Delineating environmental envelopes to improve mapping of species distributions, via a hurdle model with CART &/or MaxEnT

Species distribution modelling (SDM) typically analyses species’ presence together with some form of absence information. Ideally absences comprise observations or are inferred from comprehensive sampling. When such information is not available, then pseudo-absences are often generated from the background locations within the study region of interest containing the presences, or else absence is implied through the comparison of presences to the whole study region, e.g. as is the case in Maximum Entropy (MaxEnt) or Poisson point process modelling. However, the choice of which absence information to include can be both challenging and highly influential on SDM predictions (e.g. Oksanen and Minchin, 2002). In practice, the use of pseudo- or implied absences often leads to an imbalance where absences far outnumber presences. This leaves analysis highly susceptible to ‘naughty-noughts’: absences that occur beyond the envelope of the species, which can exert strong influence on the model and its predictions (Austin and Meyers, 1996). Also known as ‘excess zeros’, naughty noughts can be estimated via an overall proportion in simple hurdle or mixture models (Martin et al., 2005). However, absences, especially those that occur beyond the species envelope, can often be more diverse than presences. Here we consider an extension to excess zero models. The two-staged approach first exploits the compartmentalisation provided by classification trees (CTs) (as in O’Leary, 2008) to identify multiple sources of naughty noughts and simultaneously delineate several species envelopes. Then SDMs can be fit separately within each envelope, and for this stage, we examine both CTs (as in Falk et al., 2014) and the popular MaxEnt (Elith et al., 2006). We introduce a wider range of model performance measures to improve treatment of naughty noughts in SDM. We retain an overall measure of model performance, the area under the curve (AUC) of the Receiver-Operating Curve (ROC), but focus on its constituent measures of false negative rate (FNR) and false positive rate (FPR), and how these relate to the threshold in the predicted probability of presence that delimits predicted presence from absence. We also propose error rates more relevant to users of predictions: false omission rate (FOR), the chance that a predicted absence corresponds to (and hence wastes) an observed presence, and the false discovery rate (FDR), reflecting those predicted (or potential) presences that correspond to absence. A high FDR may be desirable since it could help target future search efforts, whereas zero or low FOR is desirable since it indicates none of the (often valuable) presences have been ignored in the SDM. For illustration, we chose Bradypus variegatus, a species that has previously been published as an exemplar species for MaxEnt, proposed by Phillips et al. (2006). We used CTs to increasingly refine the species envelope, starting with the whole study region (E0), eliminating more and more potential naughty noughts (E1–E3). When combined with an SDM fit within the species envelope, the best CT SDM had similar AUC and FPR to the best MaxEnt SDM, but otherwise performed better. The FNR and FOR were greatly reduced, suggesting that CTs handle absences better. Interestingly, MaxEnt predictions showed low discriminatory performance, with the most common predicted probability of presence being in the same range (0.00-0.20) for both true absences and presences. In summary, this example shows that SDMs can be improved by introducing an initial hurdle to identify naughty noughts and partition the envelope before applying SDMs. This improvement was barely detectable via AUC and FPR yet visible in FOR, FNR, and the comparison of predicted probability of presence distribution for pres/absence.

Parthenium hysterophorus L. (Asteraceae)

Parthenium is a serious problem in several tropical and sub-tropical areas around the world and particularly an emerging problem in southern Africa. It is a Weed of National Significance in Australia. The chapter summarises current knowledge about the taxonomy, biology, distribution, ecology, impacts and biological control of the weed worldwide. Queensland has led attempts to achieve biological control of parthenium since it first began foreign exploration in 1977. Since then nine insects and two rusts have been released in Queensland. Some of these have since been, or will be, used by other countries. The program has brought significant benefits to Queensland through an increase in grass biomass in some areas. Instances of non-target attack by one agent, particularly in India, are discussed with the conclusion that the effects were ultimately negligible and possibly due to parthenium pollen lodging on the leaves of non-target plants. The insects introduced for parthenium have also given a measure of control for the very closely related weeds, ragweed and Noogoora burr. The paper draws a conclusion that local climatic conditions are very important when considering whether a successful agent in one country will be useful in a second country.

