Boreal coniferous forest vegetation in the Solhomfjell area, S Norway: structure, dynamics and change, with particular reference to effects of long distance airborne pollution.

Seventeen original papers that are parts of, or spin-offs from, the TVLF project “Effects of long distance airborne pollution”, are summarized. All papers include data from the Solhomfjell reference area, a S Norwegian old-growth boreal coniferous forest with high deposition of long distance airborne pollution relative to other parts of Norway. Vegetation and environmental structure, population and vegetation dynamics, and vegetation change, was integrated by extensive use of the same sample sets. The nested sampling included 100 extended macro sample plots (64 m²) for tree species and 200 meso sample plots for the understory (1 m²; with 33 environmental variables measured and presence/absence of all species recorded in 16 subplots ). Species abundances were recorded in 50 meso plots all years 1988-93, while all 200 plots were analyzed in 1988 and 1993. Other sampling designs were used for three special studies in the Solhomfjell area. Population biology of Hylocomium splendens was studied in six of NIJOS’ ten areas for vegetational and environmental monitoring of Norwegian boreal spruce forests, as well as in the Solhomfjell area.

Gradient relationships of vegetation at the I m² scale in the Solhomfjell area are summarized by DCA ordination. Three coenoclines are identified: (I) a broad-scale gradient from xeric, lichen-dominated pine forest via subxeric pine forest dominated by ericaceous species and poor, Vaccinium myrtillus-dominated spruce forest to richer, herb-rich spruce forest, (2) a fine-scale gradient in degree of paludification ( e.g., dominance by Sphagnum spp.), most strongly visible in the bottom layer, and (3) a micro-scale gradient from dominance by small hepatics and mosses to dominance by larger mosses of the forest floor. Associated complex-gradients were identified or hypothesized by use of measured environmental variables: (1) danger of soil moisture deficiency and soil nutrient content, (2) median soil moisture, and (3) several aspects of microtopography. Several kinds of independent evidence indicates that physiological tolerance is more important than interspecific competition in the understory. Temporal variation in soil moisture and soil pH is studied, with reference to the main gradients. Persistence of cryptogams as well as vascular plant species was species-specific and size-dependent, with little variation between years. The rate of vegetation dynamics varied predictably along the main coenocline, in relation to site productivity and environmental harshness. The magnitudes of one-year change in species abundances and vegetation were used to evaluate the significance of change in the five-year period. Significant humus acidification occurred in the five-year period, most strongly in richer spruce forest where several vascular plant species declined and vegetation changed in direction of poor spruce forest. N fertilization may explain the increase of Deschampsia flexuosa in spruce forest. Significant enrichment of pine forest humus is demonstrated. Norway spruce and Scots pine showed inconsistent patterns of variation in relative crown density. Stem number and stand volume in permanent plots increased for all tree species.

Modular growth, branching patterns, and hence demography of the clonal moss Hylocomium splendens, was strongly size-dependent. Branching increased and risk of termination decreased with increasing segment size. Strong apical dominance was found. A combination of positive density-dependence of mean segment size, and regulation of segment numbers at very high densities was demonstrated. Mean segment size was strongly related to length of the photosynthetically active period. Increase of most bryophytes in the Solhomfjell area 1988-93 was primarily due to several mild, rainy winters during the period.

Advantages of an integrated approach to monitoring of boreal forests (including trees, understory and environment; the population, species and vegetation levels of organization; several spatial and temporal scales; and univariate as well as multivariate statistical methods) are emphasized. Data from other regions and longer time periods are needed to assess the extent of vegetation response to environmental change.