Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33
34
An inventory of dead wood in four plots of differing wildfire history in a tall, wet Eucalyptus obliqua forest in southern Tasmania, Australia.
G.M. Gates
1,2
, T.J. Wardlaw
3
, D.A. Ratkowsky
1,2*
, N.J. Davidson
2
and C.L.
Mohammed 1
1
School of Agricultural Science, University of Tasmania, Private Bag 54, Hobart,
Tasmania 7001, Australia
2 School of Plant Science, University of Tasmania, Private Bag 55, Hobart, Tasmania
7001, Australia
3
Forestry Tasmania, GPO Box 207, Hobart, Tasmania 7001, Australia
*e-mail: D.Ratkowsky@utas.edu.au
(corresponding author)
Abstract
Wildfires in southern Tasmania’s tall wet eucalypt forests generate dead wood of varying sizes and in many different stages of decay depending on their intensity and frequency. Large dead wood (coarse woody debris, CWD) and stags were measured and their attributes recorded in four 50x50 m plots in close proximity but with differing wildfire histories in a native wet Eucalyptus obliqua forest. Maps were drawn and are presented here of the CWD in the four plots. In addition, the diameters at breast height (DBH) of all living higher vascular plants were measured, and the positions for all stems having DBH≥10 cm were recorded. The maps of the living vegetation portray the substantial differences between the four plots.
Information from other recent studies of CWD in the same or similar forests provided a degree of replication to this study. Whereas the CWD volumes in plots of the same age since wildfire obtained from different studies proved to be very variable (reflecting the chance location of large fallen eucalypts in the plots) the proportion of plot basal area occupied by each genus of living tree was remarkably consistent, and depended upon the time since the last wildfire event. This agreement lends support to a recent model for canopy structure based upon measurement of the
DBH and relative location of each living tree in the plot, and its identification to genus level.
Keywords: dead wood, eucalypt, coarse woody debris, stags, wildfire, stand regeneration, stand development, canopy structure model.
1

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33
34
Introduction
Dead wood in forest ecosystems has been recognised for some time as being important in providing a range of ecological niches that foster biodiversity by maintaining many specialist wood-dwelling and hollow-dependent species, and as a temporary sink for forest carbon and other elements (e.g. Harmon et al . 1986; Grove et al . 2002; Stokland and Sippola 2004; Wu et al . 2005). In a forest ecosystem, dead wood consists of all dead natural structures of woody plant origin, which includes dead roots, stumps, fallen tree trunks, branches and twigs, and standing dead trees
(stags). Natural mortality of a tree occurs due to ageing and suppression caused by competition as the stand develops after disturbance. The disturbance may be natural, e.g. windthrow, wildfire, earthquakes or anthropogenic, e.g. harvesting for timber and firewood. The quantity of dead wood input into the ecosystem following any of these disturbances may be large and immediate as with a catastrophic event or a more gradual temporal process, which may be seasonal, annual or long-term
(Harmon et al . 1986). Spatial input may be within stands or across landscapes and catchments. Most inputs of dead wood occur at the local scale (e.g. within one tree length of source) but ecological processes based on dead wood habitats can operate at a range of scales. Knowledge of the natural dynamics of dead wood provides important baseline data that can be used for developing and evaluating strategies to lessen the pressure of anthropogenic disturbance on wood-inhabiting organisms
(Jonsson 2000).
In Australian eucalypt-dominated forests, fire is the major cause of large-scale natural disturbance. Wildfires vary in type (Luke and McArthur 1978), intensity (Gill
1997), size, frequency and homogeneity (Ashton 1981), resulting in differing starting points for new stand development. There is generally a lack of knowledge regarding the log accumulation rate, i.e. the time frame over which trees fall and form logs on the forest floor, the rates of log decay and how these rates differ between managed and unmanaged forests in Australia (Lindenmayer et al . 2002). This lack of knowledge makes it difficult to determine how long it may take logged areas to accumulate volumes of CWD equivalent to pre-harvesting levels and to establish silvicultural regimes that ensure that forests are sustainably managed (Lindenmayer et al . 2002).
2

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
In the commercially important wet lowland eucalypt forests of southern
Tasmania, mature trees of Eucalyptus obliqua L’Hér. frequently attain a height of over 70 m (Kirkpatrick and Backhouse 1981), enabling these forests to produce a wood volume per hectare which is amongst the highest produced by any forest in the world (Woldendorp and Keenan 2005). Mortality in these forests results not only from catastrophic fire, but also in developing stands through natural selection and suppression of smaller, weaker individuals. These smaller stems are commonly killed by insect and fungal attack or a combination of these factors. Smaller diameter CWD may also accumulate via branch wood that has fallen from the canopy. In these forests, non stand-replacing wildfires occur more frequently than stand-replacing wildfires (Alcorn et al . 2001; Turner et al . 2009). This has resulted in a mosaic of multi-aged forest stands of largely unknown dead wood complexity. A recent study employing an indirect approach (Grove et al . 2009) suggested that E. obliqua CWD in Tasmania’s southern forests decomposes very slowly in comparison with CWD from most other parts of the world.
Volumes of CWD and numbers of stags are two of the attributes used to measure stand structure and which provide quantitative evidence of habitat that can be used in biodiversity studies (McElhinny et al . 2005). The volume input, connectivity (spatial) and continuity (temporal) of CWD are important considerations in sustainable forest management (Grove et al . 2002). The aim of the present study was to inventory the dead wood present in a tall wet E . obliqua forest in southern Tasmania containing stands resulting from different wildfire events and to compare the CWD volumes therein with available information from other recent studies carried out in the same forest type (Woldendorp et al . 2002; Sohn 2007). Of particular interest is a ground-level method for modelling canopy structure in these same forests (Scanlan et al . 2010), which is based upon identification of each living tree, measurement of its diameter at breast height (DBH) and its relative location in the plot. This enables one to examine whether there is a correlation between canopy structure, as revealed using that methodology, and the quantity of CWD that arises from natural disturbance, particularly fire, within wet E . obliqua forest.
3

