Agricultural Meteorology
B. Mirkamandar; Seied Hosein Sanaei-Nejad; H. Rezaee- Pazhand; M. Farzandi
Abstract
Introduction: The behavior of daily changes of relative humidity is quite variable. We first draw the curve of this variable on a normal day. And it can be seen that the distribution of this variable is not normal. The curve of this variable is a skewed curve to the right. Therefore, the equal coefficients ...
Read More
Introduction: The behavior of daily changes of relative humidity is quite variable. We first draw the curve of this variable on a normal day. And it can be seen that the distribution of this variable is not normal. The curve of this variable is a skewed curve to the right. Therefore, the equal coefficients could be used only as an approximation for estimating the daily average of relative humidity. Climatic conditions of the meteorological stations are also another parameter to be considered. This research presents a new method for estimating the daily average of relative humidity in three climatic regions of Iran. The patterns for the sample stations in each climatic region were presented separately. Materials and Methods: E. Eccel (2012) developed an algorithm to simulate the relative humidity of the minimum daily temperature in 23 weather stations in the ALP region of Italy. In this research, the base pattern was calibrated by temperature and precipitation measurement. Ephrath, et al. (1996) developed a method for the calculation of diurnal patterns of air temperature, wind speed, global radiation, and relative humidity from available daily data. During the day, the air temperature was calculated by: (1) (2) Where S(t): Dimensionless function of time, DL: Day Length h, LSH: the time of maximum solar high h, ta: current air temperature, P: the delay in air Tmax with respect to LSH h. Farzandi, et al. (2012) presented more accurate patterns for estimating daily relative humidity from the humidity of Iranian local standard hours and daily precipitation variables, the minimum, maximum, and average daily temperature in coastal regions. The purpose was to present linear and nonlinear patterns of daily relative humidity separately for different months (12 patterns) and annually in coastal regions (the Caspian Sea, the Persian Gulf, and the Oman Sea).Mirkamandar, et al. (2020) modified new patterns of diurnal temperature based on climatically clustering in Iran. The final pattern has an interception and new coefficients to estimate the daily average of temperature. Rezaee-Pazhand, et al. (2008) introduced new patterns for estimating the daily average temperature in arid and semiarid regions of Iran. The final pattern has an interception and new coefficients to estimate the daily average of temperature. (3) Veleva, et al. (1996) showed that the atmospheric temperature-humidity complex (T-HC) of sites located in a tropical humid climate cannot be well characterized by annual average values. Better information is given by the systematic study of daily changes of temperature (T) and relative humidity (RH), which can be modeled by linear and parabolic functions. Farzandi et al. (2011) divided Iran into three climatic clusters. Which were used in the present work. First, a classification that provides climatological clustering. This clustering was used the data of annual relative humidity, temperature, precipitation, altitude, range of temperature, evaporation, and three indices of Demartonne, Ivanov, and Thornthwaite. Iran was partitioned into three clusters i.e. coastal areas, mountainous range, arid and semi-arid zone. Several clustering methods were used and the around method was found to be the best. Cophenetic correlation coefficient and Silhouette width were validation indices. Homogeneity and Heterogeneity tests for each cluster were done by L-moments. The “R”, software packages were used for clustering and validation tests. Finally, a clustering map of Iran was prepared using “GIS”. The data of 149 synoptic stations were used for this analysis. Systematic sampling was done to select sample stations. The linear regression model was fitted after screening and data preparation. A model was presented for estimating the daily average temperature in each climatic region and sampling stations in each cluster. The best models were presented by reviewing the required statistics and analyzing the residuals. The calibration and comparison of the presented patterns in this paper with commonly applied models were undertaken to calculate the mean squared error. “SPSS.22” software was used for analysis. Results and Discussion: The coefficient of determination (R2) and the Fisher statistics showed that the patterns had a good ability to estimate the daily average of relative humidity. The daily average of relative humidity patterns confirmed an interception in the equations. Standardized coefficients showed that predictor variables were not weighted in all of the patterns. The mean squares errors were a measure of the applicability of patterns. The accuracy of the estimating daily average of relative humidity recommended models in three climates was confirmed by calculating the mean squared errors. The proposed patterns of this study had less error than the common patterns. Conclusion: The relative humidity at 3 pm was more effective in estimating the daily average. The independent assumption of the residual was confirmed with the acceptable value of Durbin-Watson statistics. The averages of the residuals in each pattern was zero. According to the graphs, stabilization of variance can be seen based on the residual on each pattern in each cluster. Proposed patterns were calculated according to mathematical principles. But the common patterns did not observe these mathematical principles. The mean squares errors (MSE) of proposed patterns were less than common patterns. Therefore, the patterns presented in this study are more powerful than common patterns.
