An Artificial Neutral Network (ANN) model for predicting biodiesel kinetic viscosity as a function of temperature and chemical compositions

An Artificial Neural Network (ANN) is a computational modeling tool which has found extensive acceptance in many disciplines for modeling complex real world problems. An ANN can model problems through learning by example, rather than by fully understanding the detailed characteristics and physics of the system. In the present study, the accuracy and predictive power of an ANN was evaluated in predicting kinetic viscosity of biodiesels over a wide range of temperatures typically encountered in diesel engine operation. In this model, temperature and chemical composition of biodiesel were used as input variables. In order to obtain the necessary data for model development, the chemical composition and temperature dependent fuel properties of ten different types of biodiesels were measured experimentally using laboratory standard testing equipments following internationally recognized testing procedures. The Neural Networks Toolbox of MatLab R2012a software was used to train, validate and simulate the ANN model on a personal computer. The network architecture was optimised following a trial and error method to obtain the best prediction of the kinematic viscosity. The predictive performance of the model was determined by calculating the absolute fraction of variance (R2), root mean squared (RMS) and maximum average error percentage (MAEP) between predicted and experimental results. This study found that ANN is highly accurate in predicting the viscosity of biodiesel and demonstrates the ability of the ANN model to find a meaningful relationship between biodiesel chemical composition and fuel properties at different temperature levels. Therefore the model developed in this study can be a useful tool in accurately predict biodiesel fuel properties instead of undertaking costly and time consuming experimental tests.

The effect of endplate preselection when measuring supine versus standing cobb angle change in idiopathic scoliosis

The primary aim of this study was to determine whether endplate pre-selection makes a difference to the Cobb Angle change between supine and standing which is known to occur in idiopathic scoliosis. A secondary aim of this study was to identify which (if any) patient characteristics were correlated with supine versus standing Cobb change. The study found that pre-selecting vertebral endplates causes only has a minor effect on supine to standing Cobb change in scoliosis. There is a statistically significant relationship between supine to standing Cobb Angle change and fulcrum flexibility. Therefore, supine to standing Cobb Angle change can be considered as a measure of spinal flexibility when both standing and supine images are clinically available.

Comparison of eight published lumbar spine finite element models

Due to its ability to represent intricate systems with material nonlinearities as well as irregular loading, boundary, geometrical and material domains, the finite element (FE) method has been recognized as an important computational tool in spinal biomechanics. Current FE models generally account for a single distinct spinal geometry with one set of material properties despite inherently large inter-subject variability. The uncertainty and high variability in tissue material properties, geometry, loading and boundary conditions has cast doubt on the reliability of their predictions and comparability with reported in vitro and in vivo values. A multicenter study was undertaken to compare the results of eight well-established models of the lumbar spine that have been developed, validated and applied for many years. Models were subjected to pure and combined loading modes and their predictions were compared to in vitro and in vivo measurements for intervertebral rotations, disc pressures and facet joint forces. Under pure moment loading, the predicted L1-5 rotations of almost all models fell within the reported in vitro ranges; their median values differed on average by only 2° for flexion-extension, 1° for lateral bending and 5° for axial rotation. Predicted median facet joint forces and disc pressures were also in good agreement with previously published median in vitro values. However, the ranges of predictions were larger and exceeded the in vitro ranges, especially for facet joint forces. For all combined loading modes, except for flexion, predicted median segmental intervertebral rotations and disc pressures were in good agreement with in vivo values. The simulations yielded median facet joint forces of 0 N in flexion, 38 N in extension, 14 N in lateral bending and 60 N in axial rotation that could not be validated due to the paucity of in vivo facet joint forces. In light of high inter-subject variability, one must be cautious when generalizing predictions obtained from one deterministic model. This study demonstrates however that the predictive power increases when FE models are combined together. The median of individual numerical results can hence be used as an improved tool in order to estimate the response of the lumbar spine.

