Marija Zlata Božnar

Short-term fine resolution WRF forecast data validation in complex terrain in Slovenia

For the air pollution modelling studies over highly complex terrain, vertical wind profiles are essential. In the article we present evaluation about using the WRF model as the source of wind profile information. We tested WRF’s one day short-term forecasts at 4 km and half hour resolution running every day to obtain 3D meteorological fields and compared these data with the different meteorological stations. The results show an inadequate agreement with ground level meteorological stations, especially in basins and valleys, and a better agreement with stations situated at the top of hills and with a tower station. A novel approach to the terrain complexity characterisation of the area under examination is defined – ‘height and length of Topographic complexity, hlTc’.

Improvement of Air Pollution Forecasting Models Using Feature Determination and Pattern Selection Strategies

Air pollution forecasting models are a helpful tool for controlling pollution around sources such as large thermal power plants. Recently we developed a neural network - based, short-term SO2 pollution forecasting model for measuring sites around the Sostanj Thermal Power Plant. The most important problems that should be solved in order to improve the model performance are feature determination and pattern selection. We developed several methods to solve these two problems.

Neural Networks Predict Pollution

Air pollution is a very big problem in Slovenia. The greatest pollutants are two big thermal power plants (TPP-s) which are placed in the valleys near the coal mines. The coal has a very big percentage of the sulphur (up to 2%) and the TPP-s do not have wet desulphurisation. The results of this are the episodes of very high air pollution in the vicinity of the TPP-s. The local government intends to omit this episodes by forcing the TPP-s to reduce the power significantly (from 30% up to 70%). Such quick reduction of the TPP power is certainly an economical and technological problem.

Improving of local ozone forecasting by integrated models

This paper discuss the problem of forecasting the maximum ozone concentrations in urban microlocations, where reliable alerting of the local population when thresholds have been surpassed is necessary. To improve the forecast, the methodology of integrated models is proposed. The model is based on multilayer perceptron neural networks that use as inputs all available information from QualeAria air-quality model, WRF numerical weather prediction model and onsite measurements of meteorology and air pollution. While air-quality and meteorological models cover large geographical 3-dimensional space, their local resolution is often not satisfactory. On the other hand, empirical methods have the advantage of good local forecasts. In this paper, integrated models are used for improved 1-day-ahead forecasting of the maximum hourly value of ozone within each day for representative locations in Slovenia. The WRF meteorological model is used for forecasting meteorological variables and the QualeAria air-quality model for gas concentrations. Their predictions, together with measurements from ground stations, are used as inputs to a neural network. The model validation results show that integrated models noticeably improve ozone forecasts and provide better alert systems.

Zasavje canyon regional online air pollution modelling system in highly complex terrain – description and validation

Zasavje is an industrial region in Slovenia located along the Sava River’s steep canyon where PM10 air pollution is a major problem. In the paper a national project with the title ‘Prognostic and diagnostic integrated regional air pollution modelling system’ is described where it is shown that such a project can significantly contribute to the proper understanding of air pollution in smaller regions with a very complex topography. To achieve online efficiency some new methods to obtain high resolution short-range meteorological fields derived from meso-scale models have been developed and the implementation of the advanced Lagrangian model’s acceleration techniques and novel approaches for the whole system integration is presented. The project’s test-bed was established as a novel approach to the overall treatment of the scientific – applicative project goal.

Fireworks air pollution in Slovenia

Over the ‘Christmas–New Year’ period 12/2010–01/2011, PM10 air pollution measurements on most of the automatic measuring stations in Slovenia reached extreme values of up to 300 μg/m 3 . Analysis shows a very close correlation with the timings of the fireworks displays in the areas. This paper shows the other side of this cultural enjoyment – the effect on PM10 air pollution. As the correlation with extreme values is so drastic all over Slovenia, the question arises if these events do need some further regulation to prevent harmful health effects. Can the locations for fireworks displays be selected according to the knowledge about air pollution mechanisms in the area? Some practical examples and analyses of the situations are given for the Zasavje region in Slovenia, where two automatic stations measured extreme peaks of PM10 pollution during the period under examination.

Artificial Neural Network-Based Environmental Models

Artificial neural network-based air pollution prediction models have become very popular. The paper describes feature determination and pattern selection strategies that help to improve significantly the performance of neural-network based models.

Environmental impact assessment of a new thermal power plant Šoštanj Block 6 in highly complex terrain

Slovenia is starting an important investment – Block 6 of the thermal power plant in Sostanj located in the Velenje basin and characterised by very complex terrain. Air pollution examination as part of the environmental impact assessment is an interesting example of usage of the modern Lagrangian particle model Spray coupled with the diagnostic mass consistent wind field model Minerve for the determination of the appropriate height for the combined cooling tower. In addition, a separate study was done on a larger domain to evaluate possibilities of trans-boundary pollution, as the location is only 25 km from the Austrian border.

Selection of the data time interval for the prediction of maximum ozone concentrations

This paper highlights the problem of step-length selection for the one-step-ahead prediction of ozone called the data time interval. This is done using a case study-based comparison of two approaches for predicting the maximum daily values of tropospheric ozone. The first approach is the 1-day-ahead prediction and the second is the prediction of the maximum values based on a multi-step-ahead iteration of 1-h predictions. Gaussian process modelling is utilised for this comparison. In particular, evolving Gaussian-process models are used that update on-line with the incoming measurement data. These sorts of models have been successfully used in the past for the prediction of ozone pollution. This paper contributes an assessment of the way that the maximum ozone values are predicted. A comparison of the daily maximum ozone values forecasted by a model based on 1-day-ahead predictions with those obtained by iterated 1-h-ahead predictions of the ozone with predictions at predetermined hours of the day is given. The forecast results are in favour of the on-line model based on hourly predictions when approaching closer to the real maximum values of ozone, and in favour of the daily predictions when they are made on a daily basis.

Validation of meteorological forecasts in fine spatial and temporal resolution produced as an input for dispersion models

In conditions of complex terrain, modelling of air pollutant dispersion still has a number of scientific challenges. Ideally, appropriate meteorological data should be available for modelling. Unfortunately, for many purposes, there is no time to carry out suitable measuring campaigns. Therefore the results of prognostic weather forecasts (NWP models) are being widely used. However, these models still have quite a few disadvantages when their results are used as input for dispersion models over complex terrain. This study presents the validation of the quality of the weather forecasts in the surroundings of the Nuclear Power Plant Krsko in Slovenia, an area with highly complex terrain and the resulting complex meteorological characteristics. The forecast is available for a horizontal resolution of 2 km and half hour temporal interval and seven days in advance. The predicted meteorological parameters are validated using the measured meteorological parameters.