This data contains vegetation cover, ground cover, tree density and stand basal area data across a multi-century time-since-fire sequence derived from growth ring-size relationships in fire-sensitive Eucalyptus salubris woodlands.
Credit
We at TERN acknowledge the Traditional Owners and Custodians throughout Australia, New Zealand and all nations. We honour their profound connections to land, water, biodiversity and culture and pay our respects to their Elders past, present and emerging.
Purpose
Recurrent fire is a dominant disturbance in Mediterranean-climate landscapes. In infrequently-burnt communities, vegetation structure, habitat features and fuel availability can change over time-scales much longer than can be measured using contemporary remote-sensing approaches, creating challenges for conservation and fire management. The Great Western Woodlands (GWW) region of south-western Australia supports the world’s largest remaining area of Mediterranean-climate woodland, which in mosaic with mallee, shrublands and salt lakes cover an area of 160 000 km2. Eucalyptus woodlands in this region are typically fire-sensitive, and fire return intervals recorded over recent decades have been much shorter than the long-term average. This has led to considerable conservation concern regarding the loss of mature woodlands, and has highlighted a need to better understand how plant communities change with time since fire.
Lineage
Vegetation and ground surface cover sampling
At each site, two 50 m transects were established along north and west sides of the 50 x 50 m plot and a 70 m transect was placed diagonally through the centre starting in the north-west corner. Along these transects, 50 vertical point placements were made with a 12.5 mm diameter pole, extendable to 3 m in height, marked in 10 cm increments. Intercepts were sampled at 3 m intervals; 16 along each side and 18 on the diagonal. At each pole placement we recorded the presence or absence of an intercept between the pole and any vegetation in the following height classes: 0-12, 12-25 and 50-100 cm, and 1-2, 2-4, 4-10 and > 10 m. The presence/absence of intercepts with vegetation greater than 4 m in height was visually estimated, with the height of any intercepted vegetation checked using a hypsometer (Nikon Forestry 550). Intercepts with live and dead vegetation (entirely dead plants or dead limbs, but not individual dead leaves on otherwise live limbs) were recorded separately but note that the data provided here is total vegetation cover including intercepts with both live and dead vegetation. Point placements which did not intercept any vegetation are termed ‘foliar gaps’. At each of the 50 point placements, ground cover was recorded by placing the pole 1 m perpendicularly to each side of the transect. At each point (n = 100) ground cover was recorded as being either ‘ground fuel’, ‘bare’ (including rock) or ‘cryptogam’ based on the dominant cover type (if multiple types were present) under the pole intercept. All dead vegetation on the ground surface was classed as ground fuel, so includes shed leaves, twigs, buds, fruits, bark, branches and logs. Cryptogam cover was based on a visual field assessment of the presence or absence of soil crust organisms, including moss, lichens and cyanobacteria. In cases where ground cover placements intercepted live vegetation, the ground surface under foliage was recorded.Intercept counts were converted to a single value for the proportion of intercepts per layer/ground cover class per site.
Tree density and size sampling
Tree size data was collected by sampling 16 trees by use of a modified version of the point-centred quarter method. Diameter at D10 of trunks was measured along with the distance from the corner of the nearest tree in each of the four compass quadrants radiating from the four corners of each plot. D10 was used rather than breast height owing to the low, multiple-branching habit of Eucalyptus salubris. From these measurements, tree density per site and mean cross-sectional area per tree was calculated. Total basal area per site was calculated by multiplying mean cross-sectional area per tree by tree density.
