Search

Previous analyses of historical aerial photography and satellite imagery have shown thickening of woody cover in Australian tropical savannas, despite increasing fire frequency. The thickening has been attributed to increasing precipitation and atmospheric CO2 enrichment. These analyses involved labour‐intensive, manual classification of vegetation, and hence were limited in the extent of the areas and the number of measurement times used. Object‐based, semi‐automated classification of historical sequences of aerial photography and satellite imagery has enabled the spatio‐temporal analysis of woody cover over entire landscapes, thus facilitating measurement, monitoring and attribution of drivers of change. Using this approach, we investigated woody cover change in 4000 ha of intact mesic savanna in the Ranger uranium lease and surrounding Kakadu National Park, using imagery acquired on 10 occasions between 1950 and 2016. Unlike previous studies, we detected no overall trend in woody cover through time. Some variation in cover was related to rainfall in the previous 12 months, and there were weak effects of fire in the year of image acquisition and the antecedent 4 years. Our local‐scale study showed a mesic eucalypt savanna in northern Australia has been resilient to short‐term variation in rainfall and fire activity; however, changes in canopy cover could have occurred in other settings. When applying this semi‐automated approach to similar studies of savanna dynamics, we recommend maximising the time depth and number of measurement years, standardising the time of year for image acquisition and using many plots of 1 ha in area, rather than fewer, larger plots. [ABSTRACT FROM AUTHOR]

Question: What are the differences between plant communities recognised using supervised versus un‐supervised methods? Location: Northeastern Australia. Methods: Two classifications of savanna plant communities were formed independently with two different approaches: supervised and un‐supervised (using agglomerative hierarchical clustering). Each approach used the same vegetation datasets and, importantly, classification criteria. The communities occur on two different landscapes, with differing environmental gradients, covering an area of 53,500 km2. We compared the internal characteristics of plant communities between approaches and landscapes using four evaluation criteria: identifiability, distinctiveness, similarity of internal heterogeneity and predictability of species foliage cover. Additionally, we compared the central floristic concepts and compositional boundaries of communities identified by each approach. Results: Supervised and un‐supervised approaches recognised similar floristic community concepts. Compositional boundaries between communities were similar on the landscape with steeper environmental gradients but significantly different on the landscape with gradual environmental gradients. However, communities distinguished using supervised methods were significantly less distinct and identifiable, worse at predicting species foliage cover and significantly more variable in species composition than those identified using un‐supervised methods. Conclusions: Using supervised rather than un‐supervised approaches to distinguish plant communities can result in less recognisable communities, possibly reducing their usefulness for land management planning. Importantly, we found a large disparity between the two approaches in delineating compositional boundaries between communities on landscapes with gradual environmental gradients. This is particularly relevant to communities in biomes such as the savanna which comprises 20% of the Earth's landmass. Ecologists can be more confident using a supervised approach on landscapes with steep environmental gradients but should target landscapes with gradual environmental gradients for un‐supervised classification. [ABSTRACT FROM AUTHOR]

Question: What are the differences between plant communities recognised using supervised versus un‐supervised methods? Location: Northeastern Australia. Methods: Two classifications of savanna plant communities were formed independently with two different approaches: supervised and un‐supervised (using agglomerative hierarchical clustering). Each approach used the same vegetation datasets and, importantly, classification criteria. The communities occur on two different landscapes, with differing environmental gradients, covering an area of 53,500 km2. We compared the internal characteristics of plant communities between approaches and landscapes using four evaluation criteria: identifiability, distinctiveness, similarity of internal heterogeneity and predictability of species foliage cover. Additionally, we compared the central floristic concepts and compositional boundaries of communities identified by each approach. Results: Supervised and un‐supervised approaches recognised similar floristic community concepts. Compositional boundaries between communities were similar on the landscape with steeper environmental gradients but significantly different on the landscape with gradual environmental gradients. However, communities distinguished using supervised methods were significantly less distinct and identifiable, worse at predicting species foliage cover and significantly more variable in species composition than those identified using un‐supervised methods. Conclusions: Using supervised rather than un‐supervised approaches to distinguish plant communities can result in less recognisable communities, possibly reducing their usefulness for land management planning. Importantly, we found a large disparity between the two approaches in delineating compositional boundaries between communities on landscapes with gradual environmental gradients. This is particularly relevant to communities in biomes such as the savanna which comprises 20% of the Earth's landmass. Ecologists can be more confident using a supervised approach on landscapes with steep environmental gradients but should target landscapes with gradual environmental gradients for un‐supervised classification. [ABSTRACT FROM AUTHOR]

