Henrietta Maria and East Pen Island respec- tively. In these papers, the dominant plant association was a "general lichen-heath" com- posed of Cetraria nivalis, ...
Can. J. Bot. Downloaded from www.nrcresearchpress.com by Shanghai International Studies University on 06/05/13 For personal use only.
Studies on lichen-dominated systems. VII. Interaction of the general lichen-heath with edaphic factors DOUGLAS W. LARSON AND KENNETH A. KERSHAW Deparfmenf of Biology, McMaster Ut~iversiry,Hamilfon, Onfario Received November 5, 1973 LARSON,D. W., and K. A. KERSHAW. 1974. Studies on lichen-dominated systems. VII. Interaction of the general lichen-heath with edaphic factors. Can. J. Bot. 52: 1163-1176. The patterns of variation in vegetation and the more apparent environmental parameters within and between the ridges of a recently formed Hudson Bay coastal raised-beach system 1.5 km wide are presented and analyzed using principle-component analysis and correlation coefficients. Previous research in the area did not detect such patterns within the general lichen-heath. There is a marked develo mental sequence from the coast inland, involving increases in the soil moisture content an8t1le peat-layer thickness, but variations inde endent of any age sequence occur on an intraridge basis. Relationships are drawn between tRese variations and the biomass or cover of nine important species. LARSON,D. W., et K . A. KERSHAW. 1974. Studies on lichen-dominated systems. VII. Interaction of the general lichen-heath with edaphic factors. Can. J. Bot. 52: 1163-1176. Les auteurs presentent les patrons de variation de la vegitation et des paramktres les plus apparents du milieu, B l'intirieur et entre les cr&tes d'un systkme recent, large de 1.5 km, de plages surClevCes le long de la cbte de la baie d'Hudson. Ces patrons sont analyses a l'aide d e l'analyse des composantes principales et de coefficient de corrilation. Des recherches antkrieures dans cette rigion n'avaient pu dimontrer l'existence de tels patrons 2 I'intirieur du "general lichen-heath." I1 existe une chronosCquence marquie i partir de la cbte vers l'intirieur, comportant une augmentation de I'humidite du sol et de l'ipaisseur de la couche de tourbe. Cependant, o n rencontre 2i l'intkrieur des cr&tes des variations indipendantes de toute chronosequence. Les auteurs Ctablisseilt des corrilations entre ces variations et la biomasse ou la couverture de neuf espkces importantes. [Traduit par le journal] I
~
I
Introduction The general importance of lichen as a vegetational cover in northern regions has been established (Ahti and Hepburn 1967; Kershaw and Rouse 1971; Rouse and Kershaw 1971; Andreev 1954) but detailed studies of the structure of lichen-dominated systems have been only recently initiated. Neal and Kershaw (19730, b) and Kershaw and Rouse (1973) have studied the coastal raised-beach tundra of Hudson Bay at Cape Henrietta Maria and East Pen Island respectively. In these papers, the dominant plant association was a "general lichen-heath" composed of Cetraria nivalis, C. islandica, C. cuculIata, Cladina mitis, C. rangiferina, Alectoria ochrole~rca, and Rlzododendron lapponicum in varying abundances. The most recent work (Kershaw and Rouse 1973) separated two forms of the general lichenheath on the basis of absence of C. mitis and C. rarzgiffrina in one set of samples nearer to the coast. It was thought that the general lichen-
heath I1 was associated with the older ridges having thicker accumulations of peat. Visually apparent interridge and intraridge patterns of many species (e.g. Alectoria oclzroleuca and Cetraria rzivalis) were not detectable, however, with the methods used. This present study was designed to examine in detail the structure of the "general lichen-heath" within a series of recently exposed ridges (250800 years old) and to determine any relationships between the vegetation and the more apparent environmental variables. Tlze Researclz Site The research was conducted at the East Pen Island Research Site, details of which are given in the previous two papers in this series (Kershaw and Rouse 1973; Rouse and Kershaw 1973). Generally, the climate of the site is similar to that of Churchill with an average annual temperature of -6°C and an annual rainfall of 40 cm; prevailing winds are north-northwest. While the main research camp is situated on the node of the most recent raised-beach system
Can. J. Bot. Downloaded from www.nrcresearchpress.com by Shanghai International Studies University on 06/05/13 For personal use only.
1164
CAN. I .
BOT.VOL. 52.
(see Fig. l), the individual beach ridges used in this study were about 8 km southeast along the coast, in an area where the ridges are fanning out (Fig. 2). Ridges 3-7 in the general lichenheath of Kershaw and Rouse (1973) were in this same general area and are included in ridges 1-12 in this present work. Soils are fine to coarse sands mixed with fine gravels. A thin peat layer is usually present and continuous permafrost is found at depths varying from 0.5 to 2.0 m during the summer. Methods The five ridges under study (Fig. 3) were chosen on the basis of accessibility, adequate distance between neighboring ridges, and absence of gross topographic irregularities due to permafrost and wind erosion features. Since the ridges are laid down from the node outwards, it was also necessary t o obtain samples in a line perpendicular to the shore. This prevented any compounding of the inland vegetational change with that running along any one ridge t o its most recently formed secrion. T o allow for interridge and intraridge con~parisons,it was necessary t o develop a flexible sampling design. Each ridge was sampled by six 33-m transecrs oriented parallel to the ridge and spaced equally over the ridge profile. Six replicate 1 - m v u a d r a t s were then randomly placed within each transect and subdivided into nine equal subquadrats. Results were averaged over subquadrats t o give square95"
90"
85'
80"
HUDSON BAY
Lowlands
----
80"
85'
FIG. 1. The Hudson Bay region, showing the location of East Pen Island.
1974
meter mean values and were averaged over 1-n12quadrats t o give transect mean values. The ridge system is represented as 30 transect positions on five ridges (cf. Fig. 3). The Domin scale previously used by Neal and Kershaw (1973a, b) and Kershaw and Rouse (1973) is too insensitive t o effectively demonstrate the actual variations in abundance levels found in this vegetation. Accordingly, percentage cover was chosen as the measure. Cover has met with wide success in pl~ytosociology and is easily applied t o arctic vegetation, which characteristically has a low or prostrate habit. Of all the species encountered in the sample, nine of the more abundant were found t o have marked patterns: Cetr.ario islc111dica,C . rrivalis, C. cllcrlllc~ta,Corrriclllcir.ia cliver.gc11rs, Alectorin oclrr.ole~rcn, Clnclilirln tnitis, Dlyas ir~te~
