Probabilistic methods for seasonal forecasting in a changing climate: Cox-type regression models.

For climate risk management, cumulative distribution functions (CDFs) are an important source of information. They are ideally suited to compare probabilistic forecasts of primary (e.g. rainfall) or secondary data (e.g. crop yields). Summarised as CDFs, such forecasts allow an easy quantitative assessment of possible, alternative actions. Although the degree of uncertainty associated with CDF estimation could influence decisions, such information is rarely provided. Hence, we propose Cox-type regression models (CRMs) as a statistical framework for making inferences on CDFs in climate science. CRMs were designed for modelling probability distributions rather than just mean or median values. This makes the approach appealing for risk assessments where probabilities of extremes are often more informative than central tendency measures. CRMs are semi-parametric approaches originally designed for modelling risks arising from time-to-event data. Here we extend this original concept beyond time-dependent measures to other variables of interest. We also provide tools for estimating CDFs and surrounding uncertainty envelopes from empirical data. These statistical techniques intrinsically account for non-stationarities in time series that might be the result of climate change. This feature makes CRMs attractive candidates to investigate the feasibility of developing rigorous global circulation model (GCM)-CRM interfaces for provision of user-relevant forecasts. To demonstrate the applicability of CRMs, we present two examples for El Ni ? no/Southern Oscillation (ENSO)-based forecasts: the onset date of the wet season (Cairns, Australia) and total wet season rainfall (Quixeramobim, Brazil). This study emphasises the methodological aspects of CRMs rather than discussing merits or limitations of the ENSO-based predictors.

The infection of soybean leaves by Phakopsora pachyrhizi during conditions of discontinuous wetness.

The ability of Phakopsora pachyrhizi to cause infection under conditions of discontinuous wetness was investigated. In in vitro experiments, droplets of a uredospore suspension were deposited onto the surface of polystyrene. After an initial wetting period of either 1, 2 or 4 h, the drops were dried for different time intervals and then the wetness was restored for 11, 10 or 8 h. Germination and appressorium formation were evaluated. In in vivo experiments, soybean plants were inoculated with a uredospore suspension. Leaf wetness was interrupted for 1, 3 or 6 h after initial wetting periods of 1, 2 or 4 h. Then, the wetting was re-established for 11, 10 or 8 h, respectively. Rust severity was evaluated 14 days after inoculation. The germination of the spores and the formation of the appressoria on the soybean leaves after different periods of wetness were also quantified in vivo by scanning electron microscopy. P. pachyrhizi showed a high infective capacity during short periods of time. An interruption of wetness after 1 h caused average reductions in germination from 56 to 75% and in appressorium formation from 84 to 96%. Rust severity was lower in all of the in vivo treatments with discontinuous wetness when compared to the control plants. Rust severity was zero when the interruption of wetness occurred 4 h after the initial wetting. Wetting interruptions after 1 and 2 h reduced the average rust severity by 83 and 77%, respectively. The germination of the uredospores on the soybean leaves occurred after 2 h of wetness, with a maximum germination appearing after 4 h of wetness. Wetness interruption affected mainly the spores that had initiated the germination.