Prediction of GMA welding characteristic parameter by artificial neural network system

Nowadays, demand for automated Gas metal arc welding (GMAW) is growing and consequently need for intelligent systems is increased to ensure the accuracy of the procedure. To date, welding pool geometry has been the most used factor in quality assessment of intelligent welding systems. But, it has recently been found that Mahalanobis Distance (MD) not only can be used for this purpose but also is more efficient. In the present paper, Artificial Neural Networks (ANN) has been used for prediction of MD parameter. However, advantages and disadvantages of other methods have been discussed. The Levenberg–Marquardt algorithm was found to be the most effective algorithm for GMAW process. It is known that the number of neurons plays an important role in optimal network design. In this work, using trial and error method, it has been found that 30 is the optimal number of neurons. The model has been investigated with different number of layers in Multilayer Perceptron (MLP) architecture and has been shown that for the aim of this work the optimal result is obtained when using MLP with one layer. Robustness of the system has been evaluated by adding noise into the input data and studying the effect of the noise in prediction capability of the network. The experiments for this study were conducted in an automated GMAW setup that was integrated with data acquisition system and prepared in a laboratory for welding of steel plate with 12 mm in thickness. The accuracy of the network was evaluated by Root Mean Squared (RMS) error between the measured and the estimated values. The low error value (about 0.008) reflects the good accuracy of the model. Also the comparison of the predicted results by ANN and the test data set showed very good agreement that reveals the predictive power of the model. Therefore, the ANN model offered in here for GMA welding process can be used effectively for prediction goals.

Can perceptual indices estimate physiological strain across a range of environments and metabolic workloads when wearing explosive ordnance disposal and chemical protective clothing?

Objective Explosive ordnance disposal (EOD) often requires technicians to wear multiple protective garments in challenging environmental conditions. The accumulative effect of increased metabolic cost coupled with decreased heat dissipation associated with these garments predisposes technicians to high levels of physiological strain. It has been proposed that a perceptual strain index (PeSI) using subjective ratings of thermal sensation and perceived exertion as surrogate measures of core body temperature and heart rate, may provide an accurate estimation of physiological strain. Therefore, this study aimed to determine if the PeSI could estimate the physiological strain index (PSI) across a range of metabolic workloads and environments while wearing heavy EOD and chemical protective clothing. Methods Eleven healthy males wore an EOD and chemical protective ensemble while walking on a treadmill at 2.5, 4 and 5.5 km·h− 1 at 1% grade in environmental conditions equivalent to wet bulb globe temperature (WBGT) 21, 30 and 37 °C. WBGT conditions were randomly presented and a maximum of three randomised treadmill walking trials were completed in a single testing day. Trials were ceased at a maximum of 60-min or until the attainment of termination criteria. A Pearson's correlation coefficient, mixed linear model, absolute agreement and receiver operating characteristic (ROC) curves were used to determine the relationship between the PeSI and PSI. Results A significant moderate relationship between the PeSI and the PSI was observed [r = 0.77; p < 0.001; mean difference = 0.8 ± 1.1 a.u. (modified 95% limits of agreement − 1.3 to 3.0)]. The ROC curves indicated that the PeSI had a good predictive power when used with two, single-threshold cut-offs to differentiate between low and high levels of physiological strain (area under curve: PSI three cut-off = 0.936 and seven cut-off = 0.841). Conclusions These findings support the use of the PeSI for monitoring physiological strain while wearing EOD and chemical protective clothing. However, future research is needed to confirm the validity of the PeSI for active EOD technicians operating in the field.

Identifying candidate gene effects by restricting search space in a multivariate genetic analysis of white matter microstructure

Several genetic variants are thought to influence white matter (WM) integrity, measured with diffusion tensor imaging (DTI). Voxel based methods can test genetic associations, but heavy multiple comparisons corrections are required to adjust for searching the whole brain and for all genetic variants analyzed. Thus, genetic associations are hard to detect even in large studies. Using a recently developed multi-SNP analysis, we examined the joint predictive power of a group of 18 cholesterol-related single nucleotide polymorphisms (SNPs) on WM integrity, measured by fractional anisotropy. To boost power, we limited the analysis to brain voxels that showed significant associations with total serum cholesterol levels. From this space, we identified two genes with effects that replicated in individual voxel-wise analyses of the whole brain. Multivariate analyses of genetic variants on a reduced anatomical search space may help to identify SNPs with strongest effects on the brain from a broad panel of genes.

