Search

Western Tasmania contains large areas of treeless sedgeland interspersed with scrub and forest vegetation. Why these large treeless areas exist in a region wet enough to support rainforest remains unclear. Slower growth of woody plants in the sedgeland than the scrub and forest could be a factor, because smaller woody plants are more likely to be killed by frequent fire before they mature and set seed. However, there is little information regarding growth rates of woody plants in these communities. We characterised soils and compared growth rates of small woody plants in sedgeland, scrub and forest at two sites using diverse approaches: (i) field observational studies of seedlings from a broad range of species that established following a wildfire in 2019 at site 1; (ii) a dendrochronological study of mature myrtaceous shrubs that had established following a fire in 1983 and were killed by the 2019 fire at site 1 (iii) a field experiment in which fertiliser was applied to transplanted seedlings of the myrtaceous shrub Leptospermum scoparium at site 2. (iv) We also used a glasshouse experiment to test the effects of nutrients and waterlogging on L. scoparium seedlings. Sedgeland soils were organic and poorly drained, whereas forest soils were mineral and well drained; scrub soils were organic at the surface but well drained. Concentrations of most soil nutrients, including total P, K, Ca, Mg and Mn were highest in the forest, although N was highest in sedgeland. At both sites, woody plant species grew faster in scrub and forest than in sedgeland, by a factor of 1.7–3.4. When fertiliser was added to transplanted L. scoparium seedlings, there was no difference in growth between sedgeland and scrub. In the glasshouse experiment, growth was limited by both low soil nutrients and waterlogging. However, the waterlogging effect was not apparent in the field experiment because conditions were dry. Our study shows slow growth of woody plants in Tasmanian sedgelands associated with infertile and waterlogged organic soils. Fire disturbance sharpens the boundary but is not the cause of it. • Tasmania contains areas of treeless sedgeland interspersed with forest and scrub. • An unproven hypothesis is that tree growth is slower in sedgeland. • We conducted observational and experimental growth studies in field and glasshouse. • All studies showed slower tree growth in sedgeland than scrub or forest. • Edaphic factors (low nutrients and waterlogging) can explain slower tree growth. [ABSTRACT FROM AUTHOR]

Land-cover data provides valuable information for landscape management and can be generated using machine learning algorithms. Ensemble models or model averaging can overcome difficulties in selecting an adequate algorithm and improve model predictions, but its use is limited among ecologists. The objective of this study is to highlight the benefits and limitations of weighted and unweighted majority voting ensemble models for land-cover classification and to enable easy and wider implementation of the method by providing an R-script (for use in the R software). Using a case study of three mixed-use landscapes from southern Australia (Tasmania), land cover was classified into six classes using Landsat 8 imagery and ancillary data, and support vector machine, random forest, k-nearest neighbour and naïve Bayesian as base algorithms. The predicted classifications of the base algorithms were then averaged using both an unweighted and weighted (using the true skill statistic) majority voting ensemble algorithm. Cross-validation results showed the base algorithms achieved similar accuracy making algorithm selection difficult. The base algorithms achieved high and similar predictive accuracy when the classified land-cover and training data belong to the same geographic region but lower and different predictive accuracy when the classified land-cover and training data belong to different geographic regions. The weighted and unweighted ensemble achieved similar overall accuracy, equivalent to the best performing base algorithm. We conclude that the majority voting ensemble can be adopted to overcome difficulties in model selection during land-cover classification. • Using cross-validation resulted in no preferred algorithm for classification. • Different base algorithms achieved the highest accuracy in different scenarios. • The weighted and unweighted ensemble consistently achieved high accuracy. • The ensemble algorithm is provided as a script for use with software R. [ABSTRACT FROM AUTHOR]

Abstract: Vision plays an essential role in development and visual deficit constitutes a risk factor not only for visuo-sensory development, but also for overall development. Our experience working with visually impaired children and our consideration of these issues have allowed us to reach a number of conclusions in relation to early intervention in the visually impaired child. The first step in early intervention is to establish the origin of the visual impairment. In fact, the spectrum of visual problems in infancy is very broad and ranges from peripheral disorders due to receptor problems to cerebral visual impairment (CVI), a neurological disorder of childhood caused by damage to, or malfunctioning of, the retrochiasmatic visual pathways (optic radiations, occipital cortex, associative visual areas) in the absence of any major ocular disease. Sensory experience from the external world can influence how the visual pathways wire themselves up after birth. Visual experience is crucial, and it is also crucial to help the child learn to integrate alternative sensory information. Early intervention in visually impaired infants is mandatory, and treatment of sensory input impairments should begin as early as possible in a positive emotional setting that enhances the child''s motivation and relationship with caregivers. [Copyright &y& Elsevier]