Dead wood in tall, wet E. obliqua forest
21
22
23
24
25
17
18
19
20
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
29
30
31
32
33
26
27
28
METHODS
Study area
The study area was at the Warra Long Term Ecological Research (LTER) site in the
Huon River valley, southern Tasmania, Australia, where four 50x50 m plots were established in March-April 2006 along the ‘Bird Track’ (see Fig. 1). The plots were all within ca. 1 km of each other (lat./long. S 43º 06′, E 146º 39′) and had similar south-facing aspect, altitude, rainfall and temperature, but differed in their histories of natural disturbance by wildfire. Documented accounts, maps of fire history and fire scars on E. obliqua trees were used to determine age since fire (Turner et al .
2007). Time since wildfire for each of the four plots was estimated, respectively, to be at least 200 years (named ‘Old growth’), 108 years (named ‘1898’ as it was burnt by a fire in 1898, but later it was discovered that parts of it had also been burnt by a fire in 1934), 72 years (named ‘1934’, burnt by a fire in 1934) and 108 years/72 years (named ‘1898/1934’, burnt by fires in 1898 and in 1934). Plot names are used for convenience but also reflect, at least to some extent, the disturbance history of the plot.
Each 50x50 m plot was established in the following way. Star pickets were placed at 10 m intervals along the outer boundaries of the two opposite sides of the plot. Twine was strung from the star pickets across the plot and fibreglass rods were placed at 10 m intervals along the twine to divide the plot into 25 subplots each measuring 10x10 m. This facilitated mapping of the CWD and stags, and assessment of the vascular plants.
Vegetation classification and measurement
In each plot, all woody perennial species were identified (following nomenclature of
Buchanan 2009). The plant communities in each plot were described following the
Forest Practices Code (Forest Practices Authority 2005). Within each of the 25 subplots of each plot, all living stems were measured and, if they were at least 10 cm in diameter, mapped.
4

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
CWD Mapping
CWD as used in this study was defined to be all pieces of dead wood ≥10 cm in diameter and ≥1 m in length, a modification of the recommendation of Harmon &
Sexton (1996). This definition of CWD included stumps, suspended pieces of wood, and shards (shattered pieces of larger logs) as well as fallen trunks and branches on the forest floor. CWD originating from all woody perennial species were included in the study. CWD was consecutively numbered within each subplot. If a piece of
CWD traversed two or more subplots, its length was measured to the boundary of the subplot and it was renumbered as a separate piece of CWD in the adjacent plot. The position, orientation and attributes (see below) of every piece of CWD in each subplot were recorded for each of the four sites. This information was transcribed onto large sheets of graph paper (laminated to make it usable in wet weather) marked with plots and subplots at a scale of 1 mm equal to 10 cm. The following attributes of each piece of CWD were recorded: 1) CWD length (cm); 2) CWD diameter (cm) measured at the mid-point of the piece of CWD; 3) CWD decay class (using a scale from 1 to 5 with intervals of 0.5; see Table 1a); 4) Percent bryophyte cover on each piece of CWD (a visual score of 0-100%). Stumps were measured for decay class, height and mid diameter (i.e. the diameter mid-way between the ground and top of stump).
The system of decay classes used here for CWD (Table 1a) was devised to accommodate different wood species and to try to overcome the problems associated with the unevenness of the interval between decay classes 3 and 4. If a piece of
CWD had more than one decay class, an average was taken (after Pyle & Brown
1999). For analysis, CWD was placed into the following diameter classes: ≤15 cm,
15-30 cm, 30-60 cm, 60-90 cm, 90-120 cm, 120-150 cm, >150 cm. These classes were deemed to be most useful in forest management by Forestry Tasmania (Simon
Grove, pers. comm.; Yee 2005). During analysis, other variables were derived from length and diameter by calculation, viz . volume and surface area, assuming that the shape of a piece of CWD approximated a cylinder. For stumps, height replaced length.
5

Dead wood in tall, wet E. obliqua forest
22
23
24
25
26
27
28
29
30
31
32
33
16
17
18
19
20
21
12
13
14
15
10
11
8
9
5
6
7
3
4
1
2
Stag mapping
Stags were recorded in a similar way to CWD, except 1) height was used in place of length; 2) diameter was measured at breast height; 3) decay was assessed using a modified system to that used for CWD, see Table 1b, following Cline et al . (1980),
Spies et al.
(1988) and Motta et al.
(2006).
Statistical analyses
The majority of the statistical analyses used were of a descriptive nature, producing summary statistics. Although the CWD was measured within each subplot, enabling the data to be analysed at a subplot level, which proved useful for a subsequent survey of the macrofungi growing on wood (see Gates et al.
2011), the analyses presented here used composite pieces of CWD, obtained by concatenating the information on pieces of CWD that crossed subplot boundaries. Graphical methods presented here involved one variable at a time for decay class and % bryophyte cover on CWD, but we also used profile plots, as devised by Stokland (2001), in which two explanatory variables (diameter and decay class) were considered simultaneously.
Stags were examined by calculating the number of stags and their diameters in each plot and by recording the number of stags in each bryophyte cover class.
Results
Vegetation classification
Maps showing the positions of the most frequently occurring vascular species, being in the genera Acacia , Atherosperma , Eucalyptus , Monotoca , Nothofagus and
Pomaderris , are presented in Fig. 2. In ‘Old growth’, Nothofagus cunninghamii and
Atherosperma moschatum were predominant and Olearia argophylla absent. There were only two living eucalypts which accounted for more than half the basal area
(Table 2, Fig. 2a). This identifies ‘Old growth’ as close to, but not identical with, plant community RAIN-CT1 (Forest Practices Authority 2005), the presence of the eucalypts making the community a mixed forest rather than a true rainforest.
6

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
In ‘1898’, the presence of
Olearia argophylla and Phyllocladus aspleniifolius , the absence of Anodopetalum biglandulosum , and the sparseness of Gahnia grandis and Anopterus glandulosus (Table 2) places it near plant community WET-OB1000
(Forest Practices Authority 2005). The presence of Pomaderris apetala (193 stems) was indicative of a second fire in parts of the plot. This is evident from the map (Fig.
2b), where a broad band of living Atherosperma and Nothofagus runs from the upper right-hand corner towards the lower central and lower left-hand corner of the plot, with the remainder of the plot containing more than 150 live Pomaderris trees. The area containing the rainforest vegetation, but lacking Pomaderris , form a partition of
‘1898’ that we designate here as ‘1898R’, and is very close to being considered
‘mature’ forest (Hickey et al . 1999). The second distinct vegetation type into which
‘1898’ was partitioned, designated here as ‘1898P’, is found in the
Pomaderris abundant regions in the upper left-hand corner and the lower right-hand corner of
Fig. 2b. These areas also contain Acacia and Eucalyptus , but the typifying rainforest species Atherosperma and Nothofagus can be seen to be absent or extremely sparse.
In the ‘1934’ plot
Phyllocladus aspleniifolius , Anopterus glandulosus and
Monotoca glauca were frequent or common and Pomaderris apetala , Olearia argophylla and Acacia verticillata were sparse or absent (Table 2). This suggests that the plant community of the ‘1934’ plot is close to, but not identical with, WET-
OB1100 (Forest Practices Authority 2005). The Eucalyptus and Monotoca trees occurred throughout the entire plot, but not uniformly so (Fig. 2c).
The ‘1898/1934’ plot contained abundant
Pomaderris apetala (832 stems) but also had numerous stems of Eucryphia lucida , Phyllocladus aspleniifolius and
Anopterus glandulosus (Table 2). It is closest to WET-OB1001, which occurs on humid slopes and gullies on less fertile sites than WET-OB1000 (Forest Practices
Authority 2005). However, the large numbers of Gahnia grandis plants (132) and the presence of Olearia argophylla (17 stems), both of which reflect disturbance, are anomalous for this forest type. The distribution of Eucalyptus and Pomaderris in the left-hand portion of this plot is relatively uniform, but the clusters of rainforest species in the right-hand portion, reminiscent of the partition ‘1898R’ of ‘1898’ but on a smaller scale, suggests that the 1934 fire did not burn the plot evenly.
7