Can’t see the trees for the forest? Why IS-ServQual items matter

Despite longstanding concern with the dimensionality of the service quality construct as measured by ServQual and IS-ServQual instruments, variations on the IS-ServQual instrument have been enduringly prominent in both academic research and practice in the field of IS. We explain the continuing popularity of the instrument based on the salience of the item set for predicting overall customer satisfaction, suggesting that the preoccupation with the dimensions has been a distraction. The implicit mutual exclusivity of the items suggests a more appropriate conceptualization of IS-ServQual as a formative index. This conceptualization resolves the paradox in IS-ServQual research, that of how an instrument with such well-known and well-documented weaknesses continue to be very influential and widely used by academics and practitioners. A formative conceptualization acknowledges and addresses the criticisms of IS-ServQual, while simultaneously explaining its enduring salience by focusing on the items rather than the “dimensions.” By employing an opportunistic sample and adopting the most recent IS-ServQual instrument published in a leading IS journal (virtually, any valid IS- ServQual sample in combination with a previously tested instrument variant would suffice for study purposes), we demonstrate that when re-specified as both first-order and second-order formatives, IS-ServQual has good model quality metrics and high predictive power on customer satisfaction. We conclude that this formative specification has higher practical use and is more defensible theoretically.

Early stopping by using stochastic curtailment in a three-arm sequential trial

Summary. Interim analysis is important in a large clinical trial for ethical and cost considerations. Sometimes, an interim analysis needs to be performed at an earlier than planned time point. In that case, methods using stochastic curtailment are useful in examining the data for early stopping while controlling the inflation of type I and type II errors. We consider a three-arm randomized study of treatments to reduce perioperative blood loss following major surgery. Owing to slow accrual, an unplanned interim analysis was required by the study team to determine whether the study should be continued. We distinguish two different cases: when all treatments are under direct comparison and when one of the treatments is a control. We used simulations to study the operating characteristics of five different stochastic curtailment methods. We also considered the influence of timing of the interim analyses on the type I error and power of the test. We found that the type I error and power between the different methods can be quite different. The analysis for the perioperative blood loss trial was carried out at approximately a quarter of the planned sample size. We found that there is little evidence that the active treatments are better than a placebo and recommended closure of the trial.

Pain in juvenile idiopathic arthritis : parents and children as agents of disease management

Juvenile idiopathic arthritis (JIA) is a severe childhood disease usually characterized by long-term morbidity, unpredictable course, pain, and limitations in daily activities and social participation. The disease affects not only the child but also the whole family. The family is expected to adhere to an often very laborious regimen over a long period of time. However, the parental role is incoherently conceptualized in the research field. Pain in JIA is of somatic origin, but psychosocial factors, such as mood and self-efficacy, are critical in the perception of pain and in its impact on functioning. This study examined the factors correlating and possibly explaining pain in JIA, with a special emphasis on the mutual relations between parent- and patient-driven variables. In this patient series pain was not associated with the disease activity. The degree of pain was on average fairly low in children with JIA. When the children were clustered according to age, anxiety and depression, four distinguishable cluster groups significantly associated with pain emerged. One of the groups was described by concept vulnerability because of unfavorable variable associations. Parental depressive and anxiety symptoms accompanied by illness management had a predictive power in discriminating groups of children with varying distress levels. The parent’s and child’s perception of a child’s functional capability, distress, and somatic self-efficacy had independent explanatory power predicting the child’s pain. Of special interest in the current study was self-efficacy, which refers to the belief of an individual that he/she has the ability to engage in the behavior required for tackling the disease. In children with JIA, strong self-efficacy was related to lower levels of pain, depressive symptoms and trait anxiety. This suggests strengthening a child’s sense of self-efficacy, when helping the child to cope with his or her disease. Pain experienced by a child with JIA needs to be viewed in a multidimensional bio-psycho-social context that covers biological, environmental and cognitive behavioral mechanisms. The relations between the parent-child variables are complex and affect pain both directly and indirectly. Developing pain-treatment modalities that recognize the family as a system is also warranted.

