Northern Jarrah Forest Research Articles

A New Remote Sensing Index for Forest Dryness Monitoring Using Multi-Spectral Satellite Imagery

Canopy water content is a fundamental indicator for assessing the level of plant water stress. The correlation between changes in water content and the spectral reflectance of plant leaves at near-infrared (NIR) and short-wave infrared (SWIR) wavelengths forms the foundation for developing a new remote sensing index, the Infrared Canopy Dryness Index (ICDI), to monitor forest dryness that can be used to predict the consequences of water stress. This study introduces the index, that uses spectral reflectance analysis at near-infrared wavelengths, encapsulated by the Normalized Difference Infrared Index (NDII), in conjunction with specific canopy conditions as depicted by the widely recognized Normalized Difference Vegetation Index (NDVI). Development of the ICDI commenced with the construction of an NDII/NDVI feature space, inspired by a conceptual trapezoid model. This feature space was then parameterized, and the spatial region indicative of water stress conditions, referred to as the dry edge, was identified based on the analysis of 10,000 random observations. The ICDI was produced from the combination of the vertical distance (i.e., under consistent NDVI conditions) from an examined observation to the dry edge. Comparisons between data from drought-affected and non-drought-affected control plots in the Australian Northern Jarrah Forest affirmed that the ICDI effectively depicted forest dryness. Moreover, the index was able to detect incipient water stress several months before damage from an extended drought and heatwave. Using freely available satellite data, the index has potential for broad application in monitoring the onset of forest dryness. This will require validation of the ICDI in diverse forest systems to quantify the efficacy of the index.

Delineation and Classification of Wetlands in the Northern Jarrah Forest, Western Australia Using Remote Sensing and Machine Learning

Wetlands are under increasing pressure from threatening processes. Efforts to protect and monitor wetlands are hampered without datasets capturing the extent, type, and condition. The purpose of this study is to map the distribution of wetland type, vegetation type and vegetation condition for wetlands in the Northern Jarrah Forest region, Western Australia. A random forest algorithm implemented via Google Earth Engine (GEE) was used to classify wetlands and vegetation condition using satellite imagery, topographic indices, and soil mapping. Wetland type was classified using a hierarchical approach incorporating increasing level of detail. Wetland type was mapped as system type from the Interim Australian National Aquatic Ecosystem (ANAE) Classification framework and at hydroperiod level, with overall accuracy of 83% and 82% respectively. Vegetation type was mapped with an accuracy of 78.3%. Mapping of vegetation condition using the Vegetation Assets, States and Transitions (VAST) framework achieved an overall accuracy of 79.6%. Results show that wetlands occur in greater concentration as narrow seasonally waterlogged sites in the west, more sparsely and seasonally inundated sites in the northeast, and as broad seasonally waterlogged sites in the southeast of the study area. Wetland degradation determined through vegetation condition is concentrated in the east, and highest in seasonally waterlogged wetlands. Overall, the wetlands mapping framework implemented in this study can be used by land managers and other interested parties seeking to identify threatened and high conservation value wetlands in other areas.

Geophysics reveals forest vulnerability to drought

AbstractDrought and heat‐induced forest die‐off are being increasingly reported across the planet. As vulnerable areas tend to have thin soils and poor water holding capacities, quantification of soil depth thresholds, relative to drought intensity, has global implications for identifying forest areas at risk. Measuring soil depth at forest stand or regional scales is, however, difficult. Our aim was to quantify soil thickness across drought impacted forest stands using geophysics. We asked whether impacted sites had shallow soils and whether soil thickness was associated with drought effects and forest structure. Electrical resistivity measurements were conducted at three sites in the Northern Jarrah Forest, Western Australia, which experienced die‐off during a 2010 drought and subsequent 2011 heatwave. Multispectral imaging quantified stand structure and vegetation cover. Geophysics identified shallow bedrock in the centre of all drought sites. Soil thicknesses correlated well with stand structure and cover, consistent with increasing water limitation in thinner soils. Smaller cover by trees and shrubs, more ground cover and shorter canopies were observed in soils <20 m thick and were more likely in soils <12 m thick, while treeless areas had thin soils, <2 m thick. The apparent resilience of this forest to long‐term drying looks to be due to the region's deep soils. With predicted future drying it is expected die‐off patches will expand outwards and new die‐off patches will emerge over thin soils. Geophysics can identify areas of forest vulnerable to future drought events, suggesting the potential for landscape‐scale mapping of drought vulnerability via airborne methods.

