Joint processing of images in two spectral channels for small objects detecting

The article is devoted to the problem of detecting low contrast small-sized objects in two-color images with a powerful spatially non-stationary background. An increase of the detecting reliability is achieved through a combination of three factors: attenuation of the background based on the construction of its locally stationary model; improving the estimation of model parameters by excluding statistically significant outliers from the initial data; joint processing of two-color images with a weakened background component. A method of constructing a linear boundary for detecting a useful signal in a two-dimensional space is proposed. The performance characteristics of a two-channel detector of small-sized objects are presented.

Modeling the spatial distribution of Xylella fastidiosa. A non-stationary approach with dispersal barriers.

Spatial species distribution models often assume isotropy and stationarity, implying that spatial dependence is direction invariant and uniform throughout the study area. However, these assumptions are violated when dispersal barriers are present. Despite this, the issue of non-stationarity has been little explored in the context of plant health. The objective of this study was to evaluate the influence of barriers in the distribution of Xylella fastidiosa in the demarcated area in Alicante, Spain. Occurrence data from 2018 were analyzed through spatial Bayesian hierarchical models. The stationary model, illustrating a scenario without control interventions or geographical features, was compared with three non-stationary models: a model with mountains as physical barriers, and two models with a continuous and discontinuous perimeter barrier representing hypothetical control interventions. In the stationary model the posterior mean of the spatial range, as the distance where two observations are uncorrelated, was 4,030 m 95% CI (2,907, 5,564). This distance can be used to define the buffer zone in the demarcated area. The predicted probability of X. fastidiosa presence in the area outside the barrier was 0.46 with the stationary model, whereas it was reduced to 0.29 and 0.36 with the continuous and discontinuous barrier models, respectively. Differences between the discontinuous and continuous barrier models showed that breaks, where no control interventions were implemented, resulted in a higher predicted probability of X. fastidiosa presence in the areas with low sampling intensity. These results may help authorities prioritize the areas for surveillance and disease control.

urbanChemFoam 1.0: large-eddy simulation of non-stationary chemical transport of traffic emissions in an idealized street canyon

This paper describes a large-eddy simulation based chemical transport model, developed under the OpenFOAM framework, implemented to simulate dispersion and chemical transformation of nitrogen oxides from traffic sources in an idealized street canyon. The dynamics of the model, in terms of mean velocity and turbulent fluctuation, are evaluated using available stationary measurements. A transient model run using a photostationary reaction mechanism for nitrogen oxides and ozone subsequently follows, where non-stationary conditions for meteorology, background concentrations, and traffic emissions are applied over a 24 h period, using regional model data and measurements obtained for the city of Berlin in July 2014. Diurnal variations of pollutant concentrations indicate dependence on emission levels, background concentrations, and solar state. Comparison of vertical and horizontal profiles with corresponding stationary model runs at select times show that while there are only slight differences in velocity magnitude, visible changes in primary and secondary flow structures can be observed. In addition, temporal variations in diurnal profile and cumulative species concentration result in significant deviations in computed pollutant concentrations between transient and stationary model runs.

Frequency Analysis of Snowmelt Flood Based on GAMLSS Model in Manas River Basin, China

With the acceleration of human economic activities and dramatic changes in climate, the validity of the stationarity assumption of flood time series frequency analysis has been questioned. In this study, a framework for flood frequency analysis is developed on the basis of a tool, namely, the Generalized Additive Models for Location, Scale, and Shape (GAMLSS). We introduced this model to construct a non-stationary model with time and climate factor as covariates for the 50-year snowmelt flood time series in the Kenswat Reservoir control basin of the Manas River. The study shows that there are clear non-stationarities in the flood regime, and the characteristic series of snowmelt flood shows an increasing trend with the passing of time. The parameters of the flood distributions are modelled as functions of climate indices (temperature and rainfall). The physical mechanism was incorporated into the study, and the simulation results are similar to the actual flood conditions, which can better describe the dynamic process of snowmelt flood characteristic series. Compared with the design flood results of Kenswat Reservoir approved by the China Renewable Energy Engineering Institute in December 2008, the design value of the GAMLSS non-stationary model considers that the impact of climate factors create a design risk in dry years by underestimating the risk.

TL-Moments Approach: Application of non-Stationary GEV Model in Flood Frequency Analysis

The non-stationarity in hydrological records is a significant concerning area of interest within the field of flood risk management. Ignoring the non-stationary behaviour in flood series will result in a substantial bias in floods quantile. Hence, the non-stationary flood frequency analysis appeared to be an appropriate option to maintain the independent and identically distributed (IID) assumptions in sample observation. This paper utilized the Generalized Extreme Value (GEV) distribution to analyze extreme flood series. The time-varying moment technique, namely the L-moment and TL-moment methods are employed to estimate the non-stationary model (GEV 1, GEV 2, and GEV 3) in the flood series. The ADF test, Mann-Kendall trend test, and Spearman’s Rho test showed that two out of ten streamflow stations in Johor, Malaysia demonstrated a non-stationary behaviour in the annual maximum streamflow. Results from the simulation study demonstrate a consistent performance on the non-stationary model. Furthermore, the TL-moments method could efficiently predict the flood event estimated at quantiles of the higher return periods.

