S. Walker et Wg. Lee, Alluvial grasslands in south-eastern New Zealand: vegetation patterns, long-term and post-pastoral change, J RS NZ, 30(1), 2000, pp. 69-103
Grasslands in alluvial systems are studied in 12 valleys and intermontane b
asins in south-eastern New Zealand. Vegetation patterns and their environme
ntal determinants are described overall, and in five different valley group
s ("clusters"). Data from grazing-exclosures are used to examine effects of
the cessation of grazing on vegetation, and long-term plots and management
-induced contrasts at fencelines are used to examine grazing effects. Clima
te (minimum temperature of the coldest month and rainfall) and landscape po
sition are the determinants of the primary vegetation gradient. A secondary
vegetation gradient represents vegetation modification (i.e., the degree o
f exotic species dominance) in alluvial systems. Exclosures indicate post-p
astoral decreases in exotic species dominance, in terms of abundance, in tw
o alluvial sites, and increased exotic dominance in three alluvial sites. I
n terms of species numbers, a decrease in exotic proportion is indicated at
one site, and an increase is indicated at four sites. Differences between
the exclosed and continually grazed vegetation in native species richness,
and in the amount and proportion of native biomass, are negatively related
to the biomass of exotic species present in the ungrazed vegetation, and ar
e not related to the environmental characteristics of the sites. Historic d
ata are insufficient to identify general trends in grazed alluvial grasslan
ds. In three sites for which data are available, the dominance of exotic sp
ecies had increased, and the proportional contributions of all native speci
es, and of native grasses, had decreased with time. In terms of species num
ber, however, proportions of native species had increased at two sites, and
decreased at one site. Widespread and well established exotic grasses appe
ar to have reached their potential environmental range in the alluvial syst
ems studied, but the more recent invader, Hieracium pilosella, may not yet
occupy its potential range. The practicability of post-pastoral conservatio
n of native vegetation in alluvial systems might be indicated by the identi
fication of modification thresholds. However, threshold identification requ
ires data from a greater number of long-term plots than presently exists in
southern alluvial grasslands.