Drought then wildfire reveals a compound disturbance in a resprouting forest.

The frequency and intensity of forest disturbances, such as drought and fire, are increasing globally, with an increased likelihood of multiple disturbance events occurring in short succession. Disturbances layered over one another may influence the likelihood or intensity of subsequent events (a linked disturbance) or impact response and recovery trajectories (a compound disturbance), with substantial implications for ecological spatiotemporal vulnerability. This study evaluates evidence for disturbance interactions of drought followed by wildfire in a resprouting eucalypt-dominated forest (the Northern Jarrah Forest) in southwestern Australia. Sites were stratified by drought (high, low), from previous modeling and ground validation, and fire severity (high, moderate, unburnt), via remote sensing using the relative difference normalized burn ratio (RdNBR). Evidence of a linked disturbance was assessed via fine fuel consumption and fire severity. Compound disturbance effects were quantified at stand scale (canopy height, quadratic mean diameter, stem density) and stem scale (mortality). There was no evidence of prior drought influencing fine fuel consumption or fire severity and, hence, no evidence of a linked disturbance. However, compound disturbance effects were evident; stands previously affected by drought experienced smaller shifts in canopy height, quadratic mean diameter, and stem density than stands without prior drought impact. At the stem scale, size and fire severity were the strongest determinants of stem survival. Proportional resprouting height was greater in high drought sites than in low drought sites (p < 0.01), meaning, structurally, the low drought stands decreased in height more than the high drought stands. Thus, a legacy of the drought was evident after the wildfire. Although these resprouting eucalypt forests have been regarded as particularly resilient, this study illustrates how multiple disturbances can overwhelm the larger tree component and promote an abundance of smaller stems. We suggest that this is early evidence of a structural destabilization of these forests under a more fire-prone, hotter, and drier future climate.

Soil moisture and temperature drive emergence delays associated with fire seasonality in eucalypt forests

Abstract Many ecosystems are well adapted to fire, although the impacts of fire seasonality and its effect on post-fire recruitment are less well understood. Late summer or autumn fires within eucalypt forests with a Mediterranean-type climate allow for seedling emergence during the cooler and wetter seasons. The emergence and survival after spring fires may be impacted by higher soil temperatures and water stress, delaying recruitment until the subsequent winter period. During this delay, seeds may be exposed to predation and decay, which reduce the viable seed bank. This study examines post-fire recruitment dynamics in a eucalypt forest ecosystem (Northern Jarrah Forest (NJF) of southwestern Western Australia) and whether it may be vulnerable to human-induced changes to fire season. Here, we compare in situ post-fire seedling emergence patterns between autumn and spring burns and account for a potential ecological mechanism driving seasonal differences in emergence by determining the thermal germination requirements of seeds for 15 common species from the NJF. Our results demonstrate that 93% of species had thermal optima between 10°C and 20°C, analogous with soil temperatures measured during the germination window (late April to October). Concurrent in situ post-fire emergence was highest 144 days after an autumn (seasonal) fire, followed by a 10–72% decline. In contrast, there was no emergence within the first 200 days following a spring (aseasonal) fire. We conclude that aseasonal fire in the NJF can lead to a complete delay in recruitment in the first season post-fire, resulting in a lower inter-fire growth period and increasing the potential for further reductions in recruitment through seed predation and decay. The study suggests that aseasonal fire has an immediate and significant impact on initial recruitment in the NJF, but further research is required to determine any longer-term effects of this delay and its implications for fire management in southwestern Western Australia.