Dead wood in tall, wet E. obliqua forest
22
23
24
25
26
18
19
20
21
14
15
16
17
10
11
12
13
8
9
27
28
29
30
31
5
6
7
3
4
1
2
CWD and stag volumes
For each plot separately, minimum, median and maximum CWD subplot volumes are given in Table 3, which also gives the total plot volumes. Based on total volume
(m 3 /ha) of CWD present, ‘1898’ had more than twice the volume of ‘1898/1934’, whereas for stags (Table 4), ‘1898’ had the lowest volume. If all the stags had fallen to become CWD, the rank orders of the total amount of dead wood in the plots would remain unchanged, as the stag volumes in a plot are only a fraction (

21%) of the
CWD volumes. That is, ‘1898’ would still be the plot with the most total dead wood,
‘1934’ would remain second highest, ‘Old growth’ would remain next and
‘1898/1934’ would still have the least total dead wood.
CWD maps
Maps showing the positions of CWD in each plot are given in Fig. 3. There were 227 pieces of composite CWD in ‘Old growth’, 138 in ‘1898’, 212 in ‘1934’ and 237 in
‘1898/1934’. The relative sparseness of pieces of CWD in ‘1898’ compared with the other plots is readily observable.
CWD attributes vs. decay class
Most of the 814 pieces of composite CWD within the four plots fell into the middle decay classes (DC3 and DC3.5), with 522 pieces of CWD (64.1%) in these combined classes (Fig. 4). ‘Old growth’ had a very small percentage of CWD in the lower decay classes (DC≤2.5) compared with younger plots, but compensated for this in the higher decay classes (DC≥4), reflecting the shift in decay class distribution that occurs with increasing age of plot. Bryophyte cover tends to increase steadily as decay class increases in units of 0.5 from DC1–DC5 in all plots combined (Fig. 5).
Slightly deviating from the overall trend is ‘1934’, which reaches a plateau at a percentage bryophyte cover of ca. 50% in the higher decay classes.
8

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33
34
Profile plots
The percentages of pieces of CWD graphed simultaneously against diameter class and decay class in profile plots (Fig. 6) shows that there were no pieces of CWD greater than 60 cm diameter in DC<3 in any of the four plots. In addition, very large diameter CWD (DBH>120) was generally absent from the highest decay classes. In
‘Old growth’, almost all of the CWD was found in low diameter classes. ‘1898’ had more CWD in the higher diameter classes than any other plot.
Stag numbers and attributes
The ‘1934’ plot had the greatest number of stags, closely followed by ‘Old growth’, while ‘1898’ had the least number (Table 4). Although ‘1934’ had 34 stags, all but one of them was of small diameter (Fig. 7), so that the stags of that plot had the smallest average diameter of the four plots (Table 4). In contrast, although
‘1898/1934’ had only 20 stags, that plot had the highest average diameter due to four stags of large diameter, each >100 cm (Fig. 7). ‘Old growth’, with the second highest number of stags, had two large diameter ones (Fig. 7), giving it the second largest average diameter (Table 4).With respect to species composition, there is a sharp contrast between the younger stands, ‘1934’ and ‘1898/1934’, which had 16 and 12
E. obliqua stags, respectively, and the mature forests, ‘1898’ and ‘Old growth’, which had only one E. obliqua stag each (Table 4). The identifiable stags in ‘Old growth’ are mainly of Nothofagus cunninghamii , Atherosperma moschatum and
Acacia melanoxylon , typical rainforest species.
The number of stags as a function of bryophyte cover class and plot is shown in Fig. 8. Increasing bryophyte cover is associated with increasing age of plot, with few stags in ‘1934’ and ‘1898/1934’ having bryophyte cover of 25% or more.
Basal areas of living vegetation from another study
The four plots of the present study each had a unique wildfire history and therefore there is no replication with respect to age since wildfire. However, other plots in the tall wet E. obliqua forests of southern Tasmania are available for comparison. These plots are derived from the wildfire chronosequence study of Turner et al . (2007),
9

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33
34 with the basal areas measured by Scanlan et al . (2010). Three of these plots, viz .
‘OGS’, ‘1898S’ and ‘1934S’, were located adjacent to plots ‘OG’, ‘1898’ and
‘1934’, respectively, of the present study. It is believed (P. Turner, pers. comm.) that the whole of ‘1898S’ was affected by a second fire in 1934, but like that of ‘1898’ of the present study, that was not known at the time of plot establishment. Results for the basal areas of the living vegetation are given in Table 5, where, following
Scanlan et al . (2010), the woody perennial species are tabulated at the level of genus, a restriction that affects only Acacia , which combines occurrences of A. melanoxylon , A. verticillata and A. dealbata . All other genera had only a single species that was found in the plots.
Because ‘1898’ of the present study is really made up of two fire histories, it was split into two parts for inclusion in Table 5, with ‘1898P’ based on subplots where Pomaderris was the predominant understorey species, and ‘1898R’ based on subplots where rainforest species were predominant. A younger plot (‘1966S’) used in the chronosequence study of Turner et al . (2007) is also included in Table 5, as it had a southerly aspect, in common with all other plots in that table. There is excellent agreement between ‘OGS’ and ‘OG’ in the proportions of all the genera that appeared in those plots. There is also good agreement between ‘1934S’ and
‘1934’ for
Eucalyptus , Nothofagus and Acacia , but there was slightly more
Atherosperma and less Monotoca in ‘1934S’ than in ‘1934’. The three plots,
‘1898P’, ‘1898S’ and ‘1898/1934’, had similar fire histories, as all were burnt twice
(by wildfires in 1898 and in 1934) and therefore might be expected to have had similar stand compositions. The results in Table 5 indicate that this is largely the case, with ‘1898P’ agreeing reasonably well with ‘1898/1934’, except for the presence of slightly more eucalypts and virtually no rainforest species. ‘1898S’ also broadly agrees with ‘1898/1934’, although the former had more Acacia and
Pomaderris , but less Eucalyptus , than the latter plot.
The youngest plot, ‘1966S’, had the highest proportion of Eucalyptus of any of the plots in Table 5, and the two oldest plots, ‘OG’ and ‘OGS’, had the lowest proportion. These results are consistent with a progressive decline of the percentage basal area of Eucalyptus with increasing plot age of a chronosequence.
Volumes of CWD from other studies
10

