The detection of. biotic changes in the Tekapo riverbed after habitat restoration ......................................................................................................................... ' A thesis submitted in partial fulfilment of the degree of Master of Science in Environmental Science at the University of Canterbury, Christchurch, New Zealand by Lisa ~inclair University of Canterbury 1995 dedicated t6 mum and Mf Abstract The impact of bulldozing on terrestrial invertebrates in the Mackenzie Basin (South Canterbury) was investigated. Willow trees (Salb; spp) had been bulldozed from the Tekapo delta to recreate braided riverbed habitat for riverine birds. Poor survival of these birds is attributed to predation by cats and ferrets. Predators supplement their predominantly rabbit diet with birds, lizards and invertebrates when rabbits are scarce. An increase in predation pressure on riverine birds could result if invertebrate prey decreased in availability following bulldozing. Invertebrates were sampled with pitfall traps across the Ohau and Tekapo deltas, before and after willow clearance. Eighteen sites represented different time periods since willow clearance. A comprehensive vegetation-environmental variable survey was also made. Data were analysed with the simple statistical measures of species richness, evenness and diversity, and the more complex methods of cluster analysis and ordination to determine if distinct invertebrate and plant communities existed. Simple statistics were inadequate to distinguish between 'Shingle', 'Willow', 'bulldozed/disturbed' and 'braided riverbed' habitats. Community classification and ordination techniques were more satisfactory. The inclusion of plant species data in the invertebrate ordination accounted for twice the variation explained by the macro-environmental data and by the invertebrate ordination alone. Improvement in habitat for riverine birds was measured by the appearance of braided riverbed taxa in the bulldozed areas. Very few 'braided riverbed' invertebrate or plant taxa appeared in the bulldozed areas even two years after willow clearance. The invertebrate prey of mammalian predators were predominantly weta but these contributed little to the diet by weight. Seven undescribed species of invertebrate were identified as having conservation value. These included a new species of Hemiandrus (a ground weta), and Prodontria (a chafer beetle). Table of Contents. List of Tables List of Figures List of Appendices Abstract Chapter 1. Introduction page Chapter 2. Study site page 9 Chapter 3. Methods page Chapter 4. Results page 36 Chapter 5. Discussion and recommendations page 115 Acknowledgements References Appendices List of Tables. Table 4.1.1. Species richness scores (S) and Chi-square significance 39 tests for sites: Inv = Invertebrate; Plt = Plant. Table 4.1.2. Invertebrate, plant, and total richness scores per habitat. LJ 1 Chi-square levels of significance between samples (1993 vs 1994). Table 4.1.3. Invertebrate species richness scores per habitat in 1993; Ltl Chi-square levels of significance amongst richness scores. Table ""·.I.·""·A 1 A Invertebrate species richness scores per habitat in 1994; Chi-square levels of significance amongst richness scores. Table 4.1.5. Plant species richness scores per habitat in 1993; ~ :L Chi-square levels of significance amongst richness scores. Table 4.1.6. Plant species richness scores per habitat in 1994; 41... Chi-square levels of significance amongst richness scores. Table 4.2.1. Invertebrate species Evenness I equitability per site 49 ( Inverse of Two Dominant Species index). Table 4.2.2. Invertebrate species Evenness I equitibility per habitat, so measured by the inverse of the Two Dominant Species index. Table 4.3.1. Invertebrate species diversity per site measured by 5.:t Margalef' s Index. Table 4.3.2. Invertebrate species diversity per habitat measured by 5 6 Margalef' s Index. Table 4.4.1. Number of invertebrate tax.a per habitat from five orders. s 7 Table 4.5.1. Temporal stability (persistence) of invertebrate tax.a per '1 I habitat between sample dates. Table 4.5.2. Temporal stability (persistence) of plant taxa per habitat . '7 I between sample dates. Table 4.6.1. Environmental data for each habitat from two sample dates. <o4 Table 4.7.1. Invertebrate site by species two-way table. 70 Table 4. 7.2. Plant site by species two-way table. 13 Table 4.8.1. Cumulative percentage of variance explained by the 14- first few ordination axes. Table 4. 9.1. Invertebrates as indicators of improvement in habitat quaility (change towards the Ohau state), measured by the appearance of braided habitat species in the Experimental and Recovering treatment habitats. Table 4.9. la. 'Braided riverbed' invertebrate taxa represented in the q5 Experimental Shingle habitat after bulldozing (1994). Table 4.9.lb. 