Uta Matthes

Light Attenuation by Limestone Rock and Its Constraint on the Depth Distribution of Endolithic Algae and Cyanobacteria

To understand how light attenuation in limestone rock constrains the distribution and abundance of endolithic photoautotrophic organisms, we examined light‐level profiles in 25 rock samples containing endolithic algae and cyanobacteria. We collected samples from three representative cliff sites along the Niagara Escarpment, Canada, using methods that allowed us to make comparisons among as well as within sites. Photosynthetically active radiation profiles were generated by manufacturing cavities in the bottom of thick slabs of surface rock and enclosing radiation sensors in the cavities. The thickness of the flat layer of limestone between the sensor and a light source was then abraded in steps of ca. 0.5 mm. An exponential curve was fitted to the measurements for each sample to allow statistical comparisons of the light levels at different depths. We examined the correlations between physical characteristics of the rock and light attenuation and between light attenuation and the maximum depth, relative biomass, and taxonomic richness of endolithic algae and cyanobacteria. The depth at which 0.01% of available light remained varied from 2.1 to 4.5 mm at different sites, but there was large within‐site variability on intermediate and small spatial scales. In contrast, the maximum depth of photoautotrophic endoliths was homogeneous within sites but different among sites (1.1–3.5 mm). Although the endoliths all have access to light, there was no significant correlation between the limit of their depth penetration and threshold quantum flux densities, indicating that the mechanisms controlling distribution and abundance of endolithic photoautotrophs are as complicated as those found for aquatic algae and cyanobacteria.

Community structure of epilithic lichens on the cliffs of the Niagara Escarpment, Ontario, Canada

The spatial variation in epilithic lichen community structure was investigated as part of a larger study of the vegetation and ecology of the tall limestone cliffs within the Niagara Escarpment Biosphere Reserve in southern Ontario, Canada. The cover of all lichen taxa was visually estimated for a total of 199 samples taken from the top, middle, or bottom of the cliff face at five sites. Twelve environmental variables were also measured. Twenty-seven lichen taxa were identified on the samples. Multivariate ordinations of species composition (DCA, CCA, PCCA) revealed variation in community structure on multiple scales, but no groupings of sites that would have suggested the presence of several distinct species assemblages. A gradient in species composition from north to south, most clearly reflected in the decreasing cover of foliose and fruticose species, may reflect a gradient in human disturbance. There was also intermediate-scale patchiness in species composition in a horizontal plane across cliffs, but despite earlier claims made in the literature, no evidence of vertical zonation of the lichens on cliffs was found. Species composition also responded to small-scale factors possibly related to exposure, light, or moisture. Unlike community composition, the total cover of all lichens was homogeneous over large spatial scales and varied only on a small scale, illustrating that scale as well as resolution of a study may influence the ecological patterns seen. More than half of the species found on the Niagara Escarpment are rare on rock substrates elsewhere in southern Ontario, and two are new for North America (Candelariella heidelbergensis (Nyl.) Poelt and Lecanora perpruinosa Fröberg). The result that cliffs support a distinct flora containing many rare species suggests that they are a reservoir for biodiversity not just for vascular plants, but also for lichens.

THE FORMATION AND POSSIBLE ECOLOGICAL FUNCTION OF STEM STRIPS IN THUJA OCCIDENTALIS

The formation of stem strips through partial cambial dieback occurs in a number of long‐lived woody species and appears to be associated with old age, slow growth, and habitat adversity. The reason for this association is unknown, and investigations of the cause of stem stripping have been impeded by the strong correlation among the associated factors themselves. Hydraulic sectoring of the xylem is thought to be necessary for stem stripping to occur, and we hypothesize that loss of hydraulic conductance by permanent cavitation of tracheids in parts of the xylem is the first step in stem strip formation. The aim of this study was to identify characteristics of Thuja occidentalis that are directly associated with the presence of stem strips. We used a large data set of 842 T. occidentalis collected randomly from a favorable (cliff‐top plateau) and an unfavorable (cliff face) habitat to determine whether age, growth rate, and habitat were directly or indirectly related to the presence of stem strips. To determine whether the diameter of intertracheid pits, which directly controls cavitation resistance, is a possible anatomical mechanism for achieving hydraulic sectoring, we investigated whether pit size showed a pattern that paralleled the occurrence of stem stripping. The results showed that the frequency of stem strips was significantly correlated with tree age, while growth rate and habitat adversity were only indirectly correlated with stem stripping through their correlations with age. Earlywood tracheids of old trees had the largest pits, indicating that these trees are more susceptible to cavitation‐related water deficits than young trees and supporting the hypothesis that cavitation of tracheids is involved in stem strip formation. We discuss possible reasons for the correlation between age and stem stripping and present a graphical conceptual model that may explain the ecological function of stem stripping in trees growing in heterogeneous habitat conditions.

Can Stem Strips Be Induced? An Experimental Investigation of Cliff‐Face Thuja occidentalis

Previous studies have suggested that strip‐bark growth in ancient Thuja occidentalis on cliffs in southern Ontario, Canada, may be initiated by root mortality following rockfall. We conducted a long‐term manipulative field study on 65 cliff‐face trees to test whether stem stripping could be experimentally induced by either severing or excavating part of the root system. Further evidence of the proposed mechanism was sought by monitoring the responses of individual branches on timescales ranging from hours to years, including gas exchange, growth, canopy condition, and mortality. The trees were harvested after 5–7 yr to identify and date any stem strips. The results showed that partial root excavation, but not severing, significantly increased the frequency of stem stripping. In two cut and four excavated trees, the treatments produced partial branch and cambial mortality, supporting the proposed mechanism in the long term. However, the majority of trees did not stem strip, despite major root loss, and shorter‐term responses either were not affected at all (growth) or they affected all branches equally throughout the crown (gas exchange, canopy condition). Three control trees formed stem strips during the study period as well, leading us to conclude that experimentally imposed root loss slightly increased the likelihood of cambial mortality but was neither required for nor the only mechanism leading to stem stripping.