Technical report: vegetation changes following the short-term exclusion of grazing in the Traprock region

[Summary]: This study is the continuation of earlier research examining the effects of vegetation management on woodland communities in the Traprock region and will contribute to an understanding of the ‘biodiversity potential’ of managed land units on properties (including open paddocks, scattered treed areas, and remnant woodland). A number of grazing exclosures were established on selected properties with the aim to monitor biodiversity changes over-time following the removal of grazing. Eighteen study sites across 10 properties in the Traprock region were selected for this study. The experimental design consisted of: 2 vegetation types (grassy box (Eucalyptus melliodora, E. microcarpa, or E. moluccana) woodland (L) and ironbark (E. crebra) /gum (E. dealbata) (U) woodland); 3 mature (overstorey) tree densities (20 trees/ha [high](H), and; 3 exclosures (full exclosure (1) [2.5m complete fence], partial exclosure (2) [1.5m three-wire fence], open (3) [corner makers]). Exclosure plots were erected in January/February 2005 and sampled in April 2005 (two months after exclosure establishment) and again in February 2006 (12 months after exclosure establishment). Within each 6 x 6 m exclosure plot, a central 2 x 2 m quadrat was sampled for plant species cover (determined subjectively). Above-ground vegetation (‘biomass’) was clipped in a 0.25m2 sample and dry weight determined. Overstorey cover and recruitment were determined within each 6 x 6 m exclosure plot. Stand structural characteristics, including foliage projective cover of distinct strata, and cover of litter, logs and rocks, and general habitat condition were also determined at each site. Patterns in floristic composition were determined using non-metric Multidimensional Scaling (nMDS). Two-way crossed Analysis of Similarity (ANOSIM) determined whether there were significant differences in floristic composition between exclosure types and mature tree density classes. nMDS was also used to assess patterns in cover data for growth forms. Two-way crossed Analysis of Variance (ANOVA) was used to determine if groups (density class, exclosure type and the interaction of density and exclosure) differed significantly for total, native, exotic, growth-form, perennial, and annual species richness and above-ground biomass. In addition, Spearman-rank correlations were performed to determine if biomass and estimates of ground cover covaried. A total of 151 plant species was recorded across all treatments with an average richness across treatments of 17 species per 4m2. No differences were apparent in overall plant composition (cover) between the exclosure treatments one year following grazing exclusion (ANOSIM, pr > 0.05). nMDS ordinations show no distinction between exclosure treatments, but patterns were observed in mature tree density treatments within vegetation types as found in previous research. Stand structure (cover of strata) showed much the same pattern as floristic composition. There were generally no detectable differences (p > 0.05) in plant above-ground biomass between exclosure treatments, although significant differences between tree density classes was indicated with a significantly higher plant biomass in low density treatments compared to high density for both vegetation types. Correlation results showed that estimates of grass cover provide a good indication of above-ground biomass (p