Dead wood in tall, wet E. obliqua forest
24
25
26
27
28
29
30
31
32
33
34
21
22
23
17
18
19
20
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
There are other plots in the tall wet E. obliqua forests of southern Tasmania for which CWD volumes have been determined. Woldendorp et al . (2002) inventoried the CWD in two wildfire-affected plots in the Warra LTER, both having a southerly aspect and which were not amongst the four plots of the present study. The size of these plots was 1 ha each, i.e. four times the size of each of the plots of the present study, and the definition of CWD was slightly different, having a lower diameter and length limit of 15 cm and 50 cm, respectively (instead of the 10 cm and 1 m limits used here). Additional plots were inventoried by Sohn (2007), viz . ‘1966S’, ‘1934S’,
‘1898S’ and ‘OGS’, all of which were used in the chronosequence study of Turner et al . (2007), and which were also subsequently used by Scanlan et al . (2010) for modelling canopy structure. The plots of Sohn (2007) were each 50x50 m, as in the present study, but the lower diameter limit of CWD was 40 cm and 1 m length
(compared to a 10 cm lower limit in the present study or a 15 cm limit in that of
Woldendorp et al . 2002). The volumes of CWD from the three studies, with ‘1898’ replaced by its partitions ‘1898P’ and ‘1898R’, are graphed in Fig. 9 on a per hectare basis against age since wildfire (year of wildfire minus year of measurement).
Despite the clearly large variance of CWD volume for plots of the same or closely similar ages, a quadratic polynomial fitted to the 11 data points (Fig. 9) produces a pattern that is broadly consistent with the model of Stamm & Grove (unpublished, but see Grove 2009), where a stand reaches a maximum CWD volume at some intermediate stage between the initiation of the fire and being designated as an old growth stand.
Discussion
The sources of CWD are (1) the stand prior to the disturbance, (2) the direct result of the disturbance itself, and (3) an ensuing gradual input from the current stand, including mortality caused by disease, suppression and competition, insect attack, and windthrow. The four plots chosen for this study, although located in relatively close proximity to each other, have different fire histories and therefore probably have different mechanisms by which the major part of their CWD was likely to have originated. In Tasmania, in the long absence of fire and in areas where the rainfall exceeds 1270 mm, ecological drift occurs (Jackson 1968). This means that the understory of eucalypt dominated forests progressively becomes dominated by cool
11

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33
34 temperate rainforest (mixed forest) and, as the eucalypts die without regeneration, the eventual outcome is climax rainforest that may take ca. 400 years to occur. The
‘Old growth’ plot fits the definition of mixed forest (old, even-aged eucalypts, with an understory of mature rainforest; see Gilbert 1959 and Wells & Hickey 1999). The live vegetation showed floristic simplification with a preponderance of mature rainforest species and two very large surviving eucalypts (Table 2, Fig. 2a). Only one stag in ‘Old growth’ was of E. obliqua origin, compared to 32 stags of rainforest and/or other species (Table 4). Pieces of CWD in ‘Old growth’ had the highest percent bryophyte cover of all the plots, a consequence of the direct relationship between decay class of the wood and percent bryophyte cover (Fig. 5). The most likely origin of the high percentage of pieces of CWD in high decay classes and of small diameter in ‘Old growth’ (Fig. 6a) was from branches breaking out of declining eucalypt crowns and from the tops of stags from climax rainforest species falling to the forest floor. The sparseness of large diameter CWD in high decay classes in ‘Old growth’ suggests that sufficient time (>300 yr., Grove et al . 2009) had elapsed for the CWD resulting from the death and falling of the original mature eucalypt stand to rot away.
The ‘1898’ plot was made up of two distinct vegetation types. The partition
‘1898R’, an area characterised by living rainforest species, had a CWD volume of
1481 m
3
/ha, due to some very large pieces of CWD of E. obliqua origin that may have resulted from trees killed by an intense and possibly stand-replacing fire in the year 1898. These trees likely fell immediately after the fire or subsequently as a result of wind or disease. Any small diameter branch wood or suppressed trees of small diameter from the regenerating stand could have had sufficient time (108 years) to rot away, which may explain the relatively low percentage of pieces of
CWD of small diameter (Fig. 6b). However, the stand may not have been old enough for the accumulation of small diameter CWD of rainforest species as in ‘Old growth’. In this rainforest partition of ‘1898’, there were three very old N. cunninghamii stags consistent with an old growth plot. The partition
‘
1898P
’
had a
CWD volume almost as high as for
‘
1898R
’
, also due to a few very large diameter trees, but in a significantly lower decay class, consistent with fallen wood being on the forest floor for a shorter period of time.
In ‘1934’, the lower average decay class of the CWD reflects the shorter time
(72 years) that the wood has been lying on the forest floor. The high number of small
12