Assessing university students' prior knowledge

The aim of this dissertation was to explore how different types of prior knowledge influence student achievement and how different assessment methods influence the observed effect of prior knowledge. The project started by creating a model of prior knowledge which was tested in various science disciplines. Study I explored the contribution of different components of prior knowledge on student achievement in two different mathematics courses. The results showed that the procedural knowledge components which require higher-order cognitive skills predicted the final grades best and were also highly related to previous study success. The same pattern regarding the influence of prior knowledge was also seen in Study III which was a longitudinal study of the accumulation of prior knowledge in the context of pharmacy. The study analysed how prior knowledge from previous courses was related to student achievement in the target course. The results implied that students who possessed higher-level prior knowledge, that is, procedural knowledge, from previous courses also obtained higher grades in the more advanced target course. Study IV explored the impact of different types of prior knowledge on students’ readiness to drop out from the course, on the pace of completing the course and on the final grade. The study was conducted in the context of chemistry. The results revealed again that students who performed well in the procedural prior-knowledge tasks were also likely to complete the course in pre-scheduled time and get higher final grades. On the other hand, students whose performance was weak in the procedural prior-knowledge tasks were more likely to drop out or take a longer time to complete the course. Study II explored the issue of prior knowledge from another perspective. Study II aimed to analyse the interrelations between academic self-beliefs, prior knowledge and student achievement in the context of mathematics. The results revealed that prior knowledge was more predictive of student achievement than were other variables included in the study. Self-beliefs were also strongly related to student achievement, but the predictive power of prior knowledge overruled the influence of self-beliefs when they were included in the same model. There was also a strong correlation between academic self-beliefs and prior-knowledge performance. The results of all the four studies were consistent with each other indicating that the model of prior knowledge may be used as a potential tool for prior knowledge assessment. It is useful to make a distinction between different types of prior knowledge in assessment since the type of prior knowledge students possess appears to make a difference. The results implied that there indeed is variation between students’ prior knowledge and academic self-beliefs which influences student achievement. This should be taken into account in instruction.

Current CAPE-15: A measure of recent psychotic-like experiences and associated distress

Aim Psychotic-like experiences (PLEs) are common in young people and are associated with both distress and adverse outcomes. The Community Assessment of Psychic Experiences-Positive Scale (CAPE-P) provides a 20-item measure of lifetime PLEs. A 15-item revision of this scale was recently published (CAPE-P15). Although the CAPE-P has been used to assess PLEs in the last 12 months, there is no version of the CAPE for assessing more recent PLEs (e.g. 3 months). This study aimed to determine the reliability and validity of the current CAPE-P15 and assess its relationship with current distress. Method A cross-sectional online survey of 489 university students (17–25 years) assessed lifetime and current substance use, current distress, and lifetime and 3-month PLEs on the CAPE-P15. Results Confirmatory factor analysis indicated that the current CAPE-P15 retained the same three-factor structure as the lifetime version consisting of persecutory ideation, bizarre experiences and perceptual abnormalities. The total score of the current version was lower than the lifetime version, but the two were strongly correlated (r = .64). The current version was highly predictive of generalized distress (r = .52) and indices that combined symptom frequency with associated distress did not confer greater predictive power than frequency alone. Conclusion This study provided preliminary data that the current CAPE-P15 provides a valid and reliable measure of current PLEs. The current CAPE-P15 is likely to have substantial practical utility if it is later shown to be sensitive to change, especially in prevention and early intervention for mental disorders in young people.

Characterising insect plant host relationships facilitates understanding multiple host use

