Sampling event

Solhomfjell forest floor species composition 1988-2018

Latest version published by University of Oslo on 31 January 2023 University of Oslo
Publication date:
31 January 2023
Published by:
University of Oslo
CC-BY 4.0

Download the latest version of this resource data as a Darwin Core Archive (DwC-A) or the resource metadata as EML or RTF:

Data as a DwC-A file download 700 records in English (10 MB) - Update frequency: unknown
Metadata as an EML file download in English (21 KB)
Metadata as an RTF file download in English (23 KB)


Økland, R.H. & Eilertsen, O. 1993. Vegetation-environment relationships of boreal coniferous forests in the Solhomfjell area, Gjerstad, S Norway. - Sommerfeltia 16: 1-254. Oslo. ISBN 82-7420-018-7. ISSN 0800-6865. The understory vegetation (vascular plants, bryophytes and lichens) in an area dominated by boreal coniferous forests is subjected to detailed ecological analysis. Two hundred meso sample plots (1 m2) are used as basis for vegetation sampling, and provided with measurements of 33 environmental variables. Species abundance is recorded as frequency in 16 subplots. Parallel DCA and 2-dimensional LNMDS ordinations of meso sample plots were largely identical, both provided two coenocline axes interpretable in ecological terms. The first axis is interpreted as the response to a broad-scale topographical complex-gradient, made up of two independent complex-gradients; (1) a topography-soil depth complex-gradient in the pine forest (running from lichen-rich pine forests to submesic Vaccinium myrtillus-dominated spruce forests), and (2) a complex-gradient in soil nutrient status in the spruce forest. The second axis, mainly affecting the species composition of the bottom layer, is interpreted as a fine-scale paludification gradient. The causes of variation along these gradients are discussed: Desiccation tolerance is considered to act directly on the physiology of vascular plant species, setting their limits towards xeric sites. Similarly, cryptogams with optima in the more mesic sites are considered to be excluded from drier sites by physiological tolerance. Limits of cryptogams towards more mesic sites are, however, considered to be set by competitive ability (growth rates) in accordance with the competitive hierarchy theory. N availability is assumed to be the most important factor for differentiation of vascular plants along the nutrient gradient, while bryophytes are expected to respond to a complex of factors, including structural properties of the humus layer. Increasing N accumulation in the humus towards xeric sites may indicate oversaturation due to deposition of airborne NO3- or NH4+. Fine-scale paludification, mainly of a soligenous type, occurred in sloping terrain with shallow soil. The cryptogams apparently make up a competitive hierarchy also along the paludification gradient. No other coenoclines could be identified by analysis of 0.0625 m2 micro sample plots, most probably because the response of vegetation to micro-scale environmental gradients (probably most important: the variation in microtopography) not essentially different from the meso-scale gradients, and because the importance of random processes increase towards finer scales. Structuring processes are discussed with reference to the observed patterns. The lack of a closed bottom layer in almost all sample plots is considered a strong indication of high importance of fine-scale disturbance and density-independent mortality in the investigated system, while interspecific competition is of lower importance. The methodology in vegetation ecological studies is discussed with particular reference to monitoring. The potential of an integrated concept using permanent plots, parallel investigation of vegetation and environmental parameters, and gradient analysis, is stressed. Several suggestions for future studies, based on this integrated approach, are made.

Data Records

The data in this sampling event resource has been published as a Darwin Core Archive (DwC-A), which is a standardized format for sharing biodiversity data as a set of one or more data tables. The core data table contains 700 records.

1 extension data tables also exist. An extension record supplies extra information about a core record. The number of records in each extension data table is illustrated below.

Event (core)

This IPT archives the data and thus serves as the data repository. The data and resource metadata are available for download in the downloads section. The versions table lists other versions of the resource that have been made publicly available and allows tracking changes made to the resource over time.


The table below shows only published versions of the resource that are publicly accessible.

How to cite

Researchers should cite this work as follows:

Halvorsen R (2023): Solhomfjell forest floor species composition 1988-2018. v1.4. University of Oslo. Dataset/Samplingevent.


Researchers should respect the following rights statement:

The publisher and rights holder of this work is University of Oslo. This work is licensed under a Creative Commons Attribution (CC-BY 4.0) License.