Will biological control of Lantana camara ever succeed? Patterns, processes & prospects

Despite biocontrol research spanning over 100 years, the hybrid weed, commonly referred to as Lantana camara, is not under adequate control. Host specificity and varietal preference of released agents, climatic suitability of a region for released agents, number of agents introduced and range or area of infestation appear to play a role in limiting biocontrol success. At least one of 41 species of mainly leaf- or flower-feeding insects has been introduced, or spread, to 41 of the 70 countries or regions where lantana occurs. Over half (26) of these species have established, achieving varying levels of herbivory and presumably some degree of control. Accurate taxonomy of the plant and adaptation of potential agents to the host plant are some of the better predictors of at least establishment success. Retrospective analysis of the hosts of introduced biocontrol agents for L. camara show that a greater proportion of agents that were collected from L. camara or Lantana urticifolia established, than agents that were collected from other species of Lantana. Of the introduced agents that had established and were oligophagous, 18 out of 22 established. The proportion of species establishing, declined with the number of species introduced. However, there was no trend when oceanic islands were treated separately from mainland areas and the result is likely an artefact of how introductions have changed over time. A calculated index of the degree of herbivory due to agents known to have caused some damage per country, was not related to land area infested with lantana for mainlands nor for oceanic islands. However, the degree of herbivory is much higher on islands than mainlands. This difference between island and mainland situations may reflect population dynamics in patchy or metapopulation landscapes. Basic systematic studies of the host remain crucial to successful biocontrol, especially of hybrid weeds like L. camara. Potential biocontrol agents should be monophages collected from the most closely related species to the target weed or be phytophages that attack several species of lantana. Suitable agents should be released in the most ideal ecoclimatic area. Since collection of biocontrol agents has been limited to a fraction of the known number of phytophagous species available, biocontrol may be improved by targeting insects that feed on stems and roots, as well as the agents that feed on leaves and flowers.

Biology and host range of Ophiomyia camarae Spencer (Diptera: Agromyzidae), a potential biocontrol agent for Lantana spp. (Verbenaceae) in Australia

The life history and host range of the herringbone leaf-mining fly Ophiomyia camarae, a potential biological control agent for Lantana spp., were investigated. Eggs were deposited singly on the underside of leaves. Although several eggs can be laid on a single leaf and a maximum of three individual mines were seen on a single leaf, only one pupa per leaf ever developed. The generation time (egg to adult) was about 38 days. Females (mean 14 days) lived longer than males (mean 9 days) and produced about 61 mines. Oviposition and larval development occurred on all five lantana phenotypes tested. Eleven plant species representing six families were tested to determine the host range. Oviposition and larval development occurred on only lantana and another nonnative plant Lippia alba (Verbenaceae), with both species supporting populations over several generations. A CLIMEX model showed that most of the coastal areas of eastern Australia south to 30°16' S (Coffs Harbour) would be suitable for O. camarae. O. camarae was approved for release in Australia in October 2007 and mines have been observed on plants at numerous field sites along the coast following releases.

Local dispersion and damage of Corymbia citriodora subsp. variegata (Myrtaceae) regrowth by eriophyid mites (Acari: Eriophyoidea).

Eriophyid mites (Acari: Eriophyoidea: Eriophyidae: Rhombacus sp. and Acalox ptychocarpi Keifer) are recently-emerged pests of commercial eucalypt plantations in subtropical Australia. They cause severe blistering, necrosis and leaf loss to Corymbia citriodora subsp. variegata (F. Muell.) K.D. Hill & L.A.S. Johnson, one of the region's most important hardwood plantation species. In this study we examine the progression, incidence and severity of these damage symptoms. We also measure within-branch colonisation by mites to identify dispersive stages, and estimate the relative abundance of the two co-occurring species. Rhombacus sp., an undescribed species, was numerically dominant, accounting for over 90% of all adult mites. Adults were the dispersive stage, moving mostly within branches, but 12% of recruitment onto new leaves occurred on previously uninfested branches. Damage incidence and severity were correlated, while older leaves had more damage than younger leaves. "Patch-type" damage was less frequent but was associated with higher mite numbers and damage scores than "spot-type" damage, while leaf discoloration symptoms related mostly to leaf age.

