Weather station

About: Weather station is a research topic. Over the lifetime, 1789 publications have been published within this topic receiving 42864 citations. The topic is also known as: meteorological station & meteorological observation post.

Smart Weather Forecasting Using Machine Learning: A Case Study in Tennessee.

Abstract: Traditionally, weather predictions are performed with the help of large complex models of physics, which utilize different atmospheric conditions over a long period of time. These conditions are often unstable because of perturbations of the weather system, causing the models to provide inaccurate forecasts. The models are generally run on hundreds of nodes in a large High Performance Computing (HPC) environment which consumes a large amount of energy. In this paper, we present a weather prediction technique that utilizes historical data from multiple weather stations to train simple machine learning models, which can provide usable forecasts about certain weather conditions for the near future within a very short period of time. The models can be run on much less resource intensive environments. The evaluation results show that the accuracy of the models is good enough to be used alongside the current state-of-the-art techniques. Furthermore, we show that it is beneficial to leverage the weather station data from multiple neighboring areas over the data of only the area for which weather forecasting is being performed.

A Predictive Model for Solar Photovoltaic Power using the Levenberg-Marquardt and Bayesian Regularization Algorithms and Real-Time Weather Data

Abstract: The stability of power production in photovoltaics (PV) power plants is an important issue for large-scale gridconnected systems This is because it affects the control and operation of the electrical grid An efficient forecasting model is proposed in this paper to predict the next-day solar photovoltaic power using the Levenberg-Marquardt (LM) and Bayesian Regularization (BR) algorithms and real-time weather data The correlations between the global solar irradiance, temperature, solar photovoltaic power, and the time of the year were studied to extract the knowledge from the available historical data for the purpose of developing a real-time prediction system The solar PV generated power data were extracted from the power plant installed on-top of the faculty of engineering building at Applied Science Private University (ASU), Amman, Jordan and weather data with real-time records were measured by ASU weather station at the same university campus Huge amounts of training, validation, and testing experiments were carried out on the available records to optimize the Neural Networks (NN) configurations and compare the performance of the LM and BR algorithms with different sets and combinations of weather data Promising results were obtained with an excellent realtime overall performance for next-day forecasting with a Root Mean Square Error (RMSE) value of 00706 using the Bayesian regularization algorithm with 28 hidden layers and all weather inputs The Levenberg-Marquardt algorithm provided a 00753 RMSE using 23 hidden layers for the same set of learning inputs This research shows that the Bayesian regularization algorithm outperforms the reported real-time prediction systems for the PV power production

July temperatures in kashmir and ladakh, india: comparisons of observations and general circulation model simulations

Abstract: The Himalayan mountain region represents a challenging laboratory to refine general circulation model simulations and to compare them with climate observations at regional and local geographical scales. This paper includes an analysis of July tem- peratures in the Great Himalaya Range of northern India. Five sources of data are analyzed: (1) coarse-scale model simulations for present climate (1 x CO2 runs), (2) regression model predictions of temperature versus elevation (after Lambert and Chitrakar, 1989), (3) spatially-interpolated observed data for the region (after Legates and Willmott, 1990), (4) individual station records in the region, and (5) local-scale field data taken in summer, 1988. As expected, July temperature simulations from general circulation models tend to generalize temperature/elevation relationships and do not appear to account for landscape surface feedbacks on air temperatures at a regional and local scale. Cautious use of climate model predictions of temperature and moisture for climate change scenario analyses is strongly encouraged. For mountain areas, field verification and weather station analyses for regional and local scales are essential for interpreting results from general circulation models.

