the remediation works. These improvements had occurred before monitoring of Sites 1-4 under this project began. Figures 8 and 9 display vegetation condition scores for years 3-5 for Site 1 and years 2-4 for Sites 2, 3 and 4, respectively. When monitoring began most sites were already in B or even A condition, having improved from an estimated D or C condition( 15-35 / 100). The CSIRO-based research at Site 1 captured vegetation data in the first two years, before this project started, and results support the observation that large improvements in land condition and related vegetation metrics were achieved in the first 1-2 years. Bartley et al., 2025( p. 26) conclude that at Site 1“ statistically significant increases in vegetation metrics( vegetation cover and biomass) occurred in the first year and this can be attributed to the remediation works. In the following years metrics tended to climb more slowly or plateau till 2022. In 2023 Site 1 biomass declined to near pretreatment levels”( Bartley et al., 2025, p. 35). This is discussed further below.
Site 5 was given the lower‘ partly stable’ status in December 2024. This is mainly because neither of the gully areas( northern section and southern section) were given a comprehensive treatment. However, the southern section provides an example of successful low impact facilitated revegetation. This is described in more detail below under key question 3, Light remediation works. While slower to‘ get going’, the vegetation that has established is suited to harsh conditions and may prove to be more long-lasting than the treatments implemented at Sites 1, 2, 3 and 4. In contrast, the treatments in the catchments at Site 5 greatly increased vegetation cover, reaching A condition in three years.
What were the primary causes of reduced success / soil stability at the sites and how can these be remedied?
The main drivers of reduced stability at the largescale sites were:
• patches of erosion( mainly sheet, rill and tunnel erosion);
• reduced ground cover( e. g. < 70 per cent); and
• Lack of biodiversity( dominance of Desmanthus sp. or Indian couch, Bothriochloa pertusa)( Table 3).
Sheet and rill erosion is primarily caused by runoff over bare ground, particularly during the wet season( December- March). As ground cover increases, this erosion will typically reduce. LCATs are designed to capture resilient ground cover, primarily the amount of perennial grass tussocks so changes will be relatively slow as perennial species replace annuals, or the reverse for declining land condition. In contrast, ground cover and biomass can change quickly in response to seasons and grazing pressure. Bartley et al., 2025 measured ground cover and biomass at Sites 1 and 2. For Site 2, the‘ Stable’ site, bare areas reduced markedly in the first two years as ground cover increased, achieving > 70 per cent cover in two years( p. 39). At Site 1, ground cover and biomass increased significantly( compared to the control site) in the first year following treatment. Subsequent years saw declines in biomass( although not statistically significant) and this was attributed to climate, rainfall and grazing( Bartley et al. 2025, p. 35).“ The increase in grazing pressure at the treatment site after year 1, reflected by higher defoliation rates post-treatment, suggests potential co-benefits for grazing productivity. Nevertheless, careful grazing management will be essential to sustain vegetation cover and biomass levels”( Bartley et al., 2025, p. 35).
At Site 1, cattle tracks were noted by the construction specialists in their regular monitoring; these were repaired during maintenance to prevent erosion developing. The grazes at Site 1 occurred in conjunction with the larger paddock and tended to be longer than graze periods at the other large-scale sites. The grazing regime was typically 100 weaners twice for two months in the dry season. The smaller areas of Sites 2, 3 and 4 led to shorter graze periods which reduced the likelihood of cattle pads which can rapidly develop into rill and gullies. Similarly, Landloch Pty Ltd( April, 2024) in their report on the performance of 11 large-scale gully sites in the BBB, noted that five were impacted by cattle or feral animals despite fences being in place.
At Sites 1, 2, 3 and 4, Desmanthus species was planted as part of the pasture seed mix after the earthworks and along with ameliorants such as gypsum and fertiliser. At the end of the monitoring project, Desmanthus sp.( most likely Progardes) dominated Sites 1, 3 and 4; reducing land condition scores. The contracted engineer for Sites 3 & 4 has also noticed the dominance of D.“ Progardes” over time at sites and the use of less
NQ DRY TROPICS 41