Warra Site, 5-minute Point-Count Bird Survey Data, Southern Forests Experimental Forest Landscape, 2010-2011 

The record contains information on the five-minute, point-count, bird survey data conducted in the Southern Forests Experimental Forest Landscape (SFEFL), Warra Site for the period between 2010 to 2011. Data such as age class of study plots, Landscape Disturbance Index (LDI), bird survey period, bird species identification details, observation distance, number of individuals, height and direction of observation at a minimum distance of 25 meters are provided. 

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. The project was funded by the Forest and Wood Products Australia Project PNC 142-0809: Persistence of mature forest biodiversity elements in a production forest landscape managed under a Regional Forest Agreement 

The data set was collected as part of a broader study examining the responses of bird, plant and beetle species to the intensity of disturbance in the landscape surrounding plots of either mature tall, wet eucalypt forest or of 25-50 year-old silvicultural regeneration following clear-fell harvesting. The study tested the null hypotheses that species persisting in patches of mature forest or re-colonising patches of silvicultural regeneration would be insensitive to the intensity of disturbance in the surrounding landscape. The study was done in 2010-2011 in the SFEFL – a 32 x 35 area anchored on the Warra Long-Term Ecological Research site and extending eastwards to the estuary of the Huon River. The surveys of the three taxonomic groups were done in fifty-six, 50 x 50 m plots – 28 each in mature eucalypt forest and silvicultural regeneration. Those plots were located to sample the measured range of average disturbance intensity (based on the mix of vegetation classes) at three scales – 0.5, 1 and 2 km, in the circular landscapes surrounding the plots. The bird data set was based on 5-minute point-count surveys of each 50 x 50 m plot-corner done on sixteen separated visits between February 2010 and March 2011. 

1) Calculation of landscape disturbance intensity: A vegetation map of the SFEFL representing the 2009 distribution of nine broad vegetation classes was produced at 1:20,000 scale from forest-types interpreted from aerial photography (Stone, 1998, Tasforests, 10: 15-32) acquired in the 2000s (Forestry Tasmania, Forest Class 2005 mapping), and converted to 50 m pixel raster format using the open-source GIS software SAGA©. Each vegetation class was assigned a disturbance rating between 1 (least disturbed) for rainforest and 10 (most disturbed) for agricultural land, and each pixel was assigned the disturbance rating corresponding to its vegetation class. Using a moving-window algorithm, the Landscape Disturbance Index (LDI) for each pixel was then calculated by averaging the disturbance rating of all pixels within a given radius of that pixel. Separate LDIs were calculated for radii of 500 m, 1 km and 2 km, to give separate LDI values for each pixel at these three landscape-scales. All LDI values were rounded down to the nearest integer and the raster maps of integer LDIs at each of the three spatial scales were overlaid to identify pixels with the same integer LDI at each of the three spatial scales to produce a “three-scale-consistent LDI” map of the SFEFL.

2) Selection of sample plots: The three-scale consistent LDI map of the SFEFL was overlain with the mapped extent of two forest age-classes: (i) mature eucalypt forest > 110 years old and never harvested; and (ii) older (25–50 year-old) wet eucalypt forest that had been silviculturally regenerated after clear-fell harvesting. Patches of these age-classes coinciding with pixels that were three-scale consistent for LDI class were identified. All locations of scale-consistent patches of mature eucalypt forest and older silvicultural regeneration were prioritized for field validation according to the following criteria:

• Low altitude (below 600 m);
• Forest dominated by E.obliqua, E.regnans, or a mixture of these species;
• At least 75 m from a road, vehicle track, or edge of a strongly contrasting vegetation type;
• Forest patch of an age-class was at least 150 m wide at a point where a plot could be located.

The final experimental design comprised seven replicate 50 x 50 m square plots of mature eucalypt forest (MAT plots) at each of four LDI classes (4–7), and seven replicate plots of older silvicultural regeneration (SILV plots) at each of four LDI classes (5–8). Mature forest within highly disturbed areas with LDI class 8 was rare, as was older silvicultural regeneration within relatively undisturbed areas with LDI class 4, so these combinations were not sampled.

3) Bird surveys: Each of the 56 plots was visited on 16 occasions during the spring to early autumn in 2009–10 and 2010–11 by the same observer (A.Hingston). The plots were surveyed using the method described in Hingston and Grove (2010, Forest Ecology & Management, 259: 459-468). Surveys were carried out between sunrise and sunset on days with fine weather and little wind (<30 km/h). Plots were visited in a different sequence on each of the 16 survey cycles, to ensure that the observations across all plots had comparable seasonal and diurnal distributions: each plot was surveyed four times before 10:00 h, four times between 10:00 h and 13:00 h, four times between 13:00 h and 16:00 h, and four times after 16:00 h. No plot was surveyed more than once per day.

Five-minute point counts were conducted at each corner of each 50 x 50 m plot, so that each plot sampling occasion comprised a twenty-minute period of observation. The identities of all birds seen or heard within 25 m horizontally of each plot corner were recorded as a presence record (because of difficulty determining the numbers of individuals in dense undergrowth and tall trees). Occurrence thus refers to the total number of separate visits on which a given species was recorded from at least one of the four corners of the plot (maximum possible = 16 per plot), treating the corners as non-independent sub-samples. If a bird was flushed from within 25 m of a plot corner as the observer approached that point, the bird was included in the data-set for that point and the 5-min survey commenced immediately. Birds that were flying more than 25 m above the vegetation were excluded from analyses unless they were aerial feeders or raptors.