Search

Wildfire occurrence and behaviour are difficult to predict locally for the next day. In the present work, we propose relying on fire spread simulations to provide a fire danger index representative of the potential for fire spread that includes not only weather but also surrounding vegetation and orography. This is achieved using an artificial neural network emulator called DeepFire, trained based on simulated fire sizes. To determine how relevant this index can be in the assessment of next-day fire danger, the application of DeepFire to fire danger mapping using actual weather forecasts is studied. DeepFire forecasts for 13 fairly large fires that occurred in Corsica are analysed and compared with corresponding forecasts using another fire danger index used in operational conditions, highlighting the differences in terms of precision and the expected results of such predictions. The weather forecasts from which the weather inputs of DeepFire are determined have high spatial resolution and high frequency, which also applies to the fire danger predictions. Additionally, input uncertainty is propagated through DeepFire, resulting in ensembles of emulated fire size. Several approaches are proposed to analyse the results and provide fire danger maps and ratings using this new simulation-based prediction system. [ABSTRACT FROM AUTHOR]

The influence of the ground and atmosphere on sound generation and propagation from wind turbines creates uncertainty in sound level estimations. Realistic simulations of wind turbine noise thus require quantifying the overall uncertainty on sound pressure levels induced by environmental phenomena. This study proposes a method of uncertainty quantification using a quasi-Monte Carlo method of sampling influential input data (i.e., environmental parameters) to feed an Amiet emission model coupled with a Parabolic Equation propagation model. This method allows for calculation of the probability distribution of the output data (i.e., sound pressure levels). As this stochastic uncertainty quantification method requires a large number of simulations, a metamodel of the global (emission-propagation) wind turbine noise model was built using the kriging interpolation technique to drastically reduce calculation time. When properly employed, the metamodeling technique can quantify statistics and uncertainties in sound pressure levels at locations downwind from wind turbines. This information provides better knowledge of sound pressure variability and will help to better control the quality of wind turbine noise prediction for inhomogeneous outdoor environments. [ABSTRACT FROM AUTHOR]

This study aims to produce dynamic noise maps based on a noise model and acoustic measurements. To do so, inverse modeling and joint state-parameter methods are proposed. These methods estimate the input parameters that optimize a given cost function calculated with the resulting noise map and the noise observations. The accuracy of these two methods is compared with a noise map generated with a meta-model and with a classical data assimilation method called best linear unbiased estimator. The accuracy of the data assimilation processes is evaluated using a "leave-one-out" cross-validation method. The most accurate noise map is generated computing a joint state-parameter estimation algorithm without a priori knowledge about traffic and weather and shows a reduction of approximately 26% in the root mean square error from 3.5 to 2.6 dB compared to the reference meta-model noise map with 16 microphones over an area of 3 km2. [ABSTRACT FROM AUTHOR]

Urban noise mapping generally consists of simulating the emission and attenuation of noise in an area by following rules such as common noise assessment methods. The computational cost makes these models unsuitable for applications such as uncertainty quantification, where thousands of simulations may be required. One solution is to replace the model with a meta-model that reproduces the expected noise levels with highly reduced computational costs. The strategy is to generate the meta-model in three steps. The first step is to generate a training sample exploring the large dimension model's inputs set. The second step is to reduce the dimension of the outputs. In the third step, statistical interpolators are defined between the projected values of the training sample over the reduced space of the outputs. Radial basis functions or kriging are used as interpolators. The meta-model was built using the open source software NoiseModelling. This study compares the proximity of the meta-model outputs to the model outputs against the reduced basis, the class of the kriging covariance function, and the training sample size. Simulations using the meta-model are more than 10 000 times faster than the model while maintaining the main behavior. [ABSTRACT FROM AUTHOR]

Numerical simulations of wildfire spread can provide support in deciding firefighting actions but their predictive performance is challenged by the uncertainty of model inputs stemming from weather forecasts, fuel parameterisation and other fire characteristics. In this study, we assign probability distributions to the inputs and propagate the uncertainty by running hundreds of Monte Carlo simulations. The ensemble of simulations is summarised via a burn probability map whose evaluation based on the corresponding observed burned surface is not obvious. We define several properties and introduce probabilistic scores that are common in meteorological applications. Based on these elements, we evaluate the predictive performance of our ensembles for seven fires that occurred in Corsica from mid-2017 to early 2018. We obtain fair performance in some of the cases but accuracy and reliability of the forecasts can be improved. The ensemble generation can be accomplished in a reasonable amount of time and could be used in an operational context provided that sufficient computational resources are available. The proposed probabilistic scores are also appropriate in a calibration process to improve the ensembles. We propose a new method to generate an ensemble of wildfire simulations to predict fire spread. Ensembles are evaluated with probabilistic scores that are new to the wildlfire community. Application to seven fires supports that the method may be used with the limited data availability typical of an operational context, provided that we can access sufficient computational resources. [ABSTRACT FROM AUTHOR]