Response of an invasive liana to simulated herbivory: implications for its biological control

Pre-release evaluation of the efficacy of biological control agents is often not possible in the case of many invasive species targeted for biocontrol. In such circumstances simulating herbivory could yield significant insights into plant response to damage, thereby improving the efficiency of agent prioritisation, increasing the chances of regulating the performance of invasive plants through herbivory and minimising potential risks posed by release of multiple herbivores. We adopted this approach to understand the weaknesses herbivores could exploit, to manage the invasive liana, Macfadyena unguis-cati. We simulated herbivory by damaging the leaves, stem, root and tuber of the plant, in isolation and in combination. We also applied these treatments at multiple frequencies. Plant response in terms of biomass allocation showed that at least two severe defoliation treatments were required to diminish this liana's climbing habit and reduce its allocation to belowground tuber reserves. Belowground damage appears to have negligible effect on the plant's biomass production and tuber damage appears to trigger a compensatory response. Plant response to combinations of different types of damage did not differ significantly to that from leaf damage. This suggests that specialist herbivores in the leaf-feeding guild capable of removing over 50% of the leaf tissue may be desirable in the biological control of this invasive species.

Inland Salt

This work, a small single room installation, was my contribution to the 2015 Palimpsest Biennale in Mildura, and was shown at the ADFA building in the town's centre. On an initial research trip in July, 2015, I stopped at a dried out salt lake near Wentworth. Beach-like, the lake left a tidal line along a curving shore. The vegetation looked like ocean flora; there was the glimmering saltiness of everything, and saltbush everywhere. It brought to mind Sturt's hankering for an inland sea, and his wishful voyage along the Murray, that led him out, not in. This work puts these observations together: a drawing, made from the sanded emboss of saltbush leaves, runs, riverlike, across the space. A looped film of clouds reflected in a shallow saltlake, recalls the dream of an inland sea, supported by the sound of ocean.

Functional dynamics of the nitrogen balance of sorghum. II. Grain filling period.

Maintenance of green leaf area during grain filling can increase grain yield of sorghum grown under terminal water limitation. This 'stay-green' trait has been related to the nitrogen (N) supply-demand balance during grain filling. This study quantifies the N demand of grain and N translocation rates from leaves and stem and explores effects of genotype and N stress on onset and rate of leaf senescence during the grain filling period. Three hybrids differing in potential height were grown at three levels of N supply under well-watered conditions. Vertical profiles of biomass, leaf area, and N% of leaves, stem and grain were measured at regular intervals. Weekly SPAD chlorophyll readings on main shoot leaves were correlated with observed specific leaf nitrogen (SLN) to derive seasonal patterns of leaf N content. For all hybrids, individual grain N demand was sink determined and was initially met through N translocation from the stem and rachis. Only if this was insufficient did leaf N translocation occur. Maximum N translocation rates from leaves and stem were dependent on their N status. However, the supply of N at canopy scale was also related to the amount of leaf area senescing at any one time. This supply-demand framework for N dynamics explained effects of N stress and genotype on the onset and rate of leaf senescence.

Functional dynamics of the nitrogen balance of sorghum: I. N demand of vegetative plant parts.

Stay-green, an important trait for grain yield of sorghum grown under water limitation, has been associated with a high leaf nitrogen content at the start of grain filling. This study quantifies the N demand of leaves and stems and explores effects of N stress on the N balance of vegetative plant parts of three sorghum hybrids differing in potential crop height. The hybrids were grown under well-watered conditions at three levels of N supply. Vertical profiles of biomass and N% of leaves and stems, together with leaf size and number, and specific leaf nitrogen (SLN), were measured at regular intervals. The hybrids had similar minimum but different critical and maximum SLN, associated with differences in leaf size and N partitioning, the latter associated with differences in plant height. N demand of expanding new leaves was represented by critical SLN, and structural stem N demand by minimum stem N%. The fraction of N partitioned to leaf blades increased under N stress. A framework for N dynamics of leaves and stems is developed that captures effects of N stress and genotype on N partitioning and on critical and maximum SLN.

Pest-damage relationships for Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae) on vegetative soybean.