Dead wood in tall, wet E. obliqua forest
21
22
23
24
17
18
19
20
25
26
13
14
15
16
9
10
11
12
7
8
5
6
3
4
1
2
27
28
29
30
31
32
33 diameter stags (Fig. 7) may reflect suppression mortality in the regenerating stand. A striking difference between the composition of the living stems of ‘1934’ and that of
‘1898/1934’ and of ‘1898P’, plots or parts of plots that experienced a second fire in
1934, is the presence of Monotoca glauca and the absence of Pomaderris (Table 2,
Fig. 2c). This can be attributed to a different underlying geology. Whereas the other plots are on soils derived from Jurassic dolerite, ‘1934’ is situated on Permo-Triassic sedimentary rock, which produces a more acidic soil type that favours Monotoca in place of Pomaderris (M.G. Neyland, pers. comm.).
The ‘1898/1934’ plot had the smallest volume of CWD (Table 3), which is consistent with the second fire consuming the CWD generated by the first fire.
Alternatively, perhaps the large diameter trees that were killed by the fire of 1898 did not fall immediately but remained as stags, which survived the fire of 1934 (see
Fig. 7). Any small diameter stags that resulted from regeneration after the first fire, later suppressed by competition to become small diameter CWD on the forest floor, were likely to have been consumed by the second fire. Suppression mortality in the regenerating stand following the second fire was likely to have been responsible for the many small diameter stags of E. obliqua origin (Fig. 7) , similar to ‘1934’.
The close similarity of the basal area percentages for each genus of trees and shrubs in nearby plots (or partitions of plots) having the same wildfire history (‘OG’ vs. ‘OGS’; ‘1934’ vs. ‘1934S’; ‘1898S’ vs. ‘1898P’) is evidence of a consistency in the compositions of the higher vascular species resulting from wildfire. This is of particular relevance because of the existence of a recent ground-level method for modelling canopy structure that was developed using data from these same forests
(Scanlan et al . 2010). The method requires identification of each living tree and measurement of its DBH and its relative location in the plot. Using these groundbased measurements and a series of regression equations developed from empirical data to predict the crown dimensions of trees of each of the six important canopy genera, Scanlan et al . (2010) successfully modelled canopy structure. This suggests that the canopy can be modelled at the stand or landscape level from comparatively quick and inexpensive measurements made at ground level. The consistency between the basal area compositions of the present study and those determined in similar plots by Scanlan et al . (2010) is encouraging, as it suggests that it may be possible to map and monitor structural variation across forested landscapes from readily and
13

Dead wood in tall, wet E. obliqua forest
21
22
23
24
25
26
27
28
29
30
31
32
33
34
13
14
15
16
9
10
11
12
17
18
19
20
7
8
5
6
3
4
1
2 inexpensively obtained ground-based measurements without the use of specialised equipment.
CWD volumes, unlike basal area, vary greatly among forest plots that are otherwise very similar in wildfire history and stand composition. Discrepancies in
CWD volumes of the order of magnitude observed in Fig. 9 (e.g., a range of 366–
1481 m
3
/ha for logs + stumps at an age of 72–73 years) cannot be attributed to differences in the lower diameter limit, i.e. had a lower diameter limit of 40 cm been used in the present study instead of 10 cm, 92.3% of the total volume would still have been observed, the total volume being determined mainly by large diameter logs. The amount of CWD in a 50x50 m plot is largely a matter of chance, the main contributors to CWD volume being fallen eucalypts, some of which had heights exceeding the 50 m plot length. A single fallen tree can have a big effect on CWD volume, e.g. the largest piece of CWD in ‘Old growth’ (clearly visible on the righthand side of Fig. 3a) accounted for 42.1% of the CWD volume in that plot. There were other large senescent trees and stags outside the boundary of the plot that could have fallen and landed in the plot due to chance. Therefore, the great variance of
CWD volume is not surprising, but despite that variation, the general trend of the volume data against age since wildfire broadly follows the prediction made by a stand dynamics model (Stamm & Grove, unpublished).
Conclusions
Identifying the living vegetation to genus level and measuring the DBH and position of each living tree enables maps to be drawn from which the forest stand structure can be deduced, as shown in a recent study which modelled canopy structure from these readily obtainable ground-level measurements. Unlike CWD volume, which varies considerably, the percentage basal areas occupied by the genera of the living trees are consistent among replicates of plots subjected to the same or closely similar fire history. In a small spatial study such as the current study and those of Sohn
(2007) and Turner et al . (2007), the volume of dead wood in the tall, wet E. obliqua dominated forests of southern Tasmania does not serve as a useful indicator of stand structure due to the chance nature of where senescent trees and stags fall.
Nevertheless, the data on CWD volume (for logs plus stumps) obtained from the present study and two other recent studies using wildfire history broadly support the
14

Dead wood in tall, wet E. obliqua forest
4
5
6
7
8
1
2
3
13
14
35
36
37
38
39
40
41
42
29
30
31
32
33
34
19
20
21
22
23
24
15
16
17
18
25
26
27
28
9
10
11
12 general trend predicted by a model for stand dynamics that is currently under development.
Acknowledgements
Financial and logistic support for the field work was provided by Forestry Tasmania.
One of the authors (GMG) received financial support from an Australian
Postgraduate Award, the Holsworth Wildlife Endowment Fund, the Cooperative
Research Centre (CRC) for Forestry, the Bushfire CRC, and CSIRO Ecosystem
Sciences. Additional logistic support was provided by the Schools of Agricultural
Science and Plant Science of the University of Tasmania.
References
Alcorn, P.J., Dingle, J.K. and Hickey, J.E. (2001). Age and stand structure in a multiaged wet eucalypt forest at the Warra silvicultural systems trial. Tasforests 13:
245-59.
Ashton, D.H. (1981). Fire in tall open-forests (wet sclerophyll forests). In: Fire and the Australian Biota (eds. A.M. Gill, R.H. Groves and I.R. Noble), pp. 339-66,
Australian Academy of Science, Canberra.
Buchanan, A.M. (2009). A Census of the Vascular Plants of Tasmania & Index to
The Student’s Flora of Tasmania
.
(http://www.tmag.tas.gov.au/file.aspx?id=4439).
Cline, S.P., Berg, A.B. and Wight, H.M. (1980). Snag characteristics and dynamics in Douglas-fir forests, Western Oregon. Journal of Wildlife Management 44:
773-86.
Forest Practices Authority (2005). Forest Botany Manual 6: D'Entrecasteaux
Region . Forest Practices Authority, Tasmania.
Gates, G.M., Mohammed, C., Wardlaw, T., Ratkowsky, D.A. and Davidson, N.J.
(2011). The ecology and diversity of wood-inhabiting macrofungi in a native
Eucalyptus obliqua forest of southern Tasmania, Australia. Fungal Ecology
Vol. pp
Gilbert, J.M. (1959). Forest succession in the Florentine Valley, Tasmania. Papers and Proceedings of the Royal Society of Tas mania 93: 129-51.
Gill, A.M. (1997). Eucalypts and fires: interdependent or independent? In: Eucalypt
Ecology: Individuals to Ecosystems (eds J. Williams and J. Woinarski), pp.
151-167, Cambridge University Press, Cambridge.
Grove, S. (2009). A decade of deadwoodology at Warra. Tasmanian Naturalist 131:
25-35.
Grove, S., Meggs, J. and Goodwin, A. (2002). A Review of Biodiversity Conservation
Issues Relating to Coarse Woody Debris Management in the Wet Eucalypt
Production Forests of Tasmania . Forestry Tasmania, Hobart.
15