Noise maps are a key asset in the elaboration of urban noise mitigation policies. However, simulation-based noise maps are subject to high uncertainties, and the estimation of population exposition to noise pollution generally relies on static averages over an extended period of time. This paper introduces a method to produce hourly noise maps based on temporally averaged simulation maps and mobile phone audio recordings. The data assimilation method produces an analysis noise map which is the so-called best linear unbiased estimator: it merges the simulated map and the measurements based on respective uncertainties so that the analysis map has minimum error variance. The method is illustrated through a neighborhood-wide experiment. A systematic study of the errors associated with both the simulation map and the observations (measurement error, temporal representativeness error, location error) is carried out. Two LA eq , 1 h maps are produced, corresponding, respectively, to a morning and an evening time slot. The analysis maps achieve a reduction of at least 25% of root-mean-square error. The a posteriori error variance of the maps are generally around 50% of the a priori error variance in the vicinity of the observed locations. [ABSTRACT FROM AUTHOR]

Network-based sound monitoring systems are deployed in various cities over the world and mobile applications allowing participatory sensing are now common. Nevertheless, the sparseness of the collected measurements, either in space or in time, complicates the production of sound maps. This paper describes the results of a measurement campaign that has been conducted in order to test different spatial interpolation strategies for producing sound maps. Mobile measurements have been performed while walking multiple times in every street of the XIIIth district of Paris. By adaptively constructing a noise map on the basis of these measurements, the role of the density of observations and the performance of four different interpolation strategies is investigated. Ordinary and universal Kriging methods are assessed, as well as the effect of using an alternative definition of the distance between observation locations, which takes the topology of the road network into account. The results show that a high density of observation points is necessary to obtain an interpolated sound map close to the reference map. [ABSTRACT FROM AUTHOR]

The increasing number and quality of sensors integrated in mobile phones have paved the way for sensing schemes driven by city dwellers. The sensing quality can drastically depend on the mobile phone, and appropriate calibration strategies are needed. This paper evaluates the quality of noise measurements acquired by a variety of Android phones. The Ambiciti application was developed so as to acquire a larger control over the acquisition process. Pink and narrowband noises were used to evaluate the phones' accuracy at levels ranging from background noise to 90 dB(A) inside the lab. Conclusions of this evaluation lead to the proposition of a calibration strategy that has been embedded in Ambiciti and applied to more than 50 devices during public events. A performance analysis addressed the range, accuracy, precision, and reproducibility of measurements. After identification and removal of a bias, the measurement error standard deviation is below 1.2 dB(A) within a wide range of noise levels [45 to 75 dB(A)], for 12 out of 15 phones calibrated in the lab. In the perspective of citizens-driven noise sensing, in situ experiments were carried out, while additional tests helped to produce recommendations regarding the sensing context (grip, orientation, moving speed, mitigation, frictions, wind). [ABSTRACT FROM AUTHOR]

This paper provides a formal mathematical representation of a wildfire simulation, reviews the most common scoring methods using this formalism, and proposes new methods that are explicitly designed to evaluate a forest fire simulation from ignition to extinction. These scoring or agreement methods are tested with synthetic cases in order to expose strengths and weaknesses, and with more complex fire simulations using real observations. An implementation of the methods is provided as well as an overview of the software package. The paper stresses the importance of scores that can evaluate the dynamics of a simulation, as opposed to methods relying on snapshots of the burned surfaces computed by the model. The two new methods, arrival time agreement and shape agreement, take into account the dynamics of the simulation between observation times. Although no scoring method is able to perfectly synthesise a simulation error in a single number, the analysis of the scores obtained on idealised and real simulations provides insights into the advantages of these methods for the evaluation of fire dynamics. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents the assimilation of satellite NO2 observations into a chemistry-transport model (CTM). The NO2 columns from Ozone Monitoring Instrument (OMI) aboard NASA Aura satellite are used during November–December 2005. These satellite observations are assimilated in an air quality model from Polyphemus, in order to better forecast NO2 in Europe. The optimal-interpolation method is applied to produce analyzed columns, and these analyzed columns are mapped to model concentrations assuming that the model vertical profile is perfect. Good consistency is seen in the comparisons of model simulations, satellite data and ground observations before assimilation. The model results with and without assimilation are then compared with ground observations for evaluating the assimilation effects. It is found that the assimilation can improve the NO2 forecasts by reducing their discrepancies against ground observations, indicating a better NO2 forecast obtained with OMI observations. Such improvements are seen in a cold season rather than in a warm one probably due to the longer lifetime of NOx and the initial condition changes having more impacts in winter. This also suggests that the assimilation of the short-lived species like NO2 is a complicate problem. [ABSTRACT FROM AUTHOR]

