A sensitivity study of the WRF model in wind simulation for an area of high wind energy

The performance of the Weather Research and Forecast (WRF) model in wind simulation was evaluated under different numerical and physical options for an area of Portugal, located in complex terrain and characterized by its significant wind energy resource. The grid nudging and integration time of the simulations were the tested numerical options. Since the goal is to simulate the near-surface wind, the physical parameterization schemes regarding the boundary layer were the ones under evaluation. Also, the influences of the local terrain complexity and simulation domain resolution on the model results were also studied. Data from three wind measuring stations located within the chosen area were compared with the model results, in terms of Root Mean Square Error, Standard Deviation Error and Bias. Wind speed histograms, occurrences and energy wind roses were also used for model evaluation. Globally, the model accurately reproduced the local wind regime, despite a significant underestimation of the wind speed. The wind direction is reasonably simulated by the model especially in wind regimes where there is a clear dominant sector, but in the presence of low wind speeds the characterization of the wind direction (observed and simulated) is very subjective and led to higher deviations between simulations and observations. Within the tested options, results show that the use of grid nudging in simulations that should not exceed an integration time of 2 days is the best numerical configuration, and the parameterization set composed by the physical schemes MM5–Yonsei University–Noah are the most suitable for this site. Results were poorer in sites with higher terrain complexity, mainly due to limitations of the terrain data supplied to the model. The increase of the simulation domain resolution alone is not enough to significantly improve the model performance. Results suggest that error minimization in the wind simulation can be achieved by testing and choosing a suitable numerical and physical configuration for the region of interest together with the use of high resolution terrain data, if available.

Characterization of soils in an area of prescribed fire

The prescribed fire is a technique that is often used, it has several advantages. Pedological and hydropedological techniques were tested to assess the prescribed fire changes may cause in soils. This work was performed in Tresminas area (Vila Pouca de Aguiar, Northern Portugal), during February and March 2011. In the present study we applied several techniques. For the field sampling was followed the ISO 10381-1[1], ISO 10381-2[2], and FAO rules [3], as well as were used a grid with 17 points for measuring the soil parameters. During the fire, we have tried to check, with the assistance of the Portuguese Forestry Authority, some important parameters such as, the propagation speed, the size of the flame front and the intensity of energy emitted per unit area. Before the fire, was collected carefully soil disturbed and undisturbed samples for laboratory analysis, and measured soil water content; we also have placed four sets of thermocouples for measuring soil temperature. After the fire, were collected the thermocouples and new soil samples; the water content were measured in the soil and collected ashes. In the laboratory, after preparing and sieving the samples, were determined the soil particle size. The soil pH and electrical conductivity in water was also determined. The total carbon (TC) and inorganic carbon (IC)[4] was measured by a Shimadzu TOC-Vcsn. The water content in soil has not varied significantly before and after the fire, as well as soil pH and soil electrical conductivity. The TC and IC did not change, which was expected, since the fire not overcome the 200° C. Through the various parameters, we determined that the prescribed fire didn’t affect the soil. The low temperature of the fire and its rapid implementation that lead to the possible adverse effects caused by the wild fire didn’t occurred.

Measurements of environmental background radiation in the surrounding area of a coal-fired power plant

Certain materials used and produced in a wide range of non-nuclear industries contain enhanced activity concentrations of natural radionuclides. In particular, electricity production from coal is one of the major sources of increased exposure to man from enhanced naturally occurring materials. Over the past decades there has been some discussion about the elevated natural background radiation in the area near coal-fired power plants due to high uranium and thorium content present in coal. This work describes the methodology developed to assess the radiological impact due to natural radiation background increasing levels, potentially originated by a coal-fired power plant’s operation. Gamma radiation measurements have been done with two different instruments: a scintillometer (SPP2 NF, Saphymo) and a gamma ray spectrometer with energy discrimination (Falcon 5000, Canberra). A total of 40 relevant sampling points were established at locations within 20 km from the power plant: 15 urban and 25 suburban measured stations. The highest values were measured at the sampling points near to the power plant and those located in the area within the 6 and 20 km from the stacks. This may be explained by the presence of a huge coal pile (1.3 million tons) located near the stacks contributing to the dispersion of unburned coal and, on the other hand, the height of the stacks (225 m) which may influence ash’s dispersion up to a distance of 20 km. In situ gamma radiation measurements with energy discrimination identified natural emitting nuclides as well as their decay products (212Pb, 214Pb, 226Ra 232Th, 228Ac, 234Th 234Pa, 235U, etc.). This work has been primarily done to in order to assess the impact of a coal-fired power plant operation on the background radiation level in the surrounding area. According to the results, an increase or at least an influence has been identified both qualitatively and quantitatively.