B. Mirkamandar; Seied Hosein Sanaei-Nejad; H. Rezaee-Pazhand; M. Farzandi
Abstract
Introduction: The behavior of daily changes in temperature is not straightforward. We first drew the curve of this variable on a normal day. It can be seen that the distribution of this variable was not normal. The curve of this variable was a skewed curve to the right. Therefore, the equal coefficients ...
Read More
Introduction: The behavior of daily changes in temperature is not straightforward. We first drew the curve of this variable on a normal day. It can be seen that the distribution of this variable was not normal. The curve of this variable was a skewed curve to the right. Therefore, the equal coefficients could be used only as approximation for estimating daily average temperature. Climatic conditions of the meteorological stations were also another parameter to be considered. This research presents a new method for estimating daily average of temperature in three climatic regions of Iran. The patterns for the sample stations in each climatic region were presented separately. Materials and Methods: E. Eccel (2012) developed algorithms to simulate the relative humidity of the minimum daily temperature in 23 weather stations in the ALP region of Italy. In this research, the base pattern was calibrated by temperature and precipitation measurement. Ephrath, et al. (1996) developed a method for the calculation of diurnal patterns of air temperature, wind speed, global radiation and relative humidity from available daily data. During the day, air temperature was calculated by: (1) (2) where S (t): Dimensionless function of time, DL: Day Length h, LSH: the time of maximum solar high h, ta: Current air Temperature, P: the delay in air Tmax with respect to LSH h. Farzandi, et al. (2012) presented more accurate patterns for estimating daily relative humidity from humidity of Iranian local standard hours and daily precipitation variables, the minimum, maximum and average daily temperature in coastal regions. The purpose was to present linear and nonlinear patterns of daily relative humidity separately for different months (12 patterns) and annually in coastal regions (the Caspian Sea, the Persian Gulf, and the Oman Sea). Rezaee-Pazhand, et al. (2008) introduced new patterns for estimating daily average temperature in arid and semiarid regions of Iran. Final pattern has interception and new coefficients for estimate daily average of temperature. (3) Veleva, et al. (1996) showed that the atmospheric temperature-humidity complex (T-HC) of sites located in a tropical humid climate cannot be well characterized by annual average values. Better information is given by the systematic study of daily changes of temperature (T) and relative humidity (RH), which can be modeled with linear and parabolic functions. Farzandi et al. (2011) divided Iran into three climatic clusters used in the present work. First a classification which provides climatological clustering. This clustering was used the data of annual relative humidity, temperature, precipitation, altitude, range of temperature, evaporation and three indices of De Martonne, Ivanov and Thornthwaite. Iran was partitioned into three clusters i.e. coastal areas, mountainous range and arid and semi-arid zone. Several clustering methods were used and around method was found to be the best. Cophenetic correlation coefficient and Silhouette width were validation indices. Homogeneity and Heterogeneity tests for each cluster were done by L-moments. The “R”, software packages were used for clustering and validation testes. Finally clustering map of Iran was prepared using “GIS”. The data of 149 synoptic stations were used for this analysis. Systematic sampling was done to select sample stations. The linear regression model was fitted after screening and data preparation. A model was presented for estimating daily average of temperature in each climatic region and sampling stations in each cluster. The best models were presented by reviewing the required statistics and analyzing the residuals. The calibration and comparison of the presented patterns in this paper with commonly applied models were undertaken to calculate the mean squared error. “SPSS.22” software was used for analysis. Results and Discussion: The coefficient of determination (R2) and the Fisher statistics show that the patterns have a good ability to estimate the daily average of temperature. The daily average temperature pattern confirmed an interception in the equations. Standardized coefficients showed that predictor variables were not weighted in all of the patterns. The average values of the residuals in each pattern was zero. According to the graphs, stabilization of variance can be seen based on the residual on each pattern in each cluster. The mean squared error is a measure of the applicability of patterns. The accuracy of the estimating daily average temperature by the recommended models in three climates was confirmed by calculating the mean squared error. The proposed patterns of this study had less error than common patterns. Thus, the patterns have a good ability to estimate daily average temperature. Conclusion: The maximum temperature in calculating daily average of temperature is more effective than the minimum temperature. The standardized coefficient (Beta) of the daily average temperature patterns in coastal cluster was 48.2% for the minimum temperature and 51.8% for the maximum temperature. The largest influence of the maximum temperature was 63.1% in mountainous cluster for estimating daily average temperature. Range of the interception in the equations was from -1.735 to 0.26. The independent assumption of the residual was confirmed with the acceptable value of Durbin-Watson statistics. The average of the residuals in each patterns was zero. According to the graphs stabilization of variance can be seen based on the residual on the each pattern in each cluster. The proposed patterns were calculated according to mathematical principles but the common patterns did not consider these mathematical principles. The mean squared error (MSE) of the proposed patterns are less than common patterns. Therefore, the patterns presented in this study are more powerful than common patterns. The largest difference between the proposed patterns and the common patterns for estimate the daily average of temperature was 24% in mountainous cluster. Climatic clustering was done for states. The monthly and annual average temperature can be reliably estimated by using the data of sample stations in each state. These findings can be used to estimate daily, monthly and annual average of relative humidity in three climates and sample stations. In addition, one can employ the method for estimating daily, monthly and annual average of relative humidity and temperature based on around climatological clustering of Iran and other stations. Annual relative humidity, temperature, precipitation, altitude, range of temperature, evaporation can also be applied to estimate daily, monthly and annual average of temperature and relative humidity more accurately.