Weed biocontrol relies on host specificity testing, usually carried out under quarantine conditions to predict the future host range of candidate control agents. The predictive power of host testing can be scrutinised directly with Aconophora compressa, previously released against the weed Lantana camara L. (lantana) because its ecology in its new range (Australia) is known and includes the unanticipated use of several host species. Glasshouse based predictions of field host use from experiments designed a posteriori can therefore be compared against known field host use. Adult survival, reproductive output and egg maturation were quantified. Adult survival did not differ statistically across the four verbenaceous hosts used in Australia. Oviposition was significantly highest on fiddlewood (Citharexylum spinosum L.), followed by lantana, on which oviposition was significantly higher than on two varieties of Duranta erecta (‘‘geisha girl’’ and ‘‘Sheena’s gold’’; all Verbenaceae). Oviposition rates across Duranta varieties were not significantly different from each other but were significantly higher than on the two non-verbenaceous hosts (Jacaranda mimosifolia D. Don: Bignoneaceae (jacaranda) and Myoporum acuminatum R. Br.: Myoporaceae (Myoporum)). Production of adult A. compressa was modelled across the hosts tested. The only major discrepancy between model output and their relative abundance across hosts in the field was that densities on lantana in the field were much lower than predicted by the model. The adults may, therefore, not locate lantana under field conditions and/or adults may find lantana but leave after laying relatively few eggs. Fiddlewood is the only primary host plant of A. compressa in Australia, whereas lantana and the others are used secondarily or incidentally. The distinction between primary, secondary and incidental hosts of a herbivore species helps to predict the intensity and regularity of host use by that herbivore. Populations of the primary host plants of a released biological control agent are most likely to be consistently impacted by the herbivore, whereas secondary and incidental host plant species are unlikely to be impacted consistently. As a consequence, potential biocontrol agents should be released only against hosts to which they have been shown to be primarily adapted.

Wood properties and processing outcomes for plantation grown African mahogany (Khaya senegalensis) trees from Clare, Queensland (18 and 20 year-old trees) and Katherine, Northern Territory (14 year-old trees).