The response of vegetative soybean (Glycine max) to Helicoverpa armigera feeding was studied in irrigated field cages over three years in eastern Australia to determine the relationship between larval density and yield loss, and to develop economic injury levels. Rather than using artificial defoliation techniques, plants were infested with either eggs or larvae of H. armigera, and larvae allowed to feed until death or pupation. Larvae were counted and sized regularly and infestation intensity was calculated in Helicoverpa injury equivalent (HIE) units, where 1 HIE was the consumption of one larva from the start of the infestation period to pupation. In the two experiments where yield loss occurred, the upper threshold for zero yield loss was 7.51 ± 0.21 HIEs and 6.43 ± 1.08 HIEs respectively. In the third experiment, infestation intensity was lower and no loss of seed yield was detected up to 7.0 HIEs. The rate of yield loss/HIE beyond the zero yield loss threshold varied between Experiments 1 and 2 (-9.44 ± 0.80 g and -23.17 ± 3.18 g, respectively). H. armigera infestation also affected plant height and various yield components (including pod and seed numbers and seeds/pod) but did not affect seed size in any experiment. Leaf area loss of plants averaged 841 and 1025 cm2/larva in the two experiments compared to 214 and 302 cm2/larva for cohort larvae feeding on detached leaves at the same time, making clear that artificial defoliation techniques are unsuitable for determining H. armigera economic injury levels on vegetative soybean. Analysis of canopy leaf area and pod profiles indicated that leaf and pod loss occurred from the top of the plant downwards. However, there was an increase in pod numbers closer to the ground at higher pest densities as the plant attempted to compensate for damage. Defoliation at the damage threshold was 18.6 and 28.0% in Experiments 1 and 2, indicating that yield loss from H. armigera feeding occurred at much lower levels of defoliation than previously indicated by artificial defoliation studies. Based on these results, the economic injury level for H. armigera on vegetative soybean is approximately 7.3 HIEs/row-metre in 91 cm rows or 8.0 HIEs/m2.

Isolation and functional characterization of a lycopene beta-cyclase gene that controls fruit colour of papaya (Carica papaya L.).

The colour of papaya fruit flesh is determined largely by the presence of carotenoid pigments. Red-fleshed papaya fruit contain lycopene, whilst this pigment is absent from yellow-fleshed fruit. The conversion of lycopene (red) to beta-carotene (yellow) is catalysed by lycopene beta-cyclase. This present study describes the cloning and functional characterization of two different genes encoding lycopene beta-cyclases (lcy-beta1 and lcy-beta2) from red (Tainung) and yellow (Hybrid 1 B) papaya cultivars. A mutation in the lcy-beta2 gene, which inactivates enzyme activity, controls lycopene production in fruit and is responsible for the difference in carotenoid production between red and yellow-fleshed papaya fruit. The expression level of both lcy-beta1 and lcy-beta2 genes is similar and low in leaves, but lcy-beta2 expression increases markedly in ripe fruit. Isolation of the lcy-beta2 gene from papaya, that is preferentially expressed in fruit and is correlated with fruit colour, will facilitate marker-assisted breeding for fruit colour in papaya and should create possibilities for metabolic engineering of carotenoid production in papaya fruit to alter both colour and nutritional properties.

A Case of Simple Tool Use in Wild Liontailed Macaques (Macaca silenus)

Three members of a group of liontailed macaques (Macaea silenus) were seen to use leaves for food preparation. Other examples of prey-selection and hunting behaviour in liontailed macaques reflect individual- and group-specific skills. The absence of similar patterns in bonnet macaques (Macaca radiata) living in the same habitat might be related to differences in the social design and indicate the high significance of social aspects for the occurrence and manifestation of innovative behaviour.

