Ozan Şenkal

Precipitable water modelling using artificial neural network in Çukurova region

Precipitable water (PW) is an important atmospheric variable for climate system calculation. Local monthly mean PW values were measured by daily radiosonde observations for the time period from 1990 to 2006. Artificial neural network (ANN) method was applied for modeling and prediction of mean precipitable water data in Çukurova region, south of Turkey. We applied Levenberg–Marquardt (LM) learning algorithm and logistic sigmoid transfer function in the network. In order to train our neural network we used data of Adana station, which are assumed to give a general idea about the precipitable water of Çukurova region. Thus, meteorological and geographical data (altitude, temperature, pressure, and humidity) were used in the input layer of the network for Çukurova region. Precipitable water was the output. Correlation coefficient (R2) between the predicted and measured values for monthly mean daily sum with LM method values was found to be 94.00% (training), 91.84% (testing), respectively. The findings revealed that the ANN-based prediction technique for estimating PW values is as effective as meteorological radiosonde observations. In addition, the results suggest that ANN method values be used so as to predict the precipitable water.

The Estimation of Solar Radiation for Different Time Periods

Abstract In this study, the method of Becker and Li was proposed for the estimation of monthly global land surface temperature values from meteorological satellite (NOAA-AVHRR) data. This study introduces generalized regression neural network for the estimation of solar radiation. In order to train the neural network, meteorological satellite and geographical data for the period from 2002 for short term (Adana) and 1998–2002 for long term (İzmir) in Turkey was used. Meteorological satellite and geographical data (latitude, longitude, altitude, month, and mean land surface temperature) are used in the input layer of the network. Solar radiation is the output. Root mean squared and correlation coefficient data between estimated and ground values are found with artificial neural networks values. These values have been found to be 0.0144 MJm−2 and 99.75% (short term) and 0.1381 MJm−2 and 99.26% (long term), respectively. In recent studies, there are some effective techniques about prediction solar radiation data, which is useful to the designers of solar energy systems. Nevertheless, there is no study about the prediction of solar radiation, which has used the artificial neural networks method with land surface temperature data provided from meteorological satellite data.

Comparison of a new algorithm with the supervised classifications

In this study, a new classification algorithm in which only the selected pixels have been attempted to be classified (selected pixels classification: SPC) has been introduced and compared with the well known supervised classification methods such as maximum likelihood, minimum distance, nearest neighbour and condensed nearest neighbour. To examine the algorithm, Landsat Thematic Mapper (TM) data have been used to classify the crop cover in the selected region. It is clearly demonstrated that the SPC method has the higher accuracy with comparable CPU times.

Solar radiation and precipitable water modeling for Turkey using artificial neural networks

AbstractArtificial neural network (ANN) method was applied for modeling and prediction of mean precipitable water and solar radiation in a given location and given date (month), given altitude, temperature, pressure and humidity in Turkey (26–45ºE and 36–42ºN) during the period of 2000–2002. Resilient Propagation (RP) learning algorithms and logistic sigmoid transfer function were used in the network. To train the network, meteorological measurements taken by the Turkish State Meteorological Service (TSMS) and Wyoming University for the period from 2000 to 2002 from five stations distributed in Turkey were used as training data. Data from years (2000 and 2001) were used for training, while the year 2002 was used for testing and validating the model. The RP algorithm were first used for determination of the precipitable water and subsequently, computation of the solar radiation, in these stations Root Mean Square Error (RMSE) between the estimated and measured values for monthly mean daily sum for precipitable water and solar radiation values have been found as 0.0062 gr/cm2 and 0.0603 MJ/m2 (training cities), 0.5652 gr/cm2 and 3.2810 MJ/m2 (testing cities), respectively.

A Comparison of Two Solar Radiation Models Using Artificial Neural Networks and Remote Sensing in Turkey

This study introduces artificial neural networks for the estimation of solar radiation using model 1 (latitude, longitude, altitude, month, and meteorological land surface temperature) and model 2 (latitude, longitude, altitude, month, and satellite land surface temperature) data in Turkey. Price method was used for the estimation of land surface temperature values. Scale conjugate gradiant learning algorithms and logistic sigmoid transfer function were used in the network. R 2 with model 1 and model 2 values have been found to be 96.93 and 97.24% (training stations), 80.41 and 82.37% (testing stations), respectively. These results are sufficient to predict the solar radiation.

Estimation of the vapour pressure deficit using NOAA-AVHRR data

In this study, the calculation of vapour pressure deficit (VPD) using the National Oceanic and Atmospheric Administration Advanced Very High Resolution Radiometer (NOAA/AVHRR) satellite data set is shown. Twenty-four NOAA/AVHRR data images were arranged and turned to account for both VPD and land surface temperature (LST), which was necessary to calculate the VPD. The most accurate LST values were obtained from the Ulivieri et al. split-window algorithm with a root mean square error (RMSE) of 2.7 K, whereas the VPD values were retrieved with an RMSE of 6 mb. Furthermore, the VPD value was calculated on an average monthly basis and its correlation coefficient was found to be 0.991, while the RMSE value was calculated to be 2.67 mb. As a result, VPD can be used in studies that examine plants (germination, growth, and harvest), controlling illness outbreak, drought determination, and evapotranspiration.