Climate change or tree disease: challenges for diagnosing causes of forest die-off

Context Forest ecosystems experience compositional and structural changes as species’ environmental envelopes shift with climate change. Extreme climate events and pests/pathogens are driving these ecosystem changes. Determining which of the two potential drivers is causing a particular forest die-off can be challenging. In south-western Australia, widespread forest die-off in 2011 coincided with extremely hot and dry conditions. It occurred in a forest ecosystem that has historically experienced Phytophthora cinnamomi root disease (Phytophthora dieback). Aims To determine whether the causal agent of Phytophthora dieback, P. cinnamomi, was associated with forest die-off in the Northern Jarrah Forest. Methods A combination of direct (isolation of pathogen) and indirect (survey of susceptible indicator plant species) measurements were taken inside and outside patches of forest experiencing the die-off. Key results There was no consistent association between die-off patches and the presence of P. cinnamomi. P. cinnamomi was isolated from 3 of 33 control plots and 3 of 33 die-off plots. Although several plant species susceptible to P. cinnamomi were absent from die-off plots, the findings were inconsistent across species. This may be explained by plant tolerance to high temperatures and drought. Conclusions P. cinnamomi was not the proximate cause of the observed die-off in the Northern Jarrah Forest in 2011. Implications Novel disturbance caused by extreme climate events can mimic damage caused by certain pests/pathogens. More research is needed to determine the tolerances of plants to extreme temperature and drought conditions to disentangle abiotic and biotic drivers of tree die-off.

Bioturbation by echidna (

Bioturbation by digging animals is important for key forest ecosystem processes such as soil turnover, decomposition, nutrient cycling, water infiltration, seedling recruitment, and fungal dispersal. Despite their widespread geographic range, little is known about the role of the short-beaked echidna (Tachyglossus aculeatus) in forest ecosystems. We measured the density and size of echidna diggings in the Northern Jarrah Forest, south-western Australia, to quantify the contribution echidna make to soil turnover. We recorded an overall density of 298 echidna diggings per hectare, 21% of which were estimated to be less than 1 month old. The average size of digs was 50 ± 25 mm in depth and 160 ± 61 mm in length. After taking into account seasonal digging rates, we estimated that echidnas turn over 1.23 tonnes of soil ha−1 year−1 in this forest, representing an important role in ecosystem dynamics. Our work contributes to the growing body of evidence quantifying the role of these digging animals as critical ecosystem engineers. Given that the echidna is the only Australian digging mammal not severely impacted by population decline or range reduction, its functional contribution to health and resilience of forest ecosystems is increasingly important due to the functional loss of most Australian digging mammals.

Stem functional traits vary among co-occurring tree species and forest vulnerability to drought

Context Stem functional traits are critical for tree hydraulic infrastructure and have important consequences for forest function, particularly concerning vulnerability to drought. Methods Three stem traits, sapwood area, heartwood area, and bark area, were measured in two co-dominant forest species, Eucalyptus marginata Donn. Ex. Sm. and Corymbia calophylla (Lindl.) K.D.Hill &amp; L.A.S.Johnson, in forest patches with low and high vulnerabilities to drought in south-western Australia. Patches of high drought vulnerability experienced die-off during a heatwave and drought in 2011, while patches of low vulnerability were largely not affected. Key results Sapwood area was significantly higher in C. calophylla than in E. marginata, and C. calophylla maintained more sapwood per unit DBH than did E. marginata, especially in larger trees. There was a 29% smaller sapwood area in high drought-vulnerability patches than in low drought-vulnerability patches (including both species). The relationship between sapwood area and DBH varied by tree size. Small trees had a greater sapwood area in high drought-vulnerable patches, whereas larger trees had more sapwood in low drought-vulnerable patches. It is unclear whether sapwood area relationships reflect differences in leaf area or tree age. Conclusions Observed differences in sapwood between species may help explain their differential tolerance to drought, whereas differences between drought-vulnerability sites may suggest adaptation in the studied species. Implications Understanding the traits associated with drought vulnerability will increase our prediction of forest response to drying and warming. Strong relationships between stem traits and DBH, developed here, may help future efforts to model water-use in the Northern Jarrah Forest.