GBIF Registration

This resource has been registered with GBIF, and assigned the following GBIF UUID: b4934203-2d86-44a6-9fec-b55d7ab97749.  University of Oslo publishes this resource, and is itself registered in GBIF as a data publisher endorsed by GBIF Norway.


Sampling event; B;oreal coniferous forests; DCA; Environmental factors; Gradient; LNMDS; Monitoring; Norway; Ordination; Vegetation;


Rune Halvorsen
  • Metadata Provider
  • Originator
  • User
  • Point Of Contact
Natural History Museum, University of Oslo

Geographic Coverage

The investigation area is situated within the Solhomfjell forest reserve, Gjerstad municipality, Aust-Agder county, S Norway. The distance to the outer coastal line is ca. 38 km. The area of the reserve is 10.25 km², of which the investigation area comprises about 2 km² in the altitudinal interval 350-480 m. The UTM grid reference is 32V ML 86-92, 33-36, and the geographic position is 8°58´E, 58°58´N.

Bounding Coordinates South West [-90, -180], North East [90, 180]

Temporal Coverage

Start Date / End Date 1988-01-01 / 2018-01-01

Sampling Methods

Sampling design and recording of species A combination of systematic and restricted random sampling techniques was used for placement of sample plots. Eight transects were selected subjectively to cover the variation in boreal forest vegetation in the investigation area, as well as the variation in topography, slope, aspect, etc. Most transects were running from hilltop to valley bottom, but level transects were also included. Each tenth meter along the transects was a potential site for the lower left corner of a macro sample plot, 16 m², with its left margin along the transect line. Positions were rejected if they included (1) mires, tarns or elements of ecosystems other than forest, (2) more than 50 % naked rock, (3) walls higher than 1 m, or (4) boulder stones with diameter larger than 1 m. The total number of macro sample plots, 100, was distributed on the transects according to transect length. Within each transect, the desired number of sample plots were randomly chosen from the accepted transect positions. Each macro sample plot was divided into 16 subplots, 1 m² each. Two randomly chosen subplots, constrained to be situated along the margin of the macro plot, were fixed to be taken as meso sample plots. Meso plots were rejected if they included (1) more than 25 % naked rock, (2) walls higher than 0.25 m, (3) boulder stones with diameter larger than 0.25 m, or (4) a tree higher than 2 m, rooted within the plot. A rejected meso plot was replaced by another macro subplot, selected from a fixed priority list. Meso plot corners were permanently marked by subterranean aluminium tubes. Only the lowermost meso plot in each macro plot is included in this published data set. Each meso sample plot was divided into 16 subplots, 0.0625 m² each. As the vegetation of the field and bottom layers (vascular plants including lignified species less than 80 cm high, bryophytes and lichens) was the main object of this study, the tree and shrub layers were treated as environmental variables influencing the lower layers. Vegetation was analyzed at the meso scale (plot size 1 m²). Presence/absence of all species was recorded for each subplot. For vascular plants, presence by cover and presence by rooting were both recorded by the published data set uses presence by cover. Frequency in subplots was calculated for each species and each meso plot and recorded on a 0-16 scale. The published data set includes 7 replicates for each plot; collected at 5-year intervals (1988, 1993, 1998, 2003, 2008, 2013 and 2018). An eighth replicate is scheduled for 2023. Recording of environmental variables Environmental variables were registered for (1) background information, (2) interpretation of variation in vegetation, and (3) monitoring changes in vegetation and environmental variables over time. The environmental variables of the first and second groups mainly follow T. Økland (1989, 1990), and are described briefly below (see the original paper by Økland & Eilertsen 1993 for details and references). Variables exclusively belonging to the third group are the same as recorded in NIJOS' national grid of forest monitoring sites (Rørå et al. 1988, also see T. Økland 1990), and will not be further treated here. The environmental variables included can conveniently be divided into macro scale variables, meso scale variables, and meso scale humus layer variables. Below we give a brief account of the 33 recorded environmental variables. For