Khaya senegalensis, African mahogany, a high-value hardwood, was introduced in the Northern Territory (NT) in the 1950s; included in various trials there and at Weipa, Q in the 1960s-1970s; planted on ex mine sites at Weipa (160 ha) until 1985; revived in farm plantings in Queensland and in trials in the NT in the 1990s; adopted for large-scale, annual planting in the Douglas-Daly region, NT from 2006 and is to have the planted area in the NT extended to at least 20,000 ha. The recent serious interest from plantation growers, including Forest Enterprises Australia Ltd (FEA), has seen the establishment of some large scale commercial plantations. FEA initiated the current study to process relatively young plantation stands from both Northern Territory and Queensland plantations to investigate the sawn wood and veneer recovery and quality from trees ranging from 14 years (NT – 36 trees) to 18-20 years (North Queensland – 31 trees). Field measures of tree size and straightness were complemented with log end splitting assessment and cross-sectional disc sample collection for laboratory wood properties measurements including colour and shrinkage. End-splitting scores assessed on sawn logs were relatively low compared to fast grown plantation eucalypts and did not impact processing negatively. Heartwood proportion in individual trees ranged from 50% up to 92 % of butt cross-sectional disc area for the visually-assessed dark coloured central heartwood and lighter coloured transition wood combined. Dark central heartwood proportion was positively related to tree size (R2 = 0.57). Chemical tests failed to assist in determining heartwood – sapwood boundary. Mean basic density of whole disc samples was 658 kg/m3 and ranged among trees from 603 to 712 kg/m3. When freshly sawn, the heartwood of African mahogany was orange-red to red. Transition wood appeared to be pinkish and the sapwood was a pale yellow colour. Once air dried the heartwood colour generally darkens to pinkish-brown or orange-brown and the effect of prolonged time and sun exposure is to darken and change the heartwood to a red-brown colour. A portable colour measurement spectrophotometer was used to objectively assess colour variation in CIE L*, a* and b* values over time with drying and exposure to sunlight. Capacity to predict standard colour values accurately after varying periods of direct sunlight exposure using results obtained on initial air-dried surfaces decreased with increasing time to sun exposure. The predictions are more accurate for L* values which represent brightness than for variation in the a* values (red spectrum). Selection of superior breeding trees for colour is likely to be based on dried samples exposed to sunlight to reliably highlight wood colour differences. A generally low ratio between tangential and radial shrinkages was found, which was reflected in a low incidence of board distortion (particularly cupping) during drying. A preliminary experiment was carried out to investigate the quality of NIR models to predict shrinkage and density. NIR spectra correlated reasonably well with radial shrinkage and air dried density. When calibration models were applied to their validation sets, radial shrinkage was predicted to an accuracy of 76% with Standard Error of Prediction of 0.21%. There was also a strong predictive power for wood density. These are encouraging results suggesting that NIR spectroscopy has good potential to be used as a non-destructive method to predict shrinkage and wood density using 12mm diameter increment core samples. Average green off saw recovery was 49.5% (range 40 to 69%) for Burdekin Agricultural College (BAC) logs and 41.9% (range 20 to 61%) for Katherine (NT) logs. These figures are about 10% higher than compared to 30-year-old Khaya study by Armstrong et al. (2007) however they are inflated as the green boards were not docked to remove wane prior to being tallied. Of the recovered sawn, dried and dressed volume from the BAC logs, based on the cambial face of boards, 27% could potentially be used for select grade, 40% for medium feature grade and 26% for high feature grades. The heart faces had a slightly higher recovery of select (30%) and medium feature (43%) grade boards with a reduction in the volume of high feature (22%) and reject (6%) grade boards. Distribution of board grades for the NT site aged 14 years followed very similar trends to those of the BAC site boards with an average (between facial and cambial face) 27% could potentially be used for select grade, 42% for medium feature grade, 26% for high feature grade and 5% reject. Relatively to some other subtropical eucalypts, there was a low incidence of borer attack. The major grade limiting defects for both medium and high feature grade boards recovered from the BAC site were knots and wane. The presence of large knots may reflect both management practices and the nature of the genetic material at the site. This stand was not managed for timber production with a very late pruning implemented at about age 12 years. The large amount of wane affected boards is indicative of logs with a large taper and the presence of significant sweep. Wane, knots and skip were the major grade limiting defects for the NT site reflecting considerable amounts of sweep with large taper as might be expected in younger trees. The green veneer recovered from billets of seven Khaya trees rotary peeled on a spindleless lathe produced a recovery of 83% of green billet volume. Dried veneer recovery ranged from 40 to 74 % per billet with an average of 64%. All of the recovered grades were suitable for use in structural ply in accordance to AS/NZ 2269: 2008. The majority of veneer sheets recovered from all billets was C grade (27%) with 20% making D grade and 13% B grade. Total dry sliced veneer recovery from the logs of the two largest logs from each location was estimated to be 41.1%. Very positive results have been recorded in this small scale study. The amount of colour development observed and the very reasonable recoveries of both sawn and veneer products, with a good representation of higher grades in the product distribution, is encouraging. The prospects for significant improvement in these results from well managed and productive stands grown for high quality timber should be high. Additionally, the study has shown the utility of non-destructive evaluation techniques for use in tree improvement programs to improve the quality of future plantations. A few trees combined several of the traits desired of individuals for a first breeding population. Fortunately, the two most promising trees (32, 19) had already been selected for breeding on external traits, and grafts of them are established in the seed orchard.

Plant species diversity of buffer zones in agricultural landscapes: in search of determinants from the local to regional scale