Vegetation classification in south-western Australia’s Mediterranean jarrah forest: new data, old units, and a conservation conundrum

Conservation reserve selection is guided by vegetation classification and mapping. New survey data and improvements in the availability of archived data through online data-sharing platforms enable updated classifications and the critique of existing conservation criteria. In the Northern Jarrah Forest Region of south-western Australia, percentage-based targets using ‘forest ecosystem units’ (15% of each unit) and the systematic conservation planning principles of ‘comprehensiveness, adequacy and representativeness’ underpin the State’s reserve network. To assess the degree of community-level heterogeneity within the forest ecosystem units, new survey data for the forest (30 000 plots) were classified using a non-hierarchical clustering algorithm. Results were assigned to the National Vegetation Information System, and community groups defined at the Association level (Level V). Significant community level heterogeneity was found, including 15 communities in the dominant ‘jarrah woodland’ unit, and 13 in the ‘shrub, herb and sedgelands’ unit. Overall, this research highlights limitations in the current reserve system, including the influence of scale on percentage-based targets and ‘representativeness’. A multi-scale approach to reserve selection, based on a quantitative, floristic, hierarchical classification system, would improve the level of scientific rigour underlying decision-making.

Assessing the viability of tri-trophic food chain model in designing a conservation plan: The case of dwindling Quokka population

The extinction and persistence dynamics of quokka (Setonix brachyurus) population at northern jarrah forest of Australia is investigated using mathematical modelling. Predator's management demands a comprehensive understanding of the ecological circumstances associated with predation. Predation by red foxes (Vulpes vulpes) has a significant role in reducing the native animal population. This paper mainly focuses on the extinction dynamics of quokka population and its reduction by red foxes, by qualitative and quantitative analyses of a designed tri-trophic food-chain model composing a prey (quokka), a mesopredator (red fox) and apex predator (dingo). Existence of solution are analysed and shown to be uniformly bounded. We applied the concept of basic reproduction number from epidemiology to the food chain model, to derive a condition for extinction and persistence of predator population. Global stability of the predator-free equilibrium is established by geometric approach. We use Partial Rank Correlation Coefficient (PRCC) for performing global sensitivity analysis to identify most influential model parameter responsible for quokkas prevalence and mortality. Results of numerical simulation for both deterministic and stochastic model confirms the analytical finding and support those of previous studies. The outcome of this work shows that persistence and existence of quokka depend on the demographic impacts of environmental stochasticity on its own population. Based on our results, many conservation strategies are suggested to improve the overall growth of the species. We believe that declination of red fox and the presence of dingoes are important for preserving the uniform occurrence of quokkas.

Changes in structure of over‐ and midstory tree species in a Mediterranean‐type forest after an extreme drought‐associated heatwave

AbstractWorldwide, extreme climatic events such as drought and heatwaves are associated with forest mortality. However, the precise drivers of tree mortality at individual and stand levels vary considerably, with substantial gaps in knowledge across studies in biomes and continents. In 2010–2011, a drought‐associated heatwave occurred in south‐western Australia and drove sudden and rapid forest canopy collapse. Working in the Northern Jarrah (Eucalyptus marginata) Forest, we quantified the response of key overstory (E. marginata, Corymbia calophylla) and midstory (Banksia grandis, Allocasuarina fraseriana) tree species to the extreme climate event. Using transects spanning a gradient of drought impacts (minimal (50–100 m), transitional (100–150 m) and severe (30–60 m)), tree species mortality in relation to stand characteristics (stand basal area and stem density) and edaphic factors (soil depth) was determined. We show differential mortality between the two overstory species and the two midstory species corresponding to the drought‐associated heatwave. The dominant overstory species, E. marginata, had significantly higher mortality (~19%) than C. calophylla (~7%) in the severe zone. The midstory species, B. grandis, demonstrated substantially higher mortality (~59%) than A. fraseriana (~4%) in the transitional zone. Banksia grandis exhibited a substantial shift in structure in response to the drought‐associated heatwave in relation to tree size, basal area and soil depth. This study illustrates the role of climate extremes in driving ecosystem change and highlights the critical need to identify and quantify the resulting impact to help predict future forest die‐off events and to underpin forest management and conservation.

Carbon consequences of drought differ in forests that resprout.

