A Bayesian hurdle model for analysis of an insect resistance monitoring database

Motivated by the analysis of the Australian Grain Insect Resistance Database (AGIRD), we develop a Bayesian hurdle modelling approach to assess trends in strong resistance of stored grain insects to phosphine over time. The binary response variable from AGIRD indicating presence or absence of strong resistance is characterized by a majority of absence observations and the hurdle model is a two step approach that is useful when analyzing such a binary response dataset. The proposed hurdle model utilizes Bayesian classification trees to firstly identify covariates and covariate levels pertaining to possible presence or absence of strong resistance. Secondly, generalized additive models (GAMs) with spike and slab priors for variable selection are fitted to the subset of the dataset identified from the Bayesian classification tree indicating possibility of presence of strong resistance. From the GAM we assess trends, biosecurity issues and site specific variables influencing the presence of strong resistance using a variable selection approach. The proposed Bayesian hurdle model is compared to its frequentist counterpart, and also to a naive Bayesian approach which fits a GAM to the entire dataset. The Bayesian hurdle model has the benefit of providing a set of good trees for use in the first step and appears to provide enough flexibility to represent the influence of variables on strong resistance compared to the frequentist model, but also captures the subtle changes in the trend that are missed by the frequentist and naive Bayesian models. © 2014 Springer Science+Business Media New York.

Conversion of sub-tropical native vegetation to introduced conifer forest: Impacts on below-ground and above-ground carbon pools

Land-use change can have a major influence on soil organic carbon (SOC) and above-ground C pools. We assessed a change from native vegetation to introduced Pinus species plantations on C pools using eight paired sites. At each site we determined the impacts on 0–50 cm below-ground (SOC, charcoal C, organic matter C, particulate organic C, humic organic C, resistant organic C) and above-ground (litter, coarse woody debris, standing trees and woody understorey plants) C pools. In an analysis across the different study sites there was no significant difference (P > 0.05) in SOC or above-ground tree C stocks between paired native vegetation and pine plantations, although significant differences did exist at specific sites. SOC (calculated based on an equivalent soil mass basis) was higher in the pine plantations at two sites, higher in the native vegetation at two sites and did not differ for the other four sites. The site to site variation in SOC across the landscape was far greater than the variation observed with a change from native vegetation to introduced Pinus plantation. Differences between sites were not explained by soil type, although tree basal area was positively correlated with 0–50 cm SOC. In fact, in the native vegetation there was a significant linear relationship between above-ground biomass and SOC that explained 88.8% of the variation in the data. Fine litter C (0–25 mm diameter) tended to be higher in the pine forest than in the adjacent native vegetation and was significantly higher in the pine forest at five of the eight paired sites. Total litter C (0–100 mm diameter) increased significantly with plantation age (R2 = 0.64). Carbon stored in understorey woody plants (2.5–10 cm DBH) was higher in the native vegetation than in the adjacent pine forest. Total site C varied greatly across the study area from 58.8 Mg ha−1 at a native heathland site to 497.8 Mg ha−1 at a native eucalypt forest site. Our findings suggest that the effects of change from native vegetation to introduced Pinus sp. forest are highly site-specific and may be positive, negative, or have no influence on various C pools, depending on local site characteristics (e.g. plantation age and type of native vegetation).

Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. IX. Simulation of soil carbon and nitrogen pools using CENTURY model

Cultivation and cropping of soils results in a decline in soil organic carbon and soil nitrogen, and can lead to reduced crop yields. The CENTURY model was used to simulate the effects of continuous cultivation and cereal cropping on total soil organic matter (C and N), carbon pools, nitrogen mineralisation, and crop yield from 6 locations in southern Queensland. The model was calibrated for each replicate from the original datasets, allowing comparisons for each replicate rather than site averages. The CENTURY model was able to satisfactorily predict the impact of long-term cultivation and cereal cropping on total organic carbon, but was less successful in simulating the different fractions and nitrogen mineralisation. The model firstly over-predicted the initial (pre-cropping) soil carbon and nitrogen concentration of the sites. To account for the unique shrinking and swelling characteristics of the Vertosol soils, the default annual decomposition rates of the slow and passive carbon pools were doubled, and then the model accurately predicted initial conditions. The ability of the model to predict carbon pool fractions varied, demonstrating the difficulty inherent in predicting the size of these conceptual pools. The strength of the model lies in the ability to closely predict the starting soil organic matter conditions, and the ability to predict the impact of clearing, cultivation, fertiliser application, and continuous cropping on total soil carbon and nitrogen.