Dead wood in tall, wet E. obliqua forest
26
27
28
29
30
31
22
23
24
25
32
33
34
35
15
16
17
18
19
20
21
5
6
7
8
1
2
3
4
9
10
11
12
13
14
42
43
44
45
46
47
48
36
37
38
39
40
41
Grove, S.J., Stamm, L. and Barry, C. (2009). Log decomposition rates in Tasmanian
Eucalyptus obliqua determined using an indirect chronosequence approach.
Forest Ecology and Management 258: 389-97.
Harmon, M.E., Franklin, J.F., Swanson, F., Sollins, P., Gregory, S.V., Lattin, J.D.,
Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W.,
Cromack Jr., K. and Cummins, K.W. (1986). Ecology of coarse woody debris in temperate ecosystems. In: Advances in Ecological Research, Vol. 15 (eds A.
MacFadyen and E.D. Ford), pp. 133-302. Academic Press, New York.
Harmon, M.E. and Sexton, J. (1996). Guidelines for Measurements of Woody
Detritus in Forest Ecosystems . Publication No. 20, U.S. Long-Term Ecological
Research Network Office, University of Washington, Seattle.
Hickey, J.E., Su, W., Rowe, P., Brown, M. J. and Edwards, L.G. (1999). Fire history of the tall wet eucalypt forests of the Warra ecological research site, Tasmania.
Australian For estry 62(1): 66-71.
Jackson, W.D. (1968). Fire, air, water and earth – an elemental ecology of Tasmania.
Proceedings of the Ecological Society of Australia 3: 9-16.
Jonsson, B.G. (2000). Availability of coarse woody debris in a boreal old-growth
Picea abies forest. Journal of Vegetation Science 11: 51-6.
Kirkpatrick, J.B. and Backhouse, S. (1981). Native Trees of Tasmania . Pandani
Press, Hobart.
Lindenmayer, D.B., Claridge, A.W., Gilmore, A.M., Michael, D. and Lindenmayer,
B.D. (2002). The ecological roles of logs in Australian forests and the potential impacts of harvesting intensification on log-using biota. Pacific Conservaion
Biol ogy 8: 121-40.
Luke, R.H. and McArthur, A.G. (1978). Bushfires in Australia . Australian
Government Publishing Service, Canberra.
McElhinny, C., Gibbons, P., Brack, C. and Bauhus, J. (2005). Forest and woodland stand structural complexity: Its definition and measurement. Forest Ecology and Management 218: 1-24.
Motta, R., Berretti, R., Lingua, E. and Piussi, P. (2006). Coarse woody debris, forest structure and regeneration in the Valbona Forest Reserve, Paneveggio, Italian
Alps. Forest Ecology and Management 235: 155-63.
Pyle, C. and Brown, M.M. (1999). Heterogeneity of wood decay classes within hardwood logs. Forest Ecology and Management 114: 253-9.
Scanlan, I, McElhinny, C. and Turner, P. (2010). A methodology for modelling canopy structure: an exploratory analysis in the tall wet eucalypt forests of southern Tasmania. Forests 1: 4-24.
Siitonen, J., Martikainen, P., Punttila, P. and Rauh, J. (2000). Coarse woody debris and stand characteristics in mature managed and old-growth boreal mesic forests in southern Finland. Forest Ecology and Management 128: 211-25.
Sohn, J.A. (2007). Variation in coarse woody debris attributes in Tasmanian tall wet
Eucalyptus obliqua (L’Herit.) forests and implications for its monitoring.
Diplomarbeit, Waldbau Institut, Albert-Ludwigs-Universität, Freiburg.
Spies, T.A., Franklin, J.F. and Thomas, T.B. (1988). Coarse woody debris in
Douglas-Fir forests of Western Oregon and Washington. Ecology 69: 1689-
1702.
Stokland, J.N. (2001). The coarse woody debris profile: an archive of recent forest history and an important biodiversity indicator. Ecological Bulletins 49: 71-83.
16

Dead wood in tall, wet E. obliqua forest
15
16
17
18
19
20
21
22
23
24
25
26
27
5
6
7
8
1
2
3
4
9
10
11
12
13
14
Stokland, J.N. and Sippola, A.-L. (2004). Monitoring protocol for wood-inhabiting fungi in the Alberta Biodiversity Monitoring Program . Alberta Biodiversity
Monitoring Program, Version 2.9.
Tedersoo, L., Kõljalg, U., Hallenberg, N. and Larsson, K.-H. (2003). Fine scale distribution of ectomycorrhizal fungi and roots across substrate layers including coarse woody debris in a mixed forest. New Phytologist 159: 153-65.
Turner, P., Airey, C. and Grove, S. J. (2007). Wildfire chronosequence project establishment report . Forestry Tasmania, Division of Forest Research and
Development.
Turner, P.A.M., Balmer, J. and Kirkpatrick, J.B. (2009). Stand replacing wildfire?
The incidence of multi-aged and even-aged Eucalyptus regnans and E. obliqua forests in southern Tasmania. Forest Ecology and Management 258: 366-75.
Wells, P. and Hickey J. (1999). Wet sclerophyll, mixed and swamp forest. In:
Vegetation of Tasmania (eds J.B. Reid, R.S. Hill, M.J. Brown and M.J.
Hovenden), pp. 224-43, Australian Biological Resources Study, Canberra.
Woldendorp, G. and Keenan, R.J. (2005). Coarse woody debris in Australian forest ecosystems: A review. Austral Ecology 30: 834-43.
Woldendorp, G., Keenan,R.J. and Ryan, M.F. (2002). Coarse Woody Debris in
Australian Ecosystems . Bureau of Rural Sciences, Department of Agriculture,
Fisheries and Forestry, Australia. 75pp.
Wu, J.-B., Guan, D.-X., IIan, S.-J., Zhang, M. and Jin C.-J. (2005). Ecological functions of coarse woody debris in forest ecosystem. Journal of Forest
Research 16 , 247-52.
Yee, M. (2005). Ecology of saproxylic beetles native to Tasmanian wet eucalypt forests.
Unpublished Ph.D. Thesis. School of Agricultural Science, University of Tasmania, Hobart.
17