'Braided riverbed' invertebrate taxa represented in the q h Experimental Willow habitat before bulldozing (1993). Table 4.9.lc. 'Braided riverbed' invertebrate taxa represented in the q 7 Experimental Willow habitat after bulldozing (1994). Table 4.9.ld. 'Braided riverbed taxa' in the Recovering Treatment area qz (one and two years after willow clearance). Table 4.9.le. Some examples of ubiquitous invertebrate taxa present in q q the majority of habitats. Table 4.9.2. Plants as indicators of an improvement in habitat quality 101 (change towards the Ohau state), measured by the appearance of braided habitat species in the Experimental and Recovering Treatment habitats. Table 4.9.2a. 'Braided riverbed' plant taxon represented in the I o .1- Experimental Shingle habitat before bulldozing (1993). Table 4.9.2b. 'Braided riverbed' plant taxa represented in the 1 oi Experimental Shingle habitat after bulldozing (1994). Table 4.9.2c. 'Braided riverbed' plant taxa represented in the 101... Experimental Willow habitat before bulldozing (1993). Table 4.9.2d. 'Braided riverbed' plant taxa represented in the 103 Experimental Willow habitat after bulldozing. Table 4.9.2e. 'Braided riverbed' plant taxa represented in the 103 Recovering Treatment area two years after clearance. Table 4.9.2f. Some exmnples of ubiquitous plant taxa present in tl-ie \ oc+ majority of habitats. Table 4.10.1. Number of taxa lost from the experimental habitats \07 after bulldozing (plants and invertebrates). Table 4.1 O.la. Number of invertebrate taxa lost from the experimental 1 o 7 habitats after bulldozing. Table 4.10.lb. Invertebrate taxa lost from experimental habitats I o "b possibly as a result of bulldozing. Table 4.10.2. Plant taxa lost from experimental habitats possibly as a result of bulldozing. Table 4.11.1. Invertebrates consumed by ferrets and cats in the 110 Mackenzie basin. Table 4.11.2. Weta abundance per habitat. List of Ji'igures. Figure 2.1. Map of the study area. ,....., Figure 2.2. & 2.3. Aerial photographs of the study area I~ Ji'igure 2.4. Site 2 from the Ohau habitat in 1993. .2.0 Figure 2.5. Site 3 from the Ohau habitat in 1993. 21 Figure 2.6. Site 3 from the Ohau habitat (note flood damage). .2. I Figure 2.7. Site 10 from the Recovering Treatment area in 1993 21. (bulldozed 1992). Figure 2.8. Site 10 from the Recovering Treatment area in 1994. 11. J?igure 2.9. Site 15 from the Experimental Shingle habitat before it 1 ?> was bulldozed (1993). Figure 2.10. Site 15 from the Experimental Shingle habitat in 1994. 2?, Figure 2.11. Site 17 from the Experimental Willow habitat before it :v+ was bulldozed (1993). Figure 2.12. Site 17 from the Experimental Willow habitat in 1994. :v1- Figure 2.13. Site 21 from the Control Shingle habitat in 1993. ;z.5 Figure 2.14. Site 24 from the Control Willow habitat in 1993. ).~ Figure 3.1. Diagram of the petri dish template used in subsampling 30 the 'micro-matrix'. Figure 4.1.1. Number of invertebrate taxa per site for two sample dates. 43 Figure 4.1.2. Number of plant taxa per site for two sample dates. 43 Figure 4.1.3. Total number of taxa per site (plants plus invertebrates). 4tt Figure 4.1.4. Number of invertebrate taxa per habitat. '+ 5 figure 4.1.5. Number of plant taxa per habitat. 4-S Figure 4.1.6. Total number of taxa per habitat (plants plus invertebrates). 4t. Figure 4.2.1. Species evenness I equitibility of invertebrate taxa per site 51 measured by the inverse of the Two Dominant Species Index. Figure 4.2.2. Species evenness I equitibility of invertebrate taxa per habitat 51 measured by the inverse of the Two Dominant Species Index. Figure 4.3.1. Invertebrate species diversity per site, measured by 54- Margalef' s Index. Figure 4.3.2. Invertebrnte species diversity per habitat, measured by 5 4- Margalef' s Index. Figure 4.4.1 a. Relative number of invertebrates in the Ohau and 58 Recovering Treatment areas from two sample dates. Figure 4.4.1 b. Relative number of invertebrates in the Experimental 54 Shingle and Control Shingle habitats from two sample dates. Figure 4.4. lc Relative number of invertebrates in the Experimental Willow '70 and Control Willow habitats from two sample dates. Figure 4.5.1. Temporal stability (persistence) of plant and invertebrate b2. taxa per habitat between sample dates. Figure 4.6.1. Environmental data for each habitat in 1993. <0£i Figure 4.6.2. Environmental data for each habitat in 1994. bS Figure 4.7.1. Invertebrate site dendrogram from log10 transformed data. 6q Figure 4. 7 .2. Plant site dendrogram from loge transformed data. Figure 4.8.la. Invertebrate site ordination, axis 1 vs axis 2.Theoretical l'l site groups. Figure 4.8.1 b. Invertebrate site ordination axis 1 vs axis 2. 1993 - 1994 1 q connected. Figure 4.8. lc. Invertebrate site ordination, axis 1 vs axis 2. Invertebrate io classification grouping. Figure 4.8.ld. Invertebrate species ordination, axis 1 vs axis 2. Species "bi identification. Figure 4.8.le. Invertebrate site ordination, axis 1 vs axis 3.