Prolonged drought and intense heat-related events trigger sudden forest die-off events and have now been reported from all forested continents. Such die-offs are concerning given that drought and heatwave events are forecast to increase in severity and duration as climate change progresses. Quantifying consequences to carbon dynamics and storage from die-off events are critical for determining the current and future mitigation potential of forests. We took stand measurements five times over 2+ years from affected and unaffected plots across the Northern Jarrah Forest, southwestern Australia, following an acute drought/heatwave in 2011. We found a significant loss of live standing carbon (49.3t ha-1 ), and subsequently a significant increase in the dead standing carbon pool by 6months post-die-off. Of the persisting live trees, 38% experienced partial mortality contributing to the rapid regrowth and replenishment (82%-88%) of labile carbon pools (foliage, twigs, and branches) within 26months. Such regrowth was not substantial in terms of net carbon changes within the timeframe of the study but does reflect the resprouting resilience of this forest type. Dead carbon generated by the die-off may persist for centuries given low fragmentation and decay rates resulting in low biogenic emission rates relative to other forest types. However, future fire may threaten persistence of both dead and live pools via combustion and mortality of live tissue and impaired regrowth capacity. Resprouting forests are commonly regarded as resilient systems, however, a changing climate could see vulnerable portions of forests become carbon sources rather than carbon sinks.

Quantifying efficacy of feral pig (Sus scrofa) population management

Abstract ContextFeral pigs (Sus scrofa) are an increasing threat to agriculture and ecological communities globally. Although ground rooting is their most readily observable sign, feral pigs typically remain highly cryptic and their abundance and impacts are difficult to quantify. AimsThe aim of the present study was to evaluate the effect of current feral pig population management practices (trapping, baiting, no feral pig management) on feral pig abundance and digging impacts, using a BACI (before–after control–impact) experimental design at a landscape scale. MethodsA monitoring program was established to quantify both the abundance and digging impacts of feral pig populations within a temperate sclerophyll forest landscape using distance sampling. Transects were established across eight drinking water catchments where the whole catchment was the unit of replication for feral pig population management. Monitoring was carried out at 6-monthly intervals for 3 years, with no feral pig population management undertaken in the first year. In total, 367 feral pigs were trapped out of three catchments subject to trapping, and 26 were baited across two catchments subject to baiting with a commercial product (PIGOUT, Animal Control Technologies Australia, Melbourne, Vic., Australia). Three catchments were exempt from feral pig population management for the duration of this study. Key resultsFeral pig density within the overall study site was estimated as 1.127pigskm–2, resulting in 4580diggingskm–2year–1. There was no significant difference in feral pig density estimates observed among population management treatments or the treatment×year interaction term. An overall decrease in feral pig density across all catchments was attributed to extreme temperature and drought conditions experienced during the study. ConclusionsFeral pig populations demonstrate high resilience to current feral pig population management practices in the present study. The annual volume of soil disturbed by the numbers of feral pigs estimated across this study area is comparable to a commercial-scale resource extraction industry. We did not find significant differences in feral pig digging density among dominant vegetation types, but larger digs were associated with swamp vegetation. ImplicationsCurrent levels of feral pig population management did not reduce pig densities across eight catchments in the northern jarrah forest; therefore, more intensive population management is needed.

Persistence of remnant patches and genetic loss at the distribution periphery in island and mainland populations of the quokka

Understanding the spatial structure of populations is important in developing effective management options for threatened species, and for managing habitat connectivity for metapopulation function, and for demographic and genetic heterogeneity. We used genetic information to investigate the structure of populations of the quokka, Setonix brachyurus, in south-west Western Australia. We hypothesised that movement between known populations would be relatively rare and result in significant genetic structuring. Genetic analyses from 412 adult individuals at 14 nuclear markers (microsatellite) from 33 sampling locations identified structure, diversity and spatial separation of quokkas across their mainland distribution and on two islands. We identified nine inferred (K = 9) populations of quokka that would be otherwise difficult to define with standard ecological techniques. The highest genetic diversity was evident in a large central population of quokka in the southern forest area and genetic diversity was lower at the peripheries of the distribution. The Rottnest Island population contained 70% of the genetic diversity of the mainland populations but the genetic diversity of animals on Bald Island was markedly lower. Populations of quokka in the northern jarrah forest were the only ones to show evidence of recent or long-term population bottlenecking. Of particular interest was the recently identified population at the Muddy Lakes area (the only remaining locality on the Swan Coastal Plain), which was identified as being genetically associated with the southern forest population. Overall, spatial and population cluster analysis showed small insular populations in the northern jarrah forest area, but in the southern forests there appears to be a large panmictic population.