Dead wood in tall, wet E. obliqua forest
1
2
3
Table 1.
CWD and stag decay classification to accommodate E. obliqua and other tree species. (a) CWD, (b) Stags.
(a) CWD
Decay class Characteristics for classifying CWD
1 CWD freshly downed, entire, cylindrical, wood hard, sound, bark intact, no sign of internal decay or external macrofungal fruit bodies
1.5 Wood has been lying on the ground for some time, cracks appearing in bark.
2 CWD remaining solid, losing some bark, some Basidiomycota fruit bodies appearing, such as Mycena spp., and some Ascomycota such as Hypoxylon spp. and Hypocrea aff . megalosulphurea , bryophyte cover sparse.
2.5 CWD with many macrofungal fruit bodies, but exhibiting no sign of softening. Category included to accommodate Pomaderris apetala .
3 CWD retaining round shape, bark may be present, bryophyte cover present but variable, some degree of heart rot, still quite firm on the outer surface, many external macrofungal fruit bodies present in season.
3.5 CWD beginning to flatten, becoming softer, often with seedling trees, wood-inhabiting macrofungal genera such as Gymnopilus , Galerina , ectomycorrhizal genera such as Laccaria , Cortinarius and Russula associated with the seedling trees (Tedersoo et al . 2003) and
4 corticioids (resupinate fungi) being commonplace, bryophyte cover substantial. Roots from nursery trees making their first appearance.
CWD half its original diameter, often with only the sides remaining but still recognizable as a log or a log that may be prolifically interspersed with roots from nursery trees of considerable size.
4.5 CWD disintegrating into splinters and losing outline.
5 CWD reduced to a pile of humus, still with very small wood fragments present, outline just visible, mound-like appearance or a ‘cage’ of roots from a nursery log with some woody humus remaining.
4
Decay class
1
2
3
4
5
(b) Stags
Characteristics for classifying stags
Stag limbs and branches all present; 100% bark present.
Stag has some loss of limbs and bark but is sound at base.
Stag distinctly rotten at base; in E . obliqua the bark can still be intact at this stage.
Stag still standing with outer bark intact but obviously very decayed inside. This category is for Nothofagus stags.
Stag reduced to a thin central core, no outer wood but still standing.
This category is for Nothofagus stags.
5
18

Dead wood in tall, wet E. obliqua forest
3
4
1
2
5
Table 2.
Live tree and woody shrub species in each plot. Numbers in brackets after each species are respectively:
1
Contribution of species to total basal area of plot, expressed as a percentage of the total;
2
Number of living stems;
3
Maximum
‘1934’:
Eucalyptus obliqua (85.7%;40;230)
Nothofagus cunninghamii (6.0%;167;35)
Monotoca glauca (3.5%;174;21)
Acacia melanoxylon (2.2%;18;32.5)
Acacia dealbata (0.8%;5;31.5)
Nematolepis squamea (0.4%;2;25.5)
Phyllocladus aspleniifolius (0.4%;22;15.5)
Tasmannia lanceolata (0.3%;29;12.5)
Pittosporum bicolor (0.2%;2;23)
Cyathodes glauca (0.1%;32;8.5)
Eucryphia lucida (0.1%;3;18)
Anopterus glandulosus (0.1%;33;11.5)
Atherosperma moschatum (<0.1%;3;3)
Coprosma quadrifida (<0.1%;7;2.5)
Pimelea drupacea (<0.1%;7;1.5)
Total 544 living stems diameter, cm. Species are listed in decreasing order of their contribution to the total basal area of the plot.
‘Old growth’:
Eucalyptus obliqua (54.0% 1 ;2 2 ;350 3 )
Nothofagus cunninghamii (24.9%;189;110)
Atherosperma moschatum (17.7%;216;80)
Acacia melanoxylon (3.0%;5;80)
Eucryphia lucida (0.4%;8;35)
Anopterus glandulosus (<0.1%;4;5)
Coprosma quadrifida (<0.1%;5;6)
Phyllocladus aspleniifolius (<0.1%;1;1)
Pimelea drupacea (<0.1%;11;1.5)
Tasmannia lanceolada (<0.1%;1;1.5)
Total 442 living stems
‘1898’:
Eucalyptus obliqua (66.3%;19;300)
Pomaderris apetala (11.5%;193;27.5)
Nothofagus cunninghamii (9.3%;86;65)
Atherosperma moschatum (5.7%;50;50)
Acacia melanoxylon (3.6%;10;60)
Olearia argophylla (3.3%;37;34)
Phyllocladus aspleniifolius (0.2%;4;18.5)
Anopterus glandulosus (<0.1%;2;1.5)
Coprosma quadrifida (<0.1%;5;4.3)
Cyathodes glauca (<0.1%;1;1)
Monotoca glauca (<0.1%;2;7.5)
Pimelea drupacea (<0.1%;8;1)
Pittosporum bicolor (<0.1%;1;7)
Total 418 living stems
‘1898/1934’:
Eucalyptus obliqua (73.9%;39;250)
Pomaderris apetala (17.2%;832;25.5)
Acacia melanoxylon (3.5%;13;70)
Eucryphia lucida (2.2%;78;36)
Nothofagus cunninghamii (1.6%;155;33)
Atherosperma moschatum (0.5%;18;17,5)
Phyllocladus aspleniifolius (0.4%;54;27)
Olearia argophylla (0.3%;17;9.5)
Acacia verticillata (0.1%;3;12)
Anopterus glandulosus (0.1%;29;10)
Coprosma quadrifida (0.1%;28;5)
Cyathodes glauca (0.1%;29;7)
Aristotelia peduncularis (<0.1%;4;3)
Monotoca glauca (<0.1%;2;1)
Pimelea drupacea (<0.1%;20;1)
Pittosporum bicolor (<0.1%;1;3)
Tasmannia lanceolata (<0.1%;11;6.5)
Total 1333 living stems
6
19

Dead wood in tall, wet E. obliqua forest
1
2
3
4
5
6
Table 3. Subplot minimum, median and maximum volume of CWD (in the 25 subplots) for each of the plots ‘Old growth’, ‘1898’, ‘1934’ and ‘1898/1934’, and the total volume and total volume/ha for the same plots.
Min volume m 3
Median volume m m 3 m 3 m 3 mm 3
Max volume m 3
3
Total volume m 3
Total volume m 3 /ha
(((m3ha?, m 3 /ha
‘OG’
0.627
3.715
34.359
209.462
837.8
‘1898’
0.429
13.534
45.939
361.717
1446.9
Plot
‘1934’
0.261
9.680
37.065
272.729
1090.9
‘1898/1934’
0.263
6.707
21.098
175.302
701.2
20

Dead wood in tall, wet E. obliqua forest
1
2
3
4
5
Table 4 . The total number of stags, number of E. obliqua stags, average stag diameter and total stag volume in the four plots (‘1898’ is divided into its two component parts).
Total no. of stags
Volume of stags (m 3 /ha)
No. of E. obliqua stags
No. of non-eucalypt stags
Ave. stag diam., eucalypts
Ave. stag diam., non-eucs
Plot
‘OG’ ‘1898R’ ‘1898P’ ‘1934’ ‘1898/1934’
33
226
1
32
150
45.2
7
4.4
1
2
2.5
20.5
3
42.3
0
7
–
47.6
34
192
16
18
34.7
23.0
20
176
12
8
23.2
90.8
21