mahboobeh farzandi; Seyed Hossein Sanaeinejad; Bijan Ghahraman; Majid Sarmad
Abstract
Introduction: Temperature and precipitation are two of the main variables in meteorology and climatology. These are basic inputs in water resource management. The length of the statistical period plays a pivotal role in the accurate analysis of these variables. Observation data at Iran's first ...
Read More
Introduction: Temperature and precipitation are two of the main variables in meteorology and climatology. These are basic inputs in water resource management. The length of the statistical period plays a pivotal role in the accurate analysis of these variables. Observation data at Iran's first synoptic station from 1330 (1951) is available at the Iranian Meteorological Organization website The historical monthly precipitation and temperature of five stations in Iran is available since 1880 with missing data. These data measured by the Embassy of the United States and Britain from the Qajar period and recorded in World Weather records books. These synoptic stations include Mashhad, Isfahan, Tehran, Bushehr, and Jask. The monthly missing data were predominantly recorded during World War II (1941-1949). Unfortunately, these data have missing. Therefore, the accuracy of simulating these variables is very important. The current research aimed to predict the missing values of monthly temperature and precipitation in Mashhad station. The stations in the neighboring countries were selected due to the distance to Mashhad, relationship, and completeness of data since 1880, as the predictive variables. Monthly precipitation of Ashgabat from Tajikistan and Sarakhs, Kooshkah, Bayram Ali, Kerki and Repetek from Turkmenistan were selected as an independent variable in the making of Missing Rainfall in Mashhad. Also, the temperature of Ashgabat, Bayram Ali, Gudan, Sarakhs, and Tajan were selected to restore the monthly temperature of the Mashhad station. This research has fitted ten multiple regression models to monthly rainfall of Mashhad station and has fitted 6 multiple regression to the monthly temperature of Mashhad. then the parameters of these patterns are optimized by genetic and Ant Colony algorithm. Also, the Artificial Neural Network (MLP) model and Support vector regression have been selected and implemented in order to simulate monthly precipitation and temperature data of Mashhad.
Materials and Methods: In statistical modeling, regression analysis is a set of statistical processes for estimating the relationships among variables. It includes many techniques for modeling and analyzing several variables when the focus is on the relationship between a dependent variable and one or more independent variables (or 'predictors'). Genetic algorithm (GA) is a metaheuristic inspired by the process of natural selection that belongs to the larger class of evolutionary algorithms (EA). Genetic algorithms are commonly used to generate high-quality solutions to optimization and search problems by relying on bio-inspired operators such as mutation, crossover, and selection. Ant colony optimization algorithm (ACO) is a probabilistic technique for solving computational problems which can be reduced to finding good paths through graphs. This algorithm is a member of the ant colony algorithms family, in swarm intelligence methods, and it constitutes some metaheuristic optimizations. Artificial neural networks are one of the main tools used in machine learning. As the “neural” part of their name suggests, they are brain-inspired systems which are intended to replicate the way that we humans learn. Neural networks consist of input and output layers, as well as (in most cases) a hidden layer consisting of units that transform the input into something that the output layer can use. They are excellent tools for finding patterns which are far too complex or numerous for a human programmer to extract and teach the machine to recognize. In machine learning, support vector machines (SVMs, also support vector networks) are supervised learning models with associated learning algorithms that analyze data used for classification and regression analysis. Given a set of training examples, each marked as belonging to one or the other of two categories, an SVM training algorithm builds a model that assigns new examples to one category or the other, making it a non-probabilistic binary linear classifier (although methods such as Platt scaling exist to use SVM in a probabilistic classification setting).