Temporal longevity of unidirectional and dynamic filters to faunal recolonization in post‐mining forest restoration

AbstractUnderstanding the factors, or filters, that preclude the recolonization of, and population persistence in, restoration areas by fauna is important for maximizing the biodiversity benefits of restoration. We investigated filters through a long‐term study examining reptile and mammal recolonization of, and population persistence in, post‐mining restoration in the northern jarrah forest of southwestern Australia. Neither mammal community composition, nor individual mammal species, differed between unmined forest and restoration but reptile communities did and some reptile species were significantly less abundant in restoration than in unmined forest. We concluded that there were no filters to mammal recolonization and persistence but that there were both unidirectional (decreasing gradually over time) and dynamic (fluctuating over time) filters to reptile recolonization of post‐mining restoration that influenced both recolonization and population persistence of some reptile species. Low volumes of coarse woody debris appeared to be a unidirectional filter that will slow recolonization by some reptile species for decades or even centuries. Canopy cover appeared to be a dynamic filter that fluctuated over a period of &gt;7 years and precluded population persistence of some reptile species that had already recolonized the post‐mining restoration. A dearth of studies made it difficult to generalize how widespread these filters were in restoration areas but studies of thinning and burning in natural forests suggested they were likely to be widespread. Although management techniques to reduce the effect of unidirectional and dynamic filters remain poorly developed, their development and adoption will improve the ability of restoration to facilitate fauna recolonization and persistence and better contribute towards biodiversity conservation.

Tell-tale testicles: observations of morphological abnormalities in small, spatially restricted mainland quokka (Setonix brachyurus) populations

The quokka (Setonix brachyurus) exists in spatially restricted populations in the northern jarrah forest in south-west Western Australia. Observations were made of adult male quokkas exhibiting morphological anomalies (cryptorchidism and micropthalmia) that may be indicative of inbreeding within these populations. Despite the presence of males with abnormalities that could potentially affect their fertility, most females captured were carrying a pouch young or feeding a joey at foot. Field researchers and managers should routinely report abnormalities seen in wild captured animals. Reduced genetic diversity of quokka populations in the northern jarrah forest may not be the key threatening process and preservation of habitat may be more important to ensure persistence of populations. Future management of this species in the northern jarrah forest should include up-to-date occurrence mapping across their range using targeted camera trap surveys and management of habitat to improve connectivity between populations.

Overstorey evapotranspiration in a seasonally dry Mediterranean eucalypt forest: Response to groundwater and mining

AbstractGroundwater levels in the northern jarrah forest have declined at rates up to 0.5 m year−1 owing to increased aridity in south‐western Australia in the last 40 years. The forest has also been mined and rehabilitated resulting in significant areas of postmining forest. We tested the impact of declining groundwater levels and mining on evapotranspiration by jarrah forest overstorey. We hypothesized that trees in jarrah forest are facultative phreatophytes (will use groundwater where available but are not reliant on it) and water use per unit overstorey leaf area index (Los) of postmining forest is the same as that of postharvest forest. We measured sapflow at 7 sites in the northern jarrah forest and measured rainfall interception by the canopy at 9 sites. Stemflow was measured at 3 sites. Shallow depth to groundwater was associated with a larger ratio of transpiration per unit leaf area (Eos/Los), but there was little difference in Eos/Los between postmining and postharvest jarrah forest. Eos/Los ranged from 250 ‐ 340 mm year−1 (m2 m−2)−1 at sites where depth to groundwater was &gt;15 m but was up to 400–500 mm year−1 (m2 m−2)−1 at some sites with shallow groundwater. Based on relationships between transpiration, rainfall interception, and Los, it is possible to estimate overstorey evapotranspiration in jarrah forest from Los, especially if spatial layers are available for depth to groundwater. We conclude that jarrah forest is conservative in its water use and likely to be resilient to a drying climate. Management implications for the northern jarrah forest are briefly discussed.

