W. N. Venables

Quantitative approaches in climate change ecology

Contemporary impacts of anthropogenic climate change on ecosystems are increasingly being recognized. Documenting the extent of these impacts requires quantitative tools for analyses of ecological observations to distinguish climate impacts in noisy data and to understand interactions between climate variability and other drivers of change. To assist the development of reliable statistical approaches, we review the marine climate change literature and provide suggestions for quantitative approaches in climate change ecology. We compiled 267 peer-reviewed articles that examined relationships between climate change and marine ecological variables. Of the articles with time series data (n = 186), 75% used statistics to test for a dependency of ecological variables on climate variables. We identified several common weaknesses in statistical approaches, including marginalizing other important non-climate drivers of change, ignoring temporal and spatial autocorrelation, averaging across spatial patterns and not reporting key metrics. We provide a list of issues that need to be addressed to make inferences more defensible, including the consideration of (i) data limitations and the comparability of data sets; (ii) alternative mechanisms for change; (iii) appropriate response variables; (iv) a suitable model for the process under study; (v) temporal autocorrelation; (vi) spatial autocorrelation and patterns; and (vii) the reporting of rates of change. While the focus of our review was marine studies, these suggestions are equally applicable to terrestrial studies. Consideration of these suggestions will help advance global knowledge of climate impacts and understanding of the processes driving ecological change.

Random and Mixed Effects

Models with mixed effects contain both fixed and random effects. Fixed effects are what we have been considering up to now; the only source of randomness in our models arises from regarding the cases as independent random samples. Thus in regression we have an additive measurement error that we assume is independent between cases, and in a GLM we observe independent binomial, Poisson, gamma ... random variates whose mean is a deterministic function of the explanatory variables.

Otosclerosis and chronic tinnitus.

Chronic subjective tinnitus is a common feature of clinical otosclerosis. Analysis of the records of 1,014 consecutive cases of clinical otosclerosis, all confirmed by stapes surgery in South Australia between 1960 and 1972, gives a preoperative prevalence of this symptom of 65%. The association of tinnitus with various predictors is considered, and a statistical analysis is presented. Tinnitus has an association with gender (p < .0001), mean preoperative bone conduction (BC) level (p = .0012), mean air conduction (AC) level (p = .0192), and mean air-bone gap (p = .0075). The associations between tinnitus and the age of the patient, the duration of deafness, the presence of Schwartzes sign, and the severity of footplate pathological involvement were all nonsignificant. The association of tinnitus with the AC and BC thresholds is unexpectedly paradoxical. An economic predictive model for tinnitus in otosclerosis has been constructed from the 2 strongly significant variables, gender and mean BC hearing level, by logistic regression. In this large series of cases, the log odds in favor of finding tinnitus are about 0.810 for male subjects and 1.394 for female subjects when the BC level is zero. The log odds fall by 0.014 for each decibel of mean BC rise.

Factors influencing the probability of residual cholesteatoma

Data from the records of 251 consecutive patients with aural cholesteatoma treated surgically by one surgeon (R.E.G.) between 1962 and 1980, using mainly wide access approaches of atticotomy and attico-antrostomy, and assiduously followed up, formed the basis of an analysis to determine what variables influence the chance of residual cholesteatoma. There were 42 cases of residual cholesteatoma: 20 in the mesotympanum (including two annular), 17 in the epitympanum, and five in the mastoid. The most important variables for the prognosis of residual disease included age, the state of the stapes, the state of the middle ear mucosa, and the number of sites in the middle ear cleft affected by cholesteatoma. A logistic model was constructed to describe the dependence of the probability of residual cholesteatoma on the two determining variables age and state of the stapes. The implications of these findings for the surgical management of cholesteatoma are discussed.

Landscape-scale pest suppression is mediated by timing of predator arrival

There is increasing evidence that biological control of agricultural pests is affected by the landscape context, although the mechanisms behind this pattern have received little attention. Ecological theory predicts that one key mechanism mediating successful pest suppression is early predator immigration to agricultural fields. However, the importance of this population process under different landscape contexts remains unknown. Here, we elucidate the relative importance of landscape context and timing of predator immigration on aphid suppression by manipulating exposure to predation in agroecosystems located across a gradient of landscape complexity in a subtropical horticultural region in Australia. Aphid suppression varied with landscape context, from populations escaping control to almost complete pest suppression. In general, we found higher aphid suppression when predators were allowed immediate and continuous access to aphids than when predators were delayed or excluded for a week, but responses varied in each landscape. Contrary to previous reports from temperate agricultural landscapes, aphid suppression was neutral or negatively associated with natural and seminatural vegetation, whereas aphid suppression was positively associated with landscapes with a higher proportion of alfalfa. When landscapes were classified according to their levels of complexity, we showed that early predation resulted in similar levels of pest suppression in simplified landscapes (i.e., with low proportions of alfalfa and habitat diversity) as late predation in complex landscapes (i.e., with high proportions of alfalfa and habitat diversity). Our data show that timing of predator arrival to agricultural fields is as important as landscape complexity for mediating pest control in agroecosystems. Furthermore, our results suggest that key distributions of suitable habitats that facilitate natural enemy movement can enhance biological control in simplified landscapes.