a fuller account, including equations and references, see the original paper by Økland & Eilertsen (1993). For each macro sample plot, the exact position of all trees, their canopy perimeters, fallen logs, stumps, boulder stones, naked rock, as well as special details, were mapped. All trees (> 2 m high) rooted within a 64 m2 plot having the 16 m² macro sample plot in the centre, and all other trees with canopies covering the macro plot, were mapped as well. Mapped trees were numbered consecutively, and subjected to the following measurements: The following variables were measured to be representative for the macro sample plots. (1) Slope (MA Slo) was measured by a compass (90° scale). (2) Aspect favourability (MA Asf) was calculated from aspect measured by a clinometer (400 degrees scale). The measurements were converted to a heat index on a linear degrees scale, following Dargie (1984), Parker (1988) and Heikkinen (1991): SSW (225 g) was considered the most favourable aspect, and given the heat index value of 0; NNE (25 g) was considered the least favourable aspect and given the heat index value of 200. Intermediate values were calculated by the following formulae: where ai is the recorded aspect in macro plot i and hi is the heat index. (3) Terrain shape (MA Ter) was scored on a six point scale: 0 - valley bottom or concave terrace, 1 - concave valleyside, 2 - plane valleyside, 3 - convex valleyside, 4 - ridge, 5 - hilltop. (4) Surface unevenness (MA Une) was scored on a four point scale (cf. Rørå et al. 1988): 1 - relatively even (6 terrain roughnesses or less within the 64 m² plot enclosing the macro plot; a roughness defined to deviate more than 0.35 m from the surrounding terrain surface), 2 - uneven (7 or more roughnesses), 3 - boulderfield, 4 - coarse, with vertical walls, clefts and cliffs. (5) Soil depth (MA S d) was scored on a four point scale, based on observations of the surface relief within the 64 m² plot (cf. Rørå et al. 1988): 1 - < 25 cm (extensive rock outcrops), 2 - 25-50 cm (localized rock outcrops), 3 - 50-100 cm (no rock outcrops, terrain uneven), 4 - > 100 cm (even surface, glaciofluvial material totally concealing unevennesses of the parent material). (6) Basal area (MA Bas) was determined by a relascope (Fitje & Strand 1973). Basal area was measured at breast height from the lower left corner of each meso sample plot (16 ME Bas), using relascope factor 2. Values for the two meso sample plots were averaged to give MA Bas. Basal area is an expression of tree density and thus gives information of the light supply to the understory. (7) Canopy cover (MA Can), was calculated by a formula given in Økland & Eilertsen (1993). The following variables were measured to be representative for the meso sample plots: (8) Slope (ME Slo) was measured by a compass (see 1). (9) Aspect favourability (ME H i) was calculated from aspect measured by a clinometer (see 2). (10) Unevenness (ME Une). For each meso sample plot, microtopography was recorded in the field as follows: A 1 m² steel frame, used for recording vegetation, was levelled, and the vertical distance from the levelled frame to the soil surface at the centre of each of the 16 subplots was measured. These 16 observations were recalculated to heights above the lowest relative level in the meso plot, zi, i = 1,...,16. The zi values can be considered as a function of x, position in sample plot from left to right (0,1,2,3) and y, position from bottom to top (0,1,2,3); zi = f(xi,yi). The plane of best fit to the observation was estimated by bivariate regression taking the zi values as the dependent variable and x and y as independent variables; the model for the systematic part of the regression being Ez = a1x + a2y + a0 The regression was used to estimated fitted values z'; zi' = a1xi + a2yi + a0. (4) The deviation of the soil surface from the plane of best fit was ki = zi' - zi. (5) In even terrain, the ki values of adjacent subplots (subplots sharing one edge) differ only slightly in absolute value. There are 24 pairs of adjacent subplots within one meso plot. The following equation was used to measure unevenness, u: u = (SUM(i,j │ki - kj│)/24 where the sum is over all pairs of adjacent subplots. (11) Convexity (ME Con). The microtopography measurements (see 10 above) were used. Convex and concave sample plots will have ki values that are systematically distributed over the plot (as a function of x and y). Convex plots will have a maximum of ki close to the centre of the plot, while concave plots will have a minimum