Population monitoring of an endemic macropod, the quokka (Setonix brachyurus), in the northern jarrah forest, Western Australia

Monitoring populations of threatened species plays a part in continued conservation and contributes to assessment of how effective management actions are. We estimated population indices and studied cohort demographics of mainland populations of quokkas (Setonix brachyurus) at 14 sites across the northern jarrah forest. One site is currently monitored through annual trapping, seven were intensively surveyed a decade previously, while six sites had no previous monitoring. Across the 14 study sites, no quokkas were detected at one site and the other population estimates ranged from 5 to 25 adults. Most females (86% of capture events) carried a pouch young or were lactating (indicating a young at foot). Quokka populations at the eight previously-surveyed sites showed variable population changes. We discuss likely contributing factors, including broad-scale fox baiting and fire. Comparative studies of native species over time are important; however, such comparison has limited capacity to explain population changes without comparable methods or where relevant contributing factors (e.g. predator numbers, habitat change) have not likewise been monitored. The threat of changing climate in the northern jarrah forest (where increasing temperatures and greater frequency of drought have been witnessed over the last decades) and implications for control of fire regimes increases the urgency for an updated review of quokka populations to guide appropriate management actions.

Habitat features act as unidirectional and dynamic filters to bat use of production landscapes

Conserving global biodiversity requires careful management of production landscapes, especially in this era of rapid environmental change. The habitat filtering framework has been used for predicting species responses to land-use changes. Habitat filters are essentially features that may slow, or limit, species use of certain habitats. We wanted to determine if this framework could identify habitat and landscape filters that predicted bat use of restored forest at the species-specific and trait group levels. We surveyed bat activity, vegetation structure, and landscape characteristics at 64 sites over two years in restored northern jarrah (Eucalyptus marginata) forests of south-western Australia. Filters to use of restored forest were present for all bats, other than the open space group. We detected dynamic filters to use of restored forest by the open space edge group and unidirectional filters for the closed space edge and closed space groups. Filters to bat use of restored forests were species-specific and related to habitat, rather than landscape, features. In landscapes with relatively low patch-matrix contrast, such as the northern jarrah forest, management actions to reduce filters should focus on habitat features. To manage for the persistence of specific species within restored forest patches, tree density and midstory cover should aim to be at, or close to, reference forest levels.

Preliminary trial of woody debris addition on the return of invertebrates to restored bauxite mines in the jarrah forest of Western Australia

SummaryWoody debris is an important component of forest ecosystems, but its use in mine site restoration has been limited and it can be slow to build up naturally. A new technique of spreading snipped wood waste onto restored mine pits prior to seeding has been subjected to a preliminary trial at Alcoa's Huntly mine site, in the northern jarrah forest of south‐western Western Australia. We examined whether the application of snipped wood during restoration encourages the return of ground‐ and litter‐dwelling invertebrates without negatively suppressing plant establishment. Invertebrates were sampled across three seasons from experimental plots treated with 0 t/ha (control), 100 t/ha or 300 t/ha snipped wood waste. Invertebrate communities in treatment plots comprised higher numbers and diversity of wood and litter decomposers such as mites, Diplopoda, Dermaptera and Blattodea than control plots. Plant responses were variable, with wood treatment resulting in lower tree and overall plant density but having no effects on plant species richness or plant cover. Wood treatment plots were associated with higher soil nitrogen than controls. We hypothesise that the use of a fine wood treatment at the lowest rate of 100 t/ha (approximately 30% wood cover) is likely to enhance the diversity and abundance of invertebrates in restored areas, with minimal effect on plant establishment. Encouraging a diverse invertebrate fauna to recolonise restoration should help speed up succession and ecosystem functions such as decomposition and nutrient cycling, and more quickly return the land to previous ecosystem values.