Use of meteorological radar to estimate leaf wetness as data input for application of territorial epidemiological model (downy mildew—Plasmopara viticola)

Abstract: The grapevine downy mildew (Plasmopara viticola) represents the most important disease of the grapevine in Friuli-Venezia Giulia Region (Italy). The development of this disease depends from the meteorological conditions and particularly by air humidity, rain and leaf wetness (LW here after). Forecast models can help the technicians of the extension services to predict the timing and the best technique to use in operative programs. Unfortunately these models require data, coming from meteorological stations which are often variable in space (e.g. rain, leaf wetness) and hardly spatializable. In the first part of this work, a case study is presented to show the great difference between maps of daily rain duration, obtained by radar, and those created by spatialization of data and obtained by weather stations. Then the possibility to use the radar rain maps appears very interesting to estimate LW over a large area. LW and daily rain measurements, obtained by 14 weather stations of Friuli-Venezia Giulia plain (Italy), are compared with rain maps obtained by polarimetric radar GPM-500 placed in Fossalon di Grado (Friuli-Venezia Giulia, Italy). The reference measurements are made during two periods: from 1/4/2000 to 30/9/2000 and from 1/4/2001 to 30/9/2001. From radar maps rain measurements estimated are extracted above each weather station and these data are integrated for every hour. These radar data of hourly rain are compared to the corresponding measurementes of LW and rain obtained by weather stations. From this analysis it appears that there is a good correlation between the number of rain hours estimated by radar and the number of LW hours measured by stations: in the observed cases, the error found is lower than 2%; then radar has a good precision to estimate LW due to rain. Therefore the use of Radar is foretold to give meteorological inputs in simulation models that can work to evaluate the development of fungal diseases. In the second part a model to daily display the infections of downy mildew all over the plain of Friuli-Venezia Giulia is described. Elements of this model are: • the creation of a daily grid of rain estimated by a meteorological polarimetric radar (GPM-500) located in Fossalon di Grado; • the creation of a daily grid of temperature, air humidity, solar radiation, wind speed using data coming from 14 synoptic meteorological stations located in the plain of the region; • the creation of a daily grid of leaf-wetness computed using the SWEB model for the data measured or estimated in the previous point; • the estimation of the P. viticola infection level using a forecast model (in this case Goidanich). • The daily graphical output in every point of the grid is: number of the actual cycles of P. viticola infections; number of days required for the next infection; annual amount of infective cycles. The model has been checked in a limited area (about 200 km2) with an high presence of grapevine (DOC Aquileia) in the period from may 2000 to september 2000. In this area the data from eight meteorological automatic stations installed for diseases–defence purposes are used as test, and has been calculated the downy mildew infections using the Goidanich model. These grids have been compared with the maps calculated using data coming from meteorological radar and synoptic stations. The outputs are similar and the proposed method can be considered a good approach in the operative use of radar in the crop protection.

South Cascade Glacier mass balance, 1935-2006

Abstract: A simple model uses daily observations of precipitation and temperature at a nearby weather station to estimate glacier-average seasonal mass-balance components at South Cascade Glacier, Washington, USA, from 1935, 24 years before measurements began at the glacier. This is 13 years earlier than measurements that can be derived using the NCEP-NCAR reanalysis database (begins 1 January 1948). Although the model's error in estimating winter balance and summer balance over 1959-2006 is greater than that of a model using the reanalysis database, its error in estimating net balance is comparable. The model uses an empirically determined precipita- tion ratio between the station and the glacier, and a seasonally varying temperature lapse rate de- termined from 9 years of measurements at the glacier. Temperature is used with a degree-day for- mulation to estimate ablation and to partition precipitation between rain and snow for estimating accumulation. Both processes are assumed to exist throughout the year, and model results are compared seasonally with adjusted observations of winter and summer balances. The published mass-balance series is adjusted to a constant-topography (1970) series in an attempt to remove the influence of changing topography on the glacier's response to climate. The reconstructed values prior to 1959 are also with respect to the 1970 glacier topography. Because precipitation is measured at the weather station, rather than being inferred from other meteorological variables, it enables us to distinguish more accurately between wet-day and dry-day conditions, including vertical lapse rates of temperature.