Abstract: We propose a method to evaluate the detection abilities of networks used for protection purposes. Such networks are designed for the detection of nuclear, biological or gaseous emissions, without constraint on the source location. Their assigned goal is to have the best chance to detect a threatening emission located anywhere in the vicinity of a domain to protect. Two sensors siting applications are addressed: sensors placed in the surroundings of a facility to protect, or sensors carried by people scattered within a small area. A network protection ability is related both to its detection scope, and to its response time. To assess the performance of such networks, two statistical indicators are therefore designed: the detection probability, computed on a large number of possible source locations, and the saturation time, which is the time when the maximum detection probability has been reached. Simulations are then carried out with the Polyphemus air quality modeling system for many emission scenarios, including 961 possible source locations, various emitted species, and a few representative meteorological situations. This allows to assess the performance of single sensors as well as full networks, and their sensitivity to parameters like meteorological conditions and source characteristics. The emitted quantity and meteorological dispersion are found to be important parameters, whereas the species type does not significantly influence the results. Two network design methods are considered: (1) networks composed of a given number of the “best” sensors according to an indicator, and (2) sensors placed in circles around the protected domain. The networks built with respect to the detection probability show good results with a limited number of sensors, while the saturation time is not reliable enough to build networks. The networks based on circles also show a good performance in the studied cases, provided there is a sufficient number of sensors. [Copyright &y& Elsevier]

This paper describes a three-dimensional chemistry transport model, POLAIR, with a special focus on numerical aspects. POLAIR is a fully modular eulerian model. Several different chemical mechanisms are available, which can deal with photochemistry (Racm, Radm, etc.), continental impact (e.g. passive transport), mercury, aerosols, etc. POLAIR is designed to enable simulations from regional scales to continental scales. A few simulations at those scales have been conducted to assess and improve the code. Beyond forward simulations, inverse modelling and data assimilation can be performed, thanks to the tangent linear and adjoint versions of POLAIR, which are available through automatic differentiation. [ABSTRACT FROM AUTHOR]

Abstract: The review of major 3-D global and regional real-time air quality forecasting (RT-AQF) models in Part I identifies several areas of improvement in meteorological forecasts, chemical inputs, and model treatments of atmospheric physical, dynamic, and chemical processes. Part II highlights several recent scientific advances in some of these areas that can be incorporated into RT-AQF models to address model deficiencies and improve forecast accuracies. Current major numerical, statistical, and computational techniques to improve forecasting skills are assessed. These include bias adjustment techniques to correct biases in forecast products, chemical data assimilation techniques for improving chemical initial and boundary conditions as well as emissions, and ensemble forecasting approaches to quantify the uncertainties of the forecasts. Several case applications of current 3-D RT-AQF models with the state-of-the-science model treatments, a detailed urban process module, and an advanced combined ensemble/data assimilation technique are presented to illustrate current model skills and capabilities. Major technical challenges and research priorities are provided. A new generation of comprehensive RT-AQF model systems, to emerge in the coming decades, will be based on state-of-the-science 3-D RT-AQF models, supplemented with efficient data assimilation techniques and sophisticated statistical models, and supported with modern numerical/computational technologies and a suite of real-time observational data from all platforms. [Copyright &y& Elsevier]

Abstract: Real-time air quality forecasting (RT-AQF), a new discipline of the atmospheric sciences, represents one of the most far-reaching development and practical applications of science and engineering, poses unprecedented scientific, technical, and computational challenges, and generates significant opportunities for science dissemination and community participations. This two-part review provides a comprehensive assessment of the history, current status, major research and outreach challenges, and future directions of RT-AQF, with a focus on the application and improvement of three-dimensional (3-D) deterministic RT-AQF models. In Part I, major milestones in the history of RT-AQF are reviewed. The fundamentals of RT-AQF are introduced. Various RT-AQF techniques with varying degrees of sophistication and skills are described comparatively. Among all techniques, 3-D RT-AQF models with online-coupled meteorology–chemistry and their transitions from mesoscale to unified model systems across scales represent a significant advancement and would greatly enhance understanding of the underlying complex interplay of meteorology, emission, and chemistry from global to urban scales in the real atmosphere. Current major 3-D global and regional RT-AQF models in the world are reviewed in terms of model systems, component models, application scales, model inputs, forecast products, horizontal grid resolutions, and model treatments of chemistry and aerosol processes. An important trend of such models is their coupling with an urban model or a computational fluid dynamic model for urban/local scale applications at 1 km or less and with an exposure model to provide real-time public health assessment and exposure predictions. Evaluation protocols are described along with examinations of current forecasting skills and areas with large biases of major RT-AQF models. [Copyright &y& Elsevier]

Abstract: Atmospheric dispersion models are usually off-line coupled to mesoscale meteorological models. This may generate the loss of mass consistency, defined as the conservation of uniform mixing ratio. We investigate in this short paper the impact of the resulting mass consistency errors. Three methods based on the renormalization of density, on fluxes computed with mass mixing ratio and on the adjustment of the vertical velocity, respectively, are aimed at reducing the mass consistency errors. They are benchmarked and applied to two test cases: air quality modeling over Europe for summer 2001 (typical of reactive dispersion) and simulation of the Chernobyl accident (typical of passive dispersion). Our tests indicate the differences between the passive and the reactive cases. The investigation of the spatial patterns (especially of the vertical distribution) discriminates the method based on the adjustment of the vertical velocity. Indeed, this method suffers from the enhancement of the numerical diffusion (illustrated in the passive case) and from the modification of the escape flux (for ozone). [Copyright &y& Elsevier]