Results and Discussion: At the first stage, several multiple regressions were fitted to monthly precipitation (with coefficients ranging from 0.63 to 0.81) and six patterns for monthly temperature (0.986-0.993). Afterward, GA and ACO were applied to improve the accuracy of the selected regression models by optimizing their parameters. At the next stage, ANN and SVR were used to estimate the monthly missing values separately. Finally, the results of the previous stages were compared using the root mean square error (RMSE), and the optimal models were applied to determine the missing values of monthly temperature and precipitation of Mashhad. The results showed that the Genetic Algorithm and Ant Colony increase the accuracy of the estimation of missing rainfall data significantly more than the previous methods. The lowest error criterion (RMSE) between regression patterns is 9.8 millimeters. By genetic algorithm, this criterion is reduced to 2.56 mm, and by ant colony algorithm to 2.559.
Conclusion: Comparison of the above methods in restoration temperature and precipitation shows that evolutionary methods (GA and ACO) are the best for estimating the missing monthly precipitation and machine learning methods (ANN and SVR) are the best to imputation missing data of monthly temperature.
nafise seyednezhad; mahboobeh farzandi; H. Rezaee-Pazhand
Abstract
Introduction: The analysis of extreme events such as first frost dates are detrimental phenomena which influence in various branches of engineering, such as agriculture. The analysis and probability predicting of these events can decrease damage of agriculture, horticulture and the others. Furthermore, ...
Read More
Introduction: The analysis of extreme events such as first frost dates are detrimental phenomena which influence in various branches of engineering, such as agriculture. The analysis and probability predicting of these events can decrease damage of agriculture, horticulture and the others. Furthermore, this phenomenon can have a relation with other thermal indexes. The analyzing of first frost dates of all synoptic stations of Khorasan Razavi province is subject of this article. The frequency analysis applied to eight distributions. Then the relationship between first frost dates and thermal index were studied. Best relation was between minimum temperature and return periods of first frost dates.
Materials and Methods: The analyzing of first frost dates (origin is March 21) of all synoptic stations of Khorasan Razavi province is subject of this article. At first data of each station were screening. The basic properties such as homogeneity, randomness, stationary, independence and outliers must be tested. The eight distribution Normal, Gumbel type 1, Gamma 2-parameter, Log normal 2 or 3 parameters, Generalized Pareto, Generalized extreme values and Pearson Type 3 fitted to data and the parameters estimated with 7 methods by the name of the several types of Moments (5 methods), maximum likelihood and the maximum Entropy. The Kolmogorov – Smirnov goodness of fit test can be used to compare the best distribution. The return periods of first frost dates are major application in frequency analysis. There is maybe a relationship between periods and thermal index such as min, max and mean temperature. This relationship can be adapted by regression methods.
Results and Discussion: The statistical analysis for prediction probabilities and return periods of the first frost dates for all synoptic stations in Khorasan Razavi province and the relationship between annual temperature indicators and this phenomenon is the aim of this article. The origin date of this phenomenon is March 21. First, data were screened. Then basic hypothesis test were applied which including the Runtest (randomness), the Mann-Whitney test (homogeneity and jump), the Wald-Wolfowitz test (independence and stationary), the Grubbs and Beck test (detection Outliers) and the three sigma methods (Outlier). The results were: 1-The Sabzevar, Mashhad and Gonabad had lower Outliers that will not cause any problem in data analysis by their skewness. The first frost data of all station were without upper outlier. 2- The independence of all stations was accepted at the 10% level. 3-All stations were Randomness, Independence and homogeneous and lack of jump. Eight probability distributions (Normal, Gumbel type 1, 2-parameter gamma, 2 and 3 parameters log-normal, the generalized Pareto, the generalized extreme values and the Pearson type 3) were applied. The skewness coefficients for all stations were more than 0.1 so Normal distribution was rejected. Also the7 methods of estimation (five different methods of moments, maximum likelihood and maximum entropy methods) were used. The ks fit test was applied. The ks for some stations were closed together at several estimations methods. The results are as follows: GPA (4 times), PT3 (4 times), LN2 (4 times), GA2 (3 times). Generalized Pareto distribution had the best fitted to data (60% of cases compared to the other functions). The results significantly indicated that the occurrence of first frost on the first day of process is in place. The first frost in the period of 2 years at all stations, not occur earlier than Aban(October 28). The 100-year return period event does not occur earlier than first of Mehr(September 22). There is no significant relationship between first frost in the period of 2 years with other factors such as altitude, latitude, longitude, temperature and precipitation as well.
Conclusion: Date of the first