We review vascular plant plot survey data, plot-based terrestrial vegetation classification protocols and schema by state and bioregions across Australia, discussing recent regional approaches in States and bioregions. A high degree of inconsistency exists in vegetation classification methods and management of plot data across jurisdictions and Australia currently lacks a unified national vegetation plot database, vegetation plot-recording protocols and agreed thematic outputs for classification. Broad-scale classifications were developed largely to support vegetation mapping based on remote sensing of image patterns. Since the 1970s, plot datasets from well-sampled regions have been subjected to a range of analyses (sometimes numerical) producing finer-scaled but idiosyncratic classifications. There are robust numerical classifications for less than 5% of the Australian continent. In poorly sampled and plot-free regions, vegetation classification has focussed on expert description or hybrid approaches using patchy quantitative outputs where they exist. A rigorous quantitative continental regional to local scale approach to classification will not be possible until major data gaps are filled. Additionally, the impact of Australia's inconsistent rainfall on species composition and abundance in analyses of combined datasets requires careful consideration. Given that vegetation classification is integral to Australian biodiversity laws and regulations, there would be benefits in extending quantitative plot coverage to data-poor areas, encouraging cross jurisdictional co-operation of classification procedures, promoting the use of comparable methodologies and sampling under different disturbance and climatic regimes. The development of a consistent vegetation classification across Australia will also require inter-state co-operation, higher standards of plot data curation and plant taxa nomenclature, agreed nation-wide classification protocols and a scientifically defensible hierarchical classification that should integrate with the International Vegetation Classification (IVC). [ABSTRACT FROM AUTHOR]

Vegetation maps are critical biodiversity planning instruments, but the classification of vegetation for mapping can be strongly biased by survey design. Standardization of survey design across different vegetation types is therefore increasingly recommended for vegetation mapping programs. However, some vegetation types have complex small-scale vegetation patterns that are important in characterizing these vegetation types, and standard designs will often not capture these patterns. The objective of this paper was to investigate the magnitude of potential map bias that results from survey design standardization and recommend approaches to deal with this bias. We surveyed upland swamps of the Greater Blue Mountains World Heritage Area Australia using two contrasting survey designs, including the standard 400 m2 single quadrat design recommended and used by authorities. We then derived a classification for these swamps and tested the effect of survey design on this classification, species richness and the type of species detected (obligate or facultative swamp species). Species richness and species type were not significantly different among survey techniques. However, more than 40% of swamps clustered differently among survey designs. Thus, one of the 10 derived communities (which is floristically consistent with a previously mapped endangered community) was indistinct, and some individual swamps misclassified using the standard survey design. An effect of landscape position on swamp floristic patterns and a significant trend for high similarity scores among swamps surveyed with multiple small quadrats compared to the standard survey design was also determined. Australian upland swamps are classified at the global scale as shrub-dominated wetlands, and complex floristic patterns have been recorded in shrub-dominated wetlands in both northern and southern hemispheres. We therefore advocate either multiple survey designs or different survey standards for upland swamp communities and other vegetation types that have complex floristic patterns at small scales. [ABSTRACT FROM AUTHOR]

Abstract: To date, industry and crop forecasters have had a good idea of the potential crop yield for a specific season, but early-season information on crop area for a shire or region has been mostly unavailable. The question of “how early and with what accuracy?” area estimates can be determined using multi-temporal Moderate Resolution Imaging Spectroradiometer (MODIS) enhanced vegetation index (EVI) imagery was investigated in this paper. The study was conducted for two shires in Queensland, Australia for the 2003 and 2004 seasons, and focused on deriving total winter crop area estimates (including wheat, barley and chickpea). A simple metric (), which measures the green-up rate of the crop canopy, was derived. Using the unsupervised k-means classification algorithm, the accumulated difference of two consecutive images (one month apart) for three EVI threshold cut-offs (, where , 500 and 750) at monthly intervals from April to October was calculated. July showed the highest pixel accuracy with percent correctly classified for all thresholds of 94% and 98% for 2003 and 2004, respectively. The differences in accuracy between the three cut-offs were minimal and the T500 threshold was selected as the preferred cut-off to avoid measuring too small or too large fluctuations in the differential EVI values. When compared to the aggregated shire data (surveyed) on crop area across shires and seasons, average percent differences for the for July and August ranged from −19% to 9%. To capture most of the variability in green-up within a region, the average of July and August was used for the early-season prediction of total winter crop area estimates. This resulted in high accuracy (R2=0.96; RMSE = 3157 ha) for predicting the total winter crop from 2000 to 2004 across both shires. This result indicated that this simple multi-temporal remote sensing approach could be used with confidence in early-season crop area prediction at least one to two months ahead of anthesis. [Copyright &y& Elsevier]

