Identifying and reversing ecological barriers to successful farmland revegetation specific to tubestock planting and direct seeding in northern New South Wales

Abstract

Revegetation in agricultural regions across the globe has intensified over past decades in an effort to reverse widespread land degradation and conserve natural ecosystems and the biodiversity they contain. Biodiversity is essential for the physical, economic, social and cultural dimensions of human well-being, but agricultural intensification has resulted in the loss of millions of hectares of forests and natural vegetation globally. In Australia, a brief, but intense history of land clearing has resulted in the loss of 50% of forest ecosystems, with over 80% of eucalypt-dominated woodlands and forests having been altered by human endeavour. The situation on the Northern Tablelands of New South Wales reflects land clearing practices throughout the country with an estimated loss of tree cover of 50% to date. Although land clearing has eased in the past decade, tree decline continues due to recurrent episodes of rural dieback. To address this problem, substantial efforts have been made by revegetation organisations, practitioners and landholders to re-establish native trees in the region, but plantings often fail. This research was conducted to identify and reverse ecological barriers preventing the success of revegetation in temperate upland pastures. The work focused on tubestock plantings and direct seeding. <br/> The first study was conducted to determine if existing native shelterbelts can be evaluated in terms of survival and growth to identify the environmental stresses influencing planted eucalypt establishment and growth on the Northern Tablelands. Most ‘on-ground’ revegetation is designed and implemented with no thought given to follow-up scientific monitoring. Monitoring is important not only to justify the large amounts of public funding directed into revegetation activities, but also because it demonstrates whether targets have been achieved, and provides opportunities to learn from and improve upon past failures. Six-year old shelterbelt plantings consisting of Eucalyptus nitens, E. pauciflora and E. viminalis were examined to identify potential biotic and abiotic stresses influencing tree performance. Topographic position, altitude, slope, temperature, soil type, soil moisture and weed control were measured and modelled in relation to tree survival and growth (height). The information theoretic approach was used to select the best-fitting model from a set of competing models. Poor weed control and subzero temperatures were identified as the predominant stresses affecting eucalypt survival during the monitoring period. Subzero temperatures also significantly influenced tree growth. <br/> Given these results, the second study compared the performance of five native tree and shrub species grown in tall Corflute® tree guards and milk cartons at three landscape positions (lower slope, mid slope and upper slope) in an open temperate pasture. Seedlings in tall guards survived better than seedlings in milk cartons at mid and upper-slope landscape positions. Height was also greater for seedlings in tall guards than milk cartons at all landscape positions. Eucalypts in particular benefited from tall guards, with height growth up to three times greater than in milk cartons. Tall guards increased the temperature surrounding seedlings inside the guards, extending the growing period. <br/> Next, the efficacy of direct seeding as a revegetation technique was investigated. Prior to conducting this investigation, three trial sites were established and monitored for 3–6 months at Bingara, Ben Lomond and Invergowrie. Recruitment and subsequent establishment was so poor that the trials were considered a failure. This study compared the effects of three sowing methods (KB seeder, modified Chatfield planter and hand sowing) and three bulking materials (rice, chicken crumble and smoked vermiculite) on the recruitment of direct-seeded acacias and eucalypts. Recruitment was highest with the KB seeder followed by the Chatfield seeder and hand-sown methods. There were no significant differences in recruitment among bulking materials. Eucalypt recruitment was low compared to the recruitment of acacias. Recruitment peaked in mid May (8 weeks post-sowing) for acacias and in early July (15 weeks post-sowing) for eucalypts, but declined markedly for both genera during the reminder of the study. Some seedling losses were incurred following the first heavy frost, but most were attributed to an invasion of rat’s tail fescue (Vulpia myuros) in late winter, and waterlogging in the lower areas of the site due to above-average rainfall between May and August. <br/> Based on the results of the previous study two subsequent investigations were designed and implemented. To address the problem of weed invasion in direct-seeded revegetation, the effect of eight herbicide oversprays on the survival of 11 native tree and shrub species was examined. Seedling survival was assessed at 1, 2, 7 and 8 weeks post-spraying. Survival was greatest in seedlings treated with imazethapyr and isoxaflutole, and least in seedlings treated with diflufenican and glyphosate. There were also significant differences in survival between species, with Dodonaea viscosa, Acacia pendula and Senna artemisioides exhibiting the highest tolerance to the greatest range of herbicides, and Atriplex nummularia, Casuarina cristata and Einadia nutans exhibiting the lowest tolerance. Species varied in their tolerance of different herbicides, due to the selective nature of the different modes of action of the active ingredients and their differential uptake, translocation and metabolism between species. <br/> To address the problem of poor eucalypt recruitment, the effects of three seed-coating treatments (coated seed, seed coated with MycoApply® and uncoated seed), four watering regimes (30 mL per day, 30 mL per 3 days, 30 mL per 5 days and no water), and two seed-sowing methods (surface-sown vs buried beneath a 5-mm vermiculite layer) on eucalypt germination and early establishment were examined. Coating the seed with microbial inoculants as well as daily watering significantly increased germination. The interaction between sowing method nested within coating treatment and watering regime was significant. Soil moisture was essential for eucalypt germination and MycoApply® appeared to enhance water capture during the critical early post-germination phase when desiccation was most likely. When soil moisture was limited, seed burial beneath vermiculite was important because it increased seed-soil contact, providing better access to soil moisture, but only in seed coated with microbial inoculants. Seed-coating treatments did not benefit seedling growth (height) because soil moisture was not limiting. <br/> The final study examined the effects of four weed-control treatments (scalping, glyphosate, sugar and glyphosate, and sawdust and glyphosate) on the recruitment of native grasses and weed suppression. Recruitment of native grass was significantly higher in scalped plots compared to other treatments. The most effective weed- control treatments were scalping and the combinations of sugar and sawdust with glyphosate. Scalping and carbon (sugar and sawdust) addition controlled weeds by depleting weed seed banks, and alleviating soil nutrient enrichment. In combination with glyphosate application, which removed above-ground biomass, these techniques were effective strategies for combatting competitive invasive weeds. <br/> A summary of the main findings, study limitations and recommendations for future research were presented in the final chapter.