Accurately mapping vegetation cover for monitoring large-scale land degradation in arid and semiarid areas using remote sensing images is promising but challenging. We propose a local nonlinear modelling method that integrates geographically weighted regression, a logistic model and optical images which resulted in more accurate predictions of vegetation cover than three traditional, widely used methods in Kangbao County, Hebei Province, northern China. The proposed method provides considerable potential to improve mapping vegetation cover for arid and semiarid areas.

Alpine meadow restoration of Qinghai-Tibetan Plateau remains a substantial challenge to either theoretical or experimental study because of the occurrence of barren patches that directly account for degradation. We modelled the process of development and connection of barren patches, and identified fragmentation and percolation thresholds in the meadow degradation process, independent of spatial scale. This research has demonstrated theoretically that the effectiveness of restoration by periodic effort depends on the existing degree of meadow degradation.

This study evaluates and discusses changes in mountain grassland ecosystem service values (ESVs) and the factors driving them based on remote sensing technology from Urumqi in northern China. The six main grassland types degraded to various levels between 1990 and 2015, and ecological service values had decreased. Grassland degradation continues due to climate change and overgrazing. The control of grazing also has a positive impact on the growth of grassland ESVs in Urumqi.

Feral camels populate 3.3 million km² of remote landscape in central Australia and managing them requires an understanding of how they move across this environment. We investigated camel movements at a finer scale than has been previously attempted, and observed there were different scales of movement in the different land systems they occupied. Better understanding how camels move in these land systems improves how they are better accessed when undertaking in harvest or culling operations.

Chital deer are an introduced ungulate in north Queensland, Australia, where rainfall and plant growth are highly seasonal, and herbivores experience a dry season decline in quality and quantity of forage. We show that chital prefer grass during the wet season and avoid grass during the dry season. Seasonal changes in diet selection occur according to changes in plant phenology which is a good correlate for diet quality.

Chital deer and cattle co-graze rangelands in northern Queensland where dry season shortfalls in forage quality and quantity are a limitation to cattle production. Grass consumption by both ungulates was ~90% of intake during the wet season but during the dry season both grass intake by chital and dry matter digestibility of grass reduced by ~50%. Using an energy model, we estimate that 100 chital consume the grass of 25 cattle during the wet season and 14 during the dry season.