The use of an acetoacetyl-CoA synthase in place of a -ketothiolase enhances poly-3-hydroxybutyrate production in sugarcane mesophyll cells

Engineering the production of polyhydroxyalkanoates (PHAs) into high biomass bioenergy crops has the potential to provide a sustainable supply of bioplastics and energy from a single plant feedstock. One of the major challenges in engineering C-4 plants for the production of poly[(R)-3-hydroxybutyrate] (PHB) is the significantly lower level of polymer produced in the chloroplasts of mesophyll (M) cells compared to bundle sheath (BS) cells, thereby limiting the full PHB yield-potential of the plant. In this study, we provide evidence that the access to substrate for PHB synthesis may limit polymer production in M chloroplasts. Production of PHB in M cells of sugarcane is significantly increased by replacing -ketothiolase, the first enzyme in the bacterial PHA pathway, with acetoacetyl-CoA synthase. This novel pathway enabled the production of PHB reaching an average of 6.3% of the dry weight of total leaf biomass, with levels ranging from 3.6 to 11.8% of the dry weight (DW) of individual leaves. These yields are more than twice the level reported in PHB-producing sugarcane containing the -ketothiolase and illustrate the importance of producing polymer in mesophyll plastids to maximize yield. The molecular weight of the polymer produced was greater than 2x10(6)Da. These results are a major step forward in engineering a high biomass C-4 grass for the commercial production of PHB.

Austropleospora osteospermi gen. et sp nov and its host specificity and distribution on Chrysanthemoides monilifera ssp rotundata in Australia.

Hendersonia osteospermi was found for the first time in Australia on leaf spots of the introduced invasive plant Chrysanthemoides monilifera ssp. rotundata (bitou bush) in coastal regions of New South Wales. Pathogenicity tests on species from 11 tribes in the family Asteraceae, demonstrated that H. osteospermi caused severe necrosis on leaves and stems of C. monilifera ssp. rotundata and its congener C. monilifera ssp. monilifera (boneseed). Small necrotic spots also developed on Osteospermum fruticosum and Dimorphotheca cuneata in the Calenduleae and on Helianthus annuus (sunflower) in the Heliantheae. None of the other plant species tested developed leaf spots, although H. osteospermi was re-isolated from senescent leaves of Cynara scolymus (globe artichoke) in the Cynareae and Vernonia cinerea in the Vernonieae. Single ascospores from ascomata of a Pleospora-like fungus found on diseased stems of bitou bush produced H. osteospermi in culture, which proved the anamorph/teleomorph connection. The ITS region of both a single-ascospore isolate and a single-conidium isolate were sequenced and found to be identical. The taxonomic status of H. osteospermi is re-examined and Austropleospora osteospermi gen. et sp. nov. is described as its teleomorph based on morphology, host range tests and DNA sequence analysis. The potential of A. osteospermi for the biological control of bitou bush and boneseed in Australia is discussed.

Nitrogen use for high productivity and sustainability in cashew.

Interest in cashew production in Australia has been stimulated by domestic and export market opportunities and suitability of large areas of tropical Australia. Economic models indicate that cashew production is profitable at 2.8 t ha-1 nut-in-shell (NIS). Balanced plant nutrition is essential to achieve economic yields in Australia, with nitrogen (N) of particular importance because of its capacity to modify growth, affect nut yield and cause environmental degradation through soil acidification and off-site contamination. The study on a commercial cashew plantation at Dimbulah, Australia, investigated the effect of N rate and timing on cashew growth, nut production, N leaching and soil chemical properties over five growth cycles (1995-1999). Nitrogen was applied during the main periods of vegetative (December-April) and reproductive (June-October) growth. Commercial NIS yields (up to 4.4 t ha-1 from individual trees) that exceeded the economic threshold of 2.8 t ha-1 were achieved. The yield response was mainly determined by canopy size as mean nut weight, panicle density and nuts per panicle were largely unaffected by N treatments. Nitrogen application confined to the main period of vegetative growth (December-April) produced a seasonal growth pattern that corresponded most consistently with highest NIS yield. This N timing also reduced late season flowering and undesirable post-November nut drop. Higher yields were not produced at N rates greater than 17 g m-2 of canopy surface area (equating to 210 kg N ha-1 for mature size trees). High yields were attained when N concentrations in Mveg leaves in May-June were about 2%, but this assessment occurs at a time when it is not feasible to correct N deficiency. The Mflor leaf of the preceding November, used in conjunction with the Mveg leaf, was proposed as a diagnostic tool to guide N rate decisions. Leaching of nitrate-N and acidification of the soil profile was recorded to 0.9 m. This is an environmental and sustainability hazard, and demonstrates that improved methods of N management are required.