Dead wood in tall, wet E. obliqua forest
6
7
8
9
10
11
12
13
1
2
3
4
5
Table 5 . Proportion of total basal area contributed by the woody perennial genera in the plots of the present study and in other similar plots (having the same stand type with the same southerly aspect) located in the tall, wet E. obliqua forests of southern
Tasmania.
Genus
Acacia
Atherosperma
Eucalyptus
Eucryphia
Leptospermum
Monotoca
Nematolepis
Nothofagus
Olearia
Phyllocladus
Pomaderris
Tasmannia
‘1966S’
4.6
1.5
41
0.7
93.1
0.1
‘1934S’
0.6
0.2
5.5
0.2
0.1
73
3.7
2.2
87.1
0.4
<0.1
‘1934’
3.5
0.4
6.0
0.4
0.3
72
3.0
<0.1
85.7
0.1
Plot name; approximate plot age, years
‘1898/1934’ ‘1898S’ ‘1898P’
<0.1
1.6
0.3
0.4
17.2
<0.1
72
3.6
0.5
73.9
2.2
73
13.5
1.7
62.8
0.1
20.8
1.0
72
1.5
80.5
<0.1
1.0
0.1
16.7
‘1898R’
0.2
20.7
1.3
0.1
108
10.0
11.9
55.8
‘OG’
>200
3.0
17.7
54.0
0.4
24.9
<0.1
<0.1
‘OGS’
>200
4.9
15.8
54.9
24.2
<0.1
Notes: plots ‘1934’, ‘1898/1934’ and ‘OG’ are those of the present study; ‘1898P’ and ‘1898R’ are partitions of ‘1898’ of the present study, the former being based upon subplots where Pomaderris predominates the understorey, and the latter being based upon subplots where rainforest species are predominant; ‘1966S’, ‘1934S’, ‘1898S’ and ‘OGS’ are plots used in the chronosequence study of Turner et al . (2007), with the basal areas determined by Scanlan et al . (2010). Blank entries indicate the genus was absent from that plot or contributed less than 0.05% to the column total, which would be 100% were it not for this fact. The age of ‘1898S’ and of ‘1898P’ reflects the fact that their most recent fire occurred in 1934.
22

Dead wood in tall, wet E. obliqua forest
1
2
3
4
5
Fig. 1.
Warra LTER site location (from http:// www.warra.com
) and the position of the plots used in this study along the ‘Bird Track’.
23

Dead wood in tall, wet E. obliqua forest
1 (a) (b)
2
3
4
5
6
7
8
(c) (d)
Fig. 2. Maps of tree species with stems

10 cm diameter, Acacia (gold diamonds),
Atherosperma (pink squares), Eucalyptus (green triangles), Monotoca (violet squares), Nothofagus (orange triangles), Pomaderris (blue squares). Other genera formed only a minor component of the live vegetation and are omitted from the figure. (a) ‘Old growth’, (b) ‘1898’, (c) ‘1934’, (d) ‘1898/1934’.
24

1
Dead wood in tall, wet E. obliqua forest
2 (a) (b)
6
7
3
4
5
(c) (d)
Fig. 3. CWD maps for the four plots at the Warra Bird Track. (a) ‘Old growth’, (b)
‘1898’, (c) ‘1934’, (d) ‘1898/1934’.
25

Dead wood in tall, wet E. obliqua forest
1
40
35
30
25
20
15
10
All plots
'Old growth'
'1898'
'1934'
'1898/1934'
5
0
1 1.5
2 2.5
3
Decay class
3.5
4 4.5
5
2
3
4
5
Fig. 4.
Percentage of pieces of CWD in each decay class for each plot separately, and for all plots combined. The percentages add up to 100% within a plot.
26

Dead wood in tall, wet E. obliqua forest
1
2
3
4
5
100
90
80
70
60
All plots
'Old growth'
'1898'
'1934'
'1898/1934'
50
40
30
20
10
0
1 1.5
2 2.5
3 3.5
4 4.5
5
Decay class
Fig. 5.
Average percent bryophyte cover in each decay class for each plot separately and for all plots combined.
27

1
Dead wood in tall, wet E. obliqua forest
25
20
CWD
15
10
5
0
(a)
Diam C
5
4.5
4
3
3.5
2
2.5
1
1.5
DecayC
25
20
CWD
15
10
5
0
(b)
Diam C
5
4
4.5
1
1.5
2
2.5
3
3.5
DecayC
8
9
10
6
7
2
3
4
5
25
20
CWD
15
10
5
0
5
4
4.5
1
1.5
2
2.5
3
3.5
DecayC
25
20
CWD
15
10
5
0
5
2
1
1.5
2.5
4
3
3.5
DecayC
4.5
(c)
Diam C
(d)
Diam C
Fig. 6. Profile plot of percentages of pieces of CWD versus decay class (DecayC) and diameter class (DiamC, cm) for each of the four plots: (a) ‘Old growth’, (b)
‘1898’, (c) ‘1934’, (d) ‘1898/1934’. See Table 1a for an explanation of the decay classes.
28

Dead wood in tall, wet E. obliqua forest
20
18
16
14
Frequency
12
10
8
6
4
2
0
20
40
60
80
100
120
Diameter, cm
140
160
180
200
220
240
'1934'
'Old growth'
'1898/1934'
'1898R'
'1898P'
1
2
3
4
Fig. 7. Diameter distribution of stags in the four plots at the Warra Bird Track
(‘1898’ is divivded into its two component parts).
29

35
30
25
20
Frequency
15
10
5
0
Dead wood in tall, wet E. obliqua forest
Bryophyte
% cover
1
2
3
4
Fig. 8. Stag numbers in bryophyte percentage cover classes (‘1898’ is divivded into its two component parts).
30

Dead wood in tall, wet E. obliqua forest
1
1600
1400
1200
1000
800
600
400
200
Woldendorp et al. (2002)
Sohn (2007)
Gates (2009)
2
3
4
5
6
7
8
0
0 25 50 75 100 125 150 175 200 225 250 275
Years since wildfire
Fig. 9. Volume of CWD (m
3
/ha) as a function of time since wildfire, comparing data from the present study with two other studies that used plots having the same stand type and the same southerly aspect in the tall, wet E. obliqua forests of southern
Tasmania.
31