Non-stationary local signal-and-noise orthogonalization

The local signal-and-noise orthogonalization method has been widely used in the seismic processing and imaging community. In the local signal-and-noise orthogonalization method, a fixed triangle smoother is used for regularizing the local orthogonalization weight, which is based on the assumption that the energy is homogeneously distributed across the whole seismic profile. The fixed triangle smoother limits the performance of the local orthogonalization method in processing complicated seismic datasets. Here, we propose a new local orthogonalization method that uses a variable triangle smoother. The non-stationary smoothing radius is obtained by solving an optimization problem, where the low-pass filtered seismic data are matched by the smoothed data in terms of the local frequency attribute. The new local orthogonalization method with non-stationary model smoothness constraint is called the non-stationary local orthogonalization method. We use several synthetic and field data examples to demonstrate the successful performance of the new method.

Non-Stationary Model of Cerebral Oxygen Transport with Unknown Sources

An inverse problem for a system of equations modeling oxygen transport in the brain is studied. The problem consists of finding the right-hand side of the equation for the blood oxygen transport, which is a linear combination of given functionals describing the average oxygen concentration in the neighborhoods of the ends of arterioles and venules. The overdetermination condition is determined by the values of these functionals evaluated on the solution. The unique solvability of the problem is proven without any smallness assumptions on the model parameters.

Barrier effects on the spatial distribution of Xylella fastidiosa in Alicante, Spain

Spatial models often assume isotropy and stationarity, implying that spatial dependence is direction invariant and uniform throughout the study area. However, these assumptions are violated when dispersal barriers are present in the form of geographical features or disease control interventions. Despite this, the issue of non-stationarity has been little explored in the context of plant health. The objective of this study was to evaluate the influence of different barriers in the distribution of the quarantine plant pathogenic bacterium Xylella fastidiosa in the demarcated area in Alicante, Spain. Occurrence data from the official surveys in 2018 were analyzed with four spatial Bayesian hierarchical models: i) a stationary model representing a scenario without any control interventions or geographical features; ii) a model with mountains as physical barriers; iii) a model with a continuous or iv) discontinuous perimeter barrier as control interventions surrounding the infested area. Barriers were assumed to be totally impermeable, so they should be interpreted as areas without host plants and in which it is not possible for infected vectors or propagating plant material to pass through. Inference and prediction were performed through the integrated nested Laplace approximation methodology and the stochastic partial differential equation approach. In the stationary model the posterior mean of the spatial range was 4,030.17 m 95% CI (2,907.41, 5,563.88), meaning that host plants that are closer to an infected plant than this distance would be at risk for X. fastidiosa. This distance can be used to define the buffer zone around the infested area in Alicante. In the non-stationary models, the posterior mean of the spatial range varied from 3,860.88 m 95% CI (2,918.61, 5,212.18) in the mountain barrier model to 6,141.08 m 95% CI (4,296.32, 9,042.99) in the continuous barrier model. Compared with the stationary model, the perimeter barrier models decreased the probability of X. fastidiosa presence in the area outside the barrier. Differences between the discontinuous and continuous barrier models showed that breaks in areas with low sampling intensity resulted in a higher probability of X. fastidiosa presence. These results may help authorities prioritize the areas for surveillance and implementation of control measures.

urbanChemFoam 1.0: Large-Eddy Simulation of Non-Stationary Chemical Transport of Traffic Emissions in an Idealized Street Canyon

This paper describes a large-eddy simulation based chemical transport model, developed under the OpenFOAM framework, implemented to simulate dispersion and chemical transformation of nitrogen oxides from traffic sources in an idealized street canyon. The dynamics of the model, in terms of mean velocity and turbulent fluctuation, are evaluated using available stationary measurements. A transient model run using a photostationary reaction mechanism for nitrogen oxides and ozone subsequently follows, where non-stationary conditions for meteorology, background concentrations, and traffic emissions are applied over a 24-hour period, using regional model data and measurements obtained for the City of Berlin in July, 2014. Diurnal variations of pollutant concentrations indicate dependence on emission levels, background concentrations, and solar state. Comparison of vertical and horizontal profiles with corresponding stationary model runs at select times show that, while there are only slight differences in velocity magnitude, visible changes in primary and secondary flow structures can be observed. In addition, temporal variations in diurnal profile and cumulative species concentration result in significant deviations in computed pollutant concentrations between transient and stationary model runs.