Summary
Although circular data occur in a wide range of scientific fields, the methodology for distributional modelling and probabilistic forecasting of circular response variables is quite limited. ...Most of the existing methods are built on generalized linear and additive models, which are often challenging to optimize and interpret. Specifically, capturing abrupt changes or interactions is not straightforward but often relevant, e.g. for modelling wind directions subject to different wind regimes. Additionally, automatic covariate selection is desirable when many predictor variables are available, as is often the case in weather forecasting. To address these challenges we suggest a general distributional approach using regression trees and random forests to obtain probabilistic forecasts for circular responses. Using trees simplifies model estimation as covariates are used only for partitioning the data and subsequently just a simple von Mises distribution is fitted in the resulting subgroups. Circular regression trees are straightforward to interpret, can capture non‐linear effects and interactions, and automatically select covariates affecting location and/or scale in the von Mises distribution. Circular random forests regularize and smooth the effects from an ensemble of trees. The new methods are applied to probabilistic wind direction forecasting at two Austrian airports, considering other common approaches as a benchmark.
The role of the atmospheric boundary layer (ABL) in the atmosphere-climate system is the exchange of heat, mass and momentum between ‘the earth’s surface’ and the atmosphere. Traditionally, it is ...understood that turbulent transport is responsible for this exchange and hence the understanding and physical description of the turbulence structure of the boundary layer is key to assess the effectiveness of earth-atmosphere exchange. This understanding is rooted in the (implicit) assumption of a scale separation or spectral gap between turbulence and mean atmospheric motions, which in turn leads to the assumption of a horizontally homogeneous and flat (HHF) surface as a reference, for which both physical understanding and model parameterizations have successfully been developed over the years. Over mountainous terrain, however, the ABL is generically inhomogeneous due to both thermal (radiative) and dynamic forcing. This inhomogeneity leads to meso-scale and even sub-meso-scale flows such as slope and valley winds or wake effects. It is argued here that these (sub)meso-scale motions can significantly contribute to the vertical structure of the boundary layer and hence vertical exchange of heat and mass between the surface and the atmosphere. If model grid resolution is not high enough the latter will have to be parameterized (in a similar fashion as gravity wave drag parameterizations take into account the momentum transport due to gravity waves in large-scale models). In this contribution we summarize the available evidence of the contribution of (sub)meso-scale motions to vertical exchange in mountainous terrain from observational and numerical modeling studies. In particular, a number of recent simulation studies using idealized topography will be summarized and put into perspective – so as to identify possible limitations and areas of necessary future research.
Intra-abdominal surgical intervention can cause the development of intra-peritoneal adhesions. To reduce this problem, different agents have been tested to minimize abdominal adhesions; however, the ...optimal adhesion prophylaxis has not been found so far. Therefore, the A-Part(®) Gel was developed as a barrier to diminish postsurgical adhesions; the aim of this randomized controlled study was a first evaluation of its safety and efficacy.
In this prospective, controlled, randomized, patient-blinded, monocenter phase I-II study, 62 patients received either the hydrogel A-Part-Gel(®) as an anti-adhesive barrier or were untreated after primary elective median laparotomy. Primary endpoint was the occurrence of peritonitis and/or wound healing impairment 28 ± 10 days postoperatively. As secondary endpoints anastomotic leakage until 28 days after surgery, adverse events and adhesions were assessed until 3 months postoperatively.
A lower rate of wound healing impairment and/or peritonitis was observed in the A-Part Gel(®) group compared to the control group: (6.5 vs. 13.8 %). The difference between the two groups was -7.3%, 90 % confidence interval -20.1, 5.4 %. Both treatment groups showed similar frequency of anastomotic leakage but incidence of adverse events and serious adverse events were slightly lower in the A-Part Gel(®) group compared to the control. Adhesion rates were comparable in both groups.
A-Part Gel(®) is safe as an adhesion prophylaxis after abdominal wall surgery but no reduction of postoperative peritoneal adhesion could be found in comparison to the control group. This may at least in part be due to the small sample size as well as to the incomplete coverage of the incision due to the used application.
NCT00646412.
A new probabilistic post-processing method for wind vectors is presented in a distributional regression framework employing the bivariate Gaussian distribution. In contrast to previous studies all ...parameters of the distribution are simultaneously modeled, namely the means and variances for both wind components and also the correlation coefficient between them employing flexible regression splines. To capture a possible mismatch between the predicted and observed wind direction, ensemble forecasts of both wind components are included using flexible two-dimensional smooth functions. This encompasses a smooth rotation of the wind direction conditional on the season and the forecasted ensemble wind direction. The performance of the new method is tested for stations located in plains, mountain foreland, and within an alpine valley employing ECMWF ensemble forecasts as explanatory variables for all distribution parameters. The rotation-allowing model shows distinct improvements in terms of predictive skill for all sites compared to a baseline model that post-processes each wind component separately. Moreover, different correlation specifications are tested and small improvements compared to the model setup with no estimated correlation could be found for stations located in alpine valleys.
While circular data occur in a wide range of scientific fields, the methodology for distributional modeling and probabilistic forecasting of circular response variables is rather limited. Most of the ...existing methods are built on the framework of generalized linear and additive models, which are often challenging to optimize and to interpret. Therefore, we suggest circular regression trees and random forests as an intuitive alternative approach that is relatively easy to fit. Building on previous ideas for trees modeling circular means, we suggest a distributional approach for both trees and forests yielding probabilistic forecasts based on the von Mises distribution. The resulting tree-based models simplify the estimation process by using the available covariates for partitioning the data into sufficiently homogeneous subgroups so that a simple von Mises distribution without further covariates can be fitted to the circular response in each subgroup. These circular regression trees are straightforward to interpret, can capture nonlinear effects and interactions, and automatically select the relevant covariates that are associated with either location and/or scale changes in the von Mises distribution. Combining an ensemble of circular regression trees to a circular regression forest yields a local adaptive likelihood estimator for the von Mises distribution that can regularize and smooth the covariate effects. The new methods are evaluated in a case study on probabilistic wind direction forecasting at two Austrian airports, considering other common approaches as a benchmark.