantial environmental variance at small spatial scales (1 m or less) in both natural and disturbed environments. We have investigated the spatial structure of physical variables at larger scales (up to 106 m). We analysed surveys of edaphic properties of Wisconsin forest soils, of the water chemistry of lakes in Ontario and Labrador, and of temperature and precipitation in northeastern North America. We found no clear indication that the variance among sites approaches some maximal value as the distance between them increases. We suggest instead that the variance of the physical environment tends to increase continually with distance. The slope of the log-log regression of variance on distance provides a means of comparing the heterogeneity of different environments with respect to a given factor, or of comparing different factors within a given environment. This slope provides a useful measure of environmental structure that can be related to the biodiversity or plasticity of native organisms.

FERN COMMUNITY ASSEMBLY: THE ROLES OF CHANCE AND THE ENVIRONMENT AT LOCAL AND INTERMEDIATE SCALES

We evaluated the roles of the abiotic environment and dispersal in the as- sembly of fern communities at contrasting spatial scales within an old-growth, temperate deciduous forest. Specifically, we examined correlations among the geographic location of sampling plots separated by either 135-3515 m (mesoscale) or 4-134 m (fine scale), the abiotic environmental characteristics of the plots, and their constituent fern species. Ferns had predictable distributions along a soil moisture gradient at both spatial scales: six of eight common fern species showed repeatable environmental optima along the soil moisture gradient. By sampling in such a way as to decouple the correlation between distance and environmental variation, we showed the dominant role of environmental variables such as soil moisture in determining fern distributions at the mesoscale. At the fine scale, however, strong spatial autocorrelation in the abiotic environment precluded assigning any definitive role for either dispersal or environmental determinism alone in affecting fern distributions. The expectations of neutral theory that are rooted in dispersal limitation and those of niche theory that are rooted in environmental adaptation converge at fine spatial scales where natural environments have strong spatial structure. The structure of the environment at fine spatial scales may foster the persistence of dispersal-limited plants in the community; neighboring environments are likely to be similar, and thus suitable for propagules dis- persing short distances. While patterns of fern distribution in this locality are not consistent with purely neutral or random models of species coexistence, alternative models that rely on strict niche requirements without accounting for dispersal effects and the inherent spatial structure of the environment are inadequate because they neglect the important interaction of these factors. This outcome supports the relevance of developing theory that considers the joint effects of environmental determinism and dispersal on the distribution and abun- dance of plant species.

Foliage quality changes during canopy development of some northern hardwood trees

SummaryThe ephemerality of high quality foliage in spring may act as a defense for trees against early season folivores, but only if the duration of high quality is so short that it is difficult for insects to synchronize their eclosion with the period of high quality foliage that follows budbreak. The rate of change in foliage quality on a day to day basis through the spring was determined for 9 species of hardwood trees in 2–3 years. Measurement of physical and chemical parameters and a bioassay with gypsy moth larvae both showed decreasing quality during the three to five weeks of canopy development in all species. Rates of decline differed among species but the patterns were similar from year to year on a degree-day scale. Growth rates of larvae raised through the first stadium on foliage of differing ages reflected these changes in foliage acceptability. Increasing toughness and declining nitrogen and water contents of leaves were correlated with changes in acceptability to larvae but explained only a small part of the variation in acceptability. The host-seeking period of gypsy moth larvae over-lapped with the availability of highly acceptable foliage of the most preferred host species. Less preferred species had more rapid declines in foliage acceptability, and hence narrower overlaps with the host-seeking period, which may provide defense against use by this generalist forest pest.