Application and evaluation of classification trees for screening unwanted plants

Risk assessment systems for introduced species are being developed and applied globally, but methods for rigorously evaluating them are still in their infancy. We explore classification and regression tree models as an alternative to the current Australian Weed Risk Assessment system, and demonstrate how the performance of screening tests for unwanted alien species may be quantitatively compared using receiver operating characteristic (ROC) curve analysis. The optimal classification tree model for predicting weediness included just four out of a possible 44 attributes of introduced plants examined, namely: (i) intentional human dispersal of propagules; (ii) evidence of naturalization beyond native range; (iii) evidence of being a weed elsewhere; and (iv) a high level of domestication. Intentional human dispersal of propagules in combination with evidence of naturalization beyond a plants native range led to the strongest prediction of weediness. A high level of domestication in combination with no evidence of naturalization mitigated the likelihood of an introduced plant becoming a weed resulting from intentional human dispersal of propagules. Unlikely intentional human dispersal of propagules combined with no evidence of being a weed elsewhere led to the lowest predicted probability of weediness. The failure to include intrinsic plant attributes in the model suggests that either these attributes are not useful general predictors of weediness, or data and analysis were inadequate to elucidate the underlying relationship(s). This concurs with the historical pessimism that we will ever be able to accurately predict invasive plants. Given the apparent importance of propagule pressure (the number of individuals of an species released), future attempts at evaluating screening model performance for identifying unwanted plants need to account for propagule pressure when collating and/or analysing datasets. The classification tree had a cross-validated sensitivity of 93.6% and specificity of 36.7%. Based on the area under the ROC curve, the performance of the classification tree in correctly classifying plants as weeds or non-weeds was slightly inferior (Area under ROC curve = 0.83 +/- 0.021 (+/- SE)) to that of the current risk assessment system in use (Area under ROC curve = 0.89 +/- 0.018 (+/- SE)), although requires many fewer questions to be answered.

Assessing the Conservation Status and Threats to Priority Plants: a threat assessment approach and case study in South-East Queensland, Australia

bstract: During the Regional Forest Agreement (RFA) process in south-east Queensland, the conservation status of, and threats to, priority vascular plant taxa in the region was assessed. Characteristics of biology, demography and distribution were used to assess the species' intrinsic risk of extinction. In contrast, the threats to the taxa (their extrinsic risk of extinction) were assessed using a decision-support protocol for setting conservation targets for taxa lacking population viability analyses and habitat modelling data. Disturbance processes known or suspected to be adversely affecting the taxa were evaluated for their intensity, extent and time-scale. Expert opinion was used to provide much of the data and to assess the recommended protection areas. Five categories of intrinsic risk of extinction were recognised for the 105 priority taxa: critically endangered (43 taxa); endangered (29); vulnerable (21); rare (10); and presumed extinct (2). Only 6 of the 103 extant taxa were found to be adequately reserved and the majority were considered inadequately protected to survive the current regimes of threatening processes affecting them. Data were insufficient to calculate a protection target for one extant taxon. Over half of the taxa require all populations to be conserved as well as active management to alleviate threatening processes. The most common threats to particular taxa were competition from weeds or native species, inappropriate fire regimes, agricultural clearing, forestry, grazing by native or feral species, drought, urban development, illegal collection of plants, and altered hydrology. Apart from drought and competition from native species, these disturbances are largely influenced or initiated by human actions. Therefore, as well as increased protection of most of the taxa, active management interventions are necessary to reduce the effects of threatening processes and to enable the persistence of the taxa.

Effects of invasive Africanized honey bees (Apis mellifera scutellata) on native stingless bee populations (Meliponinae) and traditional Mayan beekeeping in central Quintana Roo, Mexico