By 2050, 70% of the Earth’s human population will live in urban areas. Urbanization can have a devastating impact on local ecosystems, but these impacts vary across time and space. Identifying links between spatiotemporal change in urban ecosystems and neighborhood socio-economics is crucial to management aimed at maintaining flora and fauna in urban areas. Here, we tracked 20 years of socio-economic change and 15 years of vegetation change in 32 residential neighborhoods in south-eastern Australia. Regression models that explicitly accounted for a time lag between neighborhood socio-economic characteristics and vegetation response explained more variation in vegetation cover than models that ignored the effects of time. Also, relationships between vegetation and socio-economic factors were stronger in later years for the same neighborhoods suggesting the influence of socio-economics is more readily identified in established neighborhoods. Socio-economic variables alone, or in combination with biophysical variables, were better predictors of vegetation cover than only biophysical variables. Across space, vegetation cover had a negative quadratic relationship with neighborhood housing density, peaking at mid-density values, and a positive relationship with education level and immigration status (the percentage of residents with a non-Australian background). Over time, housing density had a positive relationship with vegetation cover, reflecting an increase in vegetation as neighborhoods develop. Our results highlight the need to understand temporal context when attempting to explain contemporary patterns in vegetation cover and the increasing importance of socio-economic factors in influencing cover as neighborhoods become established. [ABSTRACT FROM AUTHOR]

Time series of the vegetation index product MOD13Q1 from the Moderate Resolution Imagery Spectroradiometer (MODIS) were assessed for estimating tree foliage projective cover (FPC) and cover change from 2000 to 2006. The MOD13Q1 product consists of the enhanced vegetation index (EVI) and the normalized difference vegetation index (NDVI). There were four challenges in using the MOD13Q1 product to derive tree FPC: assessing the impact of the sensor's varying view geometry on the vegetation index values; decoupling tree and grass cover contributions to the vegetation index signal; devising a method to relate the temporally composited vegetation index pixels to Lidar estimates of tree FPC for calibration; and estimating the accuracy of the FPC and FPC change measurements using independently derived Lidar, Landsat and MODIS cover estimates. The results show that, for complex canopies, the varying view geometry influenced the vegetation indices. The EVI was more sensitive to the view angle than the NDVI, indicating that it is sensitive to vegetation structure. An existing time series method successfully extracted the evergreen vegetation index signal while simultaneously minimizing the impact of varying view geometry. The vegetation indices were better suited to monitoring tree cover change than deriving accurate single-date estimates of cover at regional to continental scales. The EVI was more suited to monitoring change in high-biomass regions (cover >50%) where the NDVI begins to saturate. [ABSTRACT FROM AUTHOR]

Natural disturbances are key processes in the vast majority of ecosystems and a range of ecological theories have been developed in an attempt to predict biotic responses to them. However, empirical support for these theories has been inconsistent and considerable additional work remains to be done to better understand the response of biodiversity to natural disturbance. We tested predictions from the intermediate disturbance hypothesis and the habitat accommodation model of succession for reptile responses to fire history and a single major fire event. We focused our work on a broad range of vegetation types spanning sedgeland to temperate rainforest located within a national park in south-eastern Australia. We found no significant relationships between reptile species richness and the number of fires over the past 35 years, the time since the last fire, or the severity of a major fire in 2003. Thus, we found no strong evidence to support the intermediate disturbance hypothesis. A correspondence analysis of reptile assemblages revealed a gradient in species responses to fire history. However, we found limited evidence for an ordered succession of reptiles. Nor could the responses of individual species be readily predicted from life history attributes. Thus, our findings were generally not consistent with predictions from the habitat accommodation model of succession. A possible explanation for the absence of a predictable sequence of recovery following disturbance might be the rapidity of post-fire recovery of many components of native vegetation cover that were found to be important for reptiles (e.g. the extent of grass cover). This would have limited the time for early successional conditions to prevail and limited opportunities for species associated with such conditions. We found that most reptile species responses were much more strongly linked to vegetation type than fire variables, emphasizing a need to understand relationships with vegetation before being able to understand possible fire effects (if and where they exist). We found the disturbance concepts we examined were limited in their ability to accurately predict reptile responses to past fire history or the impacts of a single major fire in 2003. Practical management might be best guided not by disturbance theory, but by carefully setting objectives to meet conservation goals for particular individual species of reptiles. [ABSTRACT FROM AUTHOR]

The article discusses the classification and mapping of ecological communities in Australia. Researchers in the country use the non-agglomerative, flexible unweighted pair group arithmetic averaging (UPGMA) approach to classify vegetation within a landscape context. The author explains that vegetation maps are spatial expressions of vegetation type and structural patterns. He also points out about the variation in the quality of vegetation maps. He states that the variation is caused by the quality of remote sensing images across different regions.

The article focuses on the study on repeatability of cover estimates in Australia. The study stated that cover estimation has been the standard component in assessing site habitat for planning, and management purposes. However, the repeatability of cover estimation is rare, which led to inconsideration of variability in the outcome value. The study was conducted in Murray-Sunset National Park, where observers were asked the estimate projective foliage cover of the Triodia field. Results suggested that one-off cover estimated from different observers are unrepeatable.