Impacts of gold mine waste disposal on a tropical pelagic ecosystem

We used a comparative approach to investigate the impact of the disposal of gold mine tailings into the ocean near the Lihir mine (Niolam Island, Papua New Guinea). We found abundance and diversity of zooplankton, micronekton and pelagic fish to be similar or higher in the mine region compared to the reference site. We also found relatively high trace metal concentrations in lower trophic level groups, especially zooplankton, near the mine discharge, but few differences in tissue concentrations of micronekton, baitfish and pelagic fish between the two regions. Biomagnification of some trace metals by micronekton, and of mercury by fish was evident in both regions. We conclude that ocean mine waste disposal at Niolam Island has a local impact on the smaller and less mobile pelagic communities in terms of trace metal concentrations, but has little effect on the abundance and biodiversity of the local food web.

From input to output controls in a short-lived species: the case of Australia's Northern Prawn Fishery

A management strategy evaluation (MSE) framework is developed to evaluate strategies that provide total allowable catches (TACs) when the target biomass corresponds to maximum economic yield (MEY). The framework is applied to Australia’s Northern Prawn Fishery (NPF), which has been actively managed using a tradable input-control system, but is to move to output controls based on individual transferable quotas (ITQs), with a consequence that the current management strategy needs to be replaced. Because the fleet is small, it is possible to set a TAC that cannot be taken entirely. Whereas input controls tend to self-adjust if recruitment is not accurately predicted, and consequently catch variability tends to be low, this is not the case for the reasonably variable species caught in the NPF when they are managed using TACs. The management strategy recovered the simulated stock to the target reference point when it was initially depleted, and avoided dropping the stock below the limit reference point (LRP) for five scenarios based on the current best understanding of resource status; however, in some cases, the stock was left above the target biomass because of biases in the assessment. A scenario in which the stock was forced to be initially depleted to below the limit reference point showed that the management strategy allows recovery.

Sensitivity analysis of spatially aggregated responses: A gradient-based method

We often need to report on environmental, economic and social indicators, and properties at aggregated spatial scales, e.g. average income per suburb. To do this, we invariably create reporting polygons that are somewhat arbitrary. The question arises: how much does this arbitrary subdivision of space affect the outcome? In this paper, we develop a new, gradient‐based framework for carrying out a rigorous analysis of the sensitivity of integrating functions to quantitative changes in their spatial configuration. This approach is applied to both analytical and empirical models, and it allows the reporting of a hierarchy of sensitivity measures (from global to local). We found that the concepts of a vector space representing the spatial configurations and the response (hyper‐)surface on which gradients indicate the sensitivities to be helpful in developing the sensitivity analytical framework of spatial configurations in different dimensions. This approach works well with both analytical and empirical integrating functions. This approach resulted in a clear ranking of the sensitivities of the responses to changes in the reporting regions in an existing environmental reporting application. The approach also allowed us to find which vertices, and the directions of change of those vertices, influenced the outcome most. The application of the spatial framework allows the results to be reported in a hierarchical way, from the sensitivities of an integrative response to changes in a whole reserve/reporting system, down to the sensitivity along each of the dimensions of the vertices in the spatial configuration. The results of the spatial sensitivity framework that we developed in this paper can be readily visualized by plotting the sensitivities as vectors on geographic maps. This simplifies the presentation and facilitates the uptake of the results in the situations where the spatial configurations are complicated.

Ideal colonoscopic surveillance intervals to reduce incidence of advanced adenoma and colorectal cancer.

There is limited information about the interplay between multiple risk factors contributing to the risk of advanced neoplasia. We determined the actual risk for advanced neoplasia in relation to lapsed time between colonoscopies in people enrolled in a structured surveillance program. This risk information can be used to guide the selection of optimal surveillance intervals.

Analysis of long-term hearing gains after stapes surgery with piston reconstruction for otosclerosis.

Objectives: We sought to assess the postoperative hearing gains at individual frequencies after stapes surgery with piston reconstruction for clinical otosclerosis. Methods: We analyzed the air conduction (AC) hearing gains at various audiometric frequencies in a sample of 1,168 stapes procedures with piston reconstruction performed on 911 strictly consecutive patients by one surgeon between 1963 and 1979. Assiduous follow-up of patients was attempted for at least 10 years. The audiometric results over time were stored for computer analysis. Results: The mean AC gain over the speech frequencies (0.5, 1, 2, and 3 kHz) was 35.5 dB at 1 year after operation, and the rate of deterioration over the next 19 years was 0.58 dB/y. Conclusions: The picture to emerge from the analysis is clear. The mean AC gain is maximal at nearly 40 dB for the audiometric frequencies of 0.25, 0.5, and 1 kHz. The mean AC gain at any given time after operation decreases with increasing frequency, at least for frequencies of ≥1 kHz. For any frequency there is, with few exceptions, a significant decrease in the AC gain from one time period to the next.