Buffer zones are vegetated strip-edges of agricultural fields along watercourses. As linear habitats in agricultural ecosystems, buffer strips dominate and play a leading ecological role in many areas. This thesis focuses on the plant species diversity of the buffer zones in a Finnish agricultural landscape. The main objective of the present study is to identify the determinants of floral species diversity in arable buffer zones from local to regional levels. This study was conducted in a watershed area of a farmland landscape of southern Finland. The study area, Lepsämänjoki, is situated in the Nurmijärvi commune 30 km to the north of Helsinki, Finland. The biotope mosaics were mapped in GIS. A total of 59 buffer zones were surveyed, of which 29 buffer strips surveyed were also sampled by plot. Firstly, two diversity components (species richness and evenness) were investigated to determine whether the relationship between the two is equal and predictable. I found no correlation between species richness and evenness. The relationship between richness and evenness is unpredictable in a small-scale human-shaped ecosystem. Ordination and correlation analyses show that richness and evenness may result from different ecological processes, and thus should be considered separately. Species richness correlated negatively with phosphorus content, and species evenness correlated negatively with the ratio of organic carbon to total nitrogen in soil. The lack of a consistent pattern in the relationship between these two components may be due to site-specific variation in resource utilization by plant species. Within-habitat configuration (width, length, and area) were investigated to determine which is more effective for predicting species richness. More species per unit area increment could be obtained from widening the buffer strip than from lengthening it. The width of the strips is an effective determinant of plant species richness. The increase in species diversity with an increase in the width of buffer strips may be due to cross-sectional habitat gradients within the linear patches. This result can serve as a reference for policy makers, and has application value in agricultural management. In the framework of metacommunity theory, I found that both mass effect(connectivity) and species sorting (resource heterogeneity) were likely to explain species composition and diversity on a local and regional scale. The local and regional processes were interactively dominated by the degree to which dispersal perturbs local communities. In the lowly and intermediately connected regions, species sorting was of primary importance to explain species diversity, while the mass effect surpassed species sorting in the highly connected region. Increasing connectivity in communities containing high habitat heterogeneity can lead to the homogenization of local communities, and consequently, to lower regional diversity, while local species richness was unrelated to the habitat connectivity. Of all species found, Anthriscus sylvestris, Phalaris arundinacea, and Phleum pretense significantly responded to connectivity, and showed high abundance in the highly connected region. We suggest that these species may play a role in switching the force from local resources to regional connectivity shaping the community structure. On the landscape context level, the different responses of local species richness and evenness to landscape context were investigated. Seven landscape structural parameters served to indicate landscape context on five scales. On all scales but the smallest scales, the Shannon-Wiener diversity of land covers (H') correlated positively with the local richness. The factor (H') showed the highest correlation coefficients in species richness on the second largest scale. The edge density of arable field was the only predictor that correlated with species evenness on all scales, which showed the highest predictive power on the second smallest scale. The different predictive power of the factors on different scales showed a scaledependent relationship between the landscape context and local plant species diversity, and indicated that different ecological processes determine species richness and evenness. The local richness of species depends on a regional process on large scales, which may relate to the regional species pool, while species evenness depends on a fine- or coarse-grained farming system, which may relate to the patch quality of the habitats of field edges near the buffer strips. My results suggested some guidelines of species diversity conservation in the agricultural ecosystem. To maintain a high level of species diversity in the strips, a high level of phosphorus in strip soil should be avoided. Widening the strips is the most effective mean to improve species richness. Habitat connectivity is not always favorable to species diversity because increasing connectivity in communities containing high habitat heterogeneity can lead to the homogenization of local communities (beta diversity) and, consequently, to lower regional diversity. Overall, a synthesis of local and regional factors emerged as the model that best explain variations in plant species diversity. The studies also suggest that the effects of determinants on species diversity have a complex relationship with scale.

Are high-impact species predictable? An analysis of naturalised grasses in northern Australia

Predicting which species are likely to cause serious impacts in the future is crucial for targeting management efforts, but the characteristics of such species remain largely unconfirmed. We use data and expert opinion on tropical and subtropical grasses naturalised in Australia since European settlement to identify naturalised and high-impact species and subsequently to test whether high-impact species are predictable. High-impact species for the three main affected sectors (environment, pastoral and agriculture) were determined by assessing evidence against pre-defined criteria. Twenty-one of the 155 naturalised species (14%) were classified as high-impact, including four that affected more than one sector. High-impact species were more likely to have faster spread rates (regions invaded per decade) and to be semi-aquatic. Spread rate was best explained by whether species had been actively spread (as pasture), and time since naturalisation, but may not be explanatory as it was tightly correlated with range size and incidence rate. Giving more weight to minimising the chance of overlooking high-impact species, a priority for biosecurity, meant a wider range of predictors was required to identify high-impact species, and the predictive power of the models was reduced. By-sector analysis of predictors of high impact species was limited by their relative rarity, but showed sector differences, including to the universal predictors (spread rate and habitat) and life history. Furthermore, species causing high impact to agriculture have changed in the past 10 years with changes in farming practice, highlighting the importance of context in determining impact. A rationale for invasion ecology is to improve the prediction and response to future threats. Although our study identifies some universal predictors, it suggests improved prediction will require a far greater emphasis on impact rather than invasiveness, and will need to account for the individual circumstances of affected sectors and the relative rarity of high-impact species.