The restricted polynomially-tilted pairwise interaction (RPPI) distribution gives a flexible model for compositional data. It is particularly well-suited to situations where some of the marginal distributions of the components of a composition are concentrated near zero, possibly with right skewness. This article develops a method of tractable robust estimation for the model by combining two ideas. The first idea is to use score matching estimation after an additive log-ratio transformation. The resulting estimator is automatically insensitive to zeros in the data compositions. The second idea is to incorporate suitable weights in the estimating equations. The resulting estimator is additionally resistant to outliers. These properties are confirmed in simulation studies where we further also demonstrate that our new outlier-robust estimator is efficient in high concentration settings, even in the case when there is no model contamination. An example is given using microbiome data. A user-friendly R package accompanies the article.

The conservation of natural capital on farms is being increasingly recognised as essential for addressing global biodiversity decline. At the same time, there is growing interest in the potential for natural capital on farms that generates high public benefits, to also generate private benefits, potentially fostering greater adoption of conservation practices on farms. Despite this, empirical analysis of private benefits of natural capital on farms remains limited. Addressing this gap can help in identifying conservation opportunities and selecting cost-effective policy mechanisms to deliver public good biodiversity conservation outcomes on private farmland. We present results from a hedonic model on private benefits associated with native woodland and bird biodiversity on commercial farmland in a critically endangered temperate woodland ecosystem in southeast Australia. Using quantile regression, we found significant heterogeneity in private benefits associated with such natural capital across farmland value quantiles. These private benefits increased from lower to upper land value quantiles, conditional on a range of attributes of the farmland parcel. In some instances, private costs were associated with enhancement of native woodland and bird biodiversity on farms, above even a very low baseline. We discuss the implications of these findings for targeting of investment and selection of policy mechanisms.

Despite growing research on the associations between environmental conditions and mental health, no previous study has collected both quantitative indicators of farm-scale ecology and validated measures of farmer mental health. We assessed whether on-farm factors of engaging in natural resource management (NRM), the environmental state of the farm, or perceived financial challenges were associated with mental health and wellbeing in farmers. A cross-sectional survey was conducted in an established cohort of farmers for whom ecological monitoring data were collected, with data linkage to vegetation and biodiversity indicators. Participants were 63 farmers residing or working in the Box-gum grassy woodlands ecological region of eastern Australia. Primary outcome measures were symptoms of depression, anxiety, life satisfaction and wellbeing. Based on both zero-order correlations and regression models, financial struggle or financial worry were associated with poorer mental health and wellbeing outcomes, as was younger age. There were no direct associations of vegetation cover, avian biodiversity or NRM engagement with mental health or wellbeing. There remain considerable challenges in quantifying the effects of farm ecology on mental health outcomes. Further investigation of the potential social and financial benefits of natural asset management may be warranted.

In many farming landscapes, aquatic features, such as wetlands, creeks, and dams, provide water for stock and irrigation, while also acting as habitat for a range of plants and animals. Indeed, some species threatened by land-use change may otherwise be considerably rarer-or even suffer extinction-in the absence of these habitats. Therefore, a critical issue for the maintenance of biodiversity in agricultural landscapes is the extent to which the management of aquatic systems can promote the integration of agricultural production and biodiversity conservation. We completed a cross-sectional study in southern New South Wales (southeastern Australia) to quantify the efficacy of two concurrently implemented management practices-partial revegetation and control of livestock grazing-aimed at enhancing the vegetation structure, biodiversity value, and water quality of farm dams. We found that excluding livestock for even short periods resulted in increased vegetation cover. Relative to unenhanced dams (such as those that remained unfenced), those that had been enhanced for several years were characterized by reduced levels of turbidity, nutrients, and fecal contamination. Enhanced dams also supported increased richness and abundance of macroinvertebrates. In contrast, unenhanced control dams tended to have high abundance of a few macroinvertebrate taxa. Notably, differences remained between the macroinvertebrate assemblages of enhanced dams and nearby "natural" waterbodies that we monitored as reference sites. While the biodiversity value of semilotic, natural waterbodies in the region cannot be replicated by artificial lentic systems, we consider the extensive system of farm dams in the region to represent a novel ecosystem that may nonetheless support some native macroinvertebrates. Our results show that management interventions such as fencing and grazing control can improve water quality in farm dams, improve vegetation structure around farm dams, and support greater abundance and diversity of aquatic macroinvertebrates.

Feral Apis mellifera colonies are widespread globally and cause ecological impacts as pollinators and competitors for food and nesting opportunities. The magnitude of impact depends on their population density, but knowledge of this density is poor. We document feral A. mellifera colonies at 69 per km 2 in fragmented Eucalyptus woodlands in Australia, exceeding estimates from elsewhere in the world, and matched only by one other Australian study. We surveyed 52.5 ha of woodland patches with 357 nest boxes installed to provide nesting opportunities for threatened vertebrates. Our sites covered a region of more than 140 km across with repeated surveys over 3 to 6 years. We show that nest box use by feral A. mellifera colonies is influenced by box design (p = 0.042), with weak evidence for an interactive effect of type of vegetation at a site (woodland remnants vs. replanting) and woody cover within 500 m (p = 0.091). At 69 colonies per km 2 , this density is equivalent to the recommended stocking of hives for pollination of some crops and is therefore likely to influence pollination and lead to competition with other flower visitors. Apis mellifera is also likely to be competing for hollows with cavity dependent native fauna, especially in landscapes where there has been extensive tree removal.

The effect of spatial scale on the location and abundance of species has long been a major topic of interest in ecology. Accounting for key drivers at multiple scales is critical for rigorous description of patterns of species distribution and biodiversity change. We quantified the effects of potential drivers of bird occupancy across a geographically dispersed, but heavily disturbed and fragmented ecosystem.