The Maya of the Yucatan region have a long history of keeping the native stingless bees (subfamily Meliponinae). However, market forces in the last few decades have driven the Maya to favor the use of invasive Africanized honey bees (Apis mellifera scutellata) for producing large quantities of high quality honey that has an international market. Furthermore, the native bees traditionally used by the Maya are now disappearing, along with the practice of keeping them. ^ An interdisciplinary approach was taken in order to determine the social factors behind the decrease in stingless beekeeping and the ecological driving forces behind their disappearance from the wild. Social research methods included participant observation with stingless beekeepers, Apis beekeepers, and marketing intermediaries. Ecological research methods included point observations of commonly known melliferous and polliniferous plants along transects in three communities with different degrees of human induced ecosystem disturbance. ^ The stingless bee species most important to the Maya, Melipona beecheii, has become extremely rare, and this has caused a breakdown of stingless beekeeping tradition, compounded with the pressure of the market economy, which fuels Apis beekeeping and has lessened the influence of traditional practices. The community with the heaviest amount of human induced ecosystem disturbance also had the highest degree of dominance of Apis mellifera, while the area with the most intact ecosystem had the highest diversity of stingless bees, though Apis mellifera was still the dominant species. Aggressive competitive behavior involving physical attacks by Apis mellifera against stingless bees was observed on several occasions, and this is a new observation previously unreported by science. ^

Pollination, Herbivory, and Habitat Fragmentation: Their Effects on the Reproductive Fitness of Angadenia berteroi, a Native Perennial Plant of the South Florida Pine Rocklands

Angadenia berteroi is a tropical perennial subshrub of the pine rocklands with large yellow flowers that set very few fruits. My dissertation seeks to elucidate the factors that affect the reproductive fitness of Angadenia berteroi a native species of the south Florida pine rocklands. I provide novel information on the pollination biology of this native species. I also assess the effects of herbivory on growth and the reproductive success of A. berteroi. Finally, I elucidate how habitat fragmentation and quality are correlated with reproductive fitness of this native perennial plant. Using a novel experimental approach, I determined the most effective pollinator group. I used nylon fishing line of widths corresponding to proboscis diameter of the major groups of visitors to examine pollen removal and deposition. In the field, I estimated visitation frequency and efficacy of each pollinator type. Using potted plants, I exposed flowers to single visit from different types of pollinators to measure fruit set. I performed artificial defoliation with scissors on plants growing in the greenhouse to assess the effects of defoliation before flowering as well as during flowering. Additionally, I used structural equation modelling (SEM) to elucidate how A. berteroi reproductive fitness was affected by habitat fragmentation and quality. My experiments provide evidence that Angadenia berteroi is specialized for bee pollination; though butterflies, skippers and others also visit its flowers, A. berteroi is exclusively pollinated by two native bees of the South Florida pine rocklands . This research also demonstrated that herbivory by the oleander moth may have direct and indirect effects on Angadenia berteroi growth and reproductive success. The SEM results suggested that habitat quality (litter depth and subcanopy cover) may favor reproduction in native species of the South Florida pine rocklands that are properly maintained by periodic fires and exotic control. Insights from this threatened and charismatic species may provide impetus to properly manage remaining pine rocklands in South Florida for this and other endemic understory species.

Red mangrove juvenile mangroves and mangrove peat soils at the wetland mesocosm facility. Plants were exposed to phosphorus addition and defolitation, mimicing effects of marine storms on coastal mangrove ecosystems.

http://digitalcommons.fiu.edu/fce_lter_photos/1334/thumbnail.jpg

Comparison of growth traits between abundant and uncommon forms of a non-native vine, Dolichandra unguis-cati (Bignoniaceae) in Australia

Cat’s claw creeper vine, Dolichandra unguis-cati (L.) Lohmann (syn. Macfadyena unguis-cati (L.) Gentry) (Bignoniaceae), is a major environmental weed in Australia. Two distinct forms of this weed (‘long’ and ‘short’ pod), with differences in leaf morphology and fruit size, occur in Australia. The long pod form has only been reported in less than fifteen localities in the whole of south-east Queensland, while the short pod form is widely distributed in Queensland and New South Wales. This study sought to compare growth traits such as specific leaf area, relative growth rate, stem length, shoot/root ratio, tuber biomass and branching architecture between these forms. These traits were monitored under glasshouse conditions over a period of 18 months. Short pod exhibited higher values of relative growth rates, stem length, number of tubers and specific leaf area than long pod, but only after 10 months of plant growth. Prior to this, long and short pod did not differ significantly. Higher values for these traits have been described as characteristics of successful colonizers. Results from this study could partly explain why the short pod form is more widely distributed in Australia while long pod is confined to a few localities.