in this region. The deviation from fitted values near the centre of the plot is calculated as k0 = (k6 + k7 + k10 + k11)/4, where the subscripts i refer to subplot numbers, counted from the lower left of the plot. Subplots 6, 7, 10 and 11 are the four subplots bordering on the plot centre. The mean deviation of ki from k0 for the remaining 12 subplots can be used as an index of the convexity of the meso plot: co = (SUMi (k0 - ki)/12, (8) where i is the values from 1 to 16 different from 6, 7, 10 and 11. Values of co > 0 indicate convex plots, values < 0 indicate concave plots, while values < 0 indicate plane, uneven or saddle-shaped plots. (12-14) Soil depth. Soil depth was measured as the distance possible to drive a steel rod into the soil. Measurements were magde at eight fixed points 25 cm off the ede of the meso sample plot; 2 points along each edge. The set of measurements was used to make three variables: (12) Soil depth, minimum (ME Smi), (13) Soil depth, median (ME Sme), and (14) Soil depth, maximum (ME Sma). (15) Litter index (ME Lit). Amount of litterfall was estimated for each meso sample plot by consideration of the position of the plot relative to all trees covering the plot, and to characteristics of the trees (see Økland & Eilertsen 1993 for details). (16) Basal area (ME Bas) was determined by a relascope (Fitje & Strand 1973). Basal area was measured at breast height from the lower left corner of each meso sample plot using relascope factor 1 (also see 6 MA Bas). The following set of variables were measured to be representative for the humus layer (or the upper 5 cm of the humus layer, if thicker). Two sets of samples were collected; one set for determination of soil moisture, and one set for determination of chemical and physical properties of the humus layer. Samples for determination of soil moisture were collected on 15-16 Oct 1988, after several days without precipitation. Two cores, 5 cm high and 98 cm³ each, were collected just below the sample plot. The cores were transferred to plastic bags and kept frozen until analysis. Samples for chemical and physical analysis were taken on 15-16 Sept 1988. Several (5-10) small samples, 50-100 cm³ each, were collected and mixed. They were kept in the frozen state for several months. Before analysis at Landbrukets Analysesenter, Ås they were dried at 38°C, grounded and sifted with 2 mm mesh width. Exchangeable cations were determined by adding 50 cm³ 1 M NH₄NO₃ solution to 10 g dried soil (cf. Stuanes et al. 1984). The solution was left overnight, filtered, and the sediment washed with 1 M NH₄NO₃ until the volume of extract amounted to 250 cm³. Element concentrations ((21) Ca, (22) Mg, (23) Na, (24) K, (28) Al, (30) Mn, (31) Zn, (32) P, and (33) S, were determined in the extract by a Jarrell Ash ICAP 1100 instrument. (17) Soil Moisture (Mois). Volumetric soil moisture was determined by weighting the fresh samples, drying the samples at 110°C until constant weight, and reweighting. (18) Loss on ignition (LI) was determined by ashing a sample at 550°C in a muffle furnace. (19) pH, aquous solution (pHH₂O). One part dried sample was mixed with 2.5 parts distilled water and left overnight. pH was measured the next day with an Orion SA 720 meter. (20) pH, measured in CaCl₂ (pHCaCl₂). One part dried sample was mixed with 2.5 parts 0.01 M CaCl₂, otherwise as (19). (25) Echangeable H [H₃O+]. 50 ml of the extract was titrated with 0.05 M NaOH until pH = 7.0. The volume of NaOH was corrected for the value used with pure extractant, to obtain the result. (26) Total N. Kjeldahl-N was determined by digestion of the dried sample with H₂SO₄, and use of a Se catalyst in a Tecator FIA system. (27) Total P (P-AL). One part dried sample was mixed with 20 parts of a solution 0.1 M with respect to ammoniumlactate and 0.4 M with respect to acetic acid. pH was adjusted to 3.75. P was determined in the extract by Jarell Ash ICAP 1100.

Study Extent For general description of Study area, see "Study area" above

Method step description:

  1. See sampling description

Bibliographic Citations

  1. Økland, R.H. & Eilertsen, O. 1993. Vegetation-environment relationships of boreal coniferous forests in the Solhomfjell area, Gjerstad, S Norway. ‒ Sommerfeltia 16: 1-254
  2. Økland, T., Bakkestuen, V., Økland, R.H. & Eilertsen, O. 2004. Changes in forest understory vegetation in Norway related to long-term soil acidification and climatic change. ‒ J. Veg. Sci. 15: 437-448.

Additional Metadata

Alternative Identifiers b4934203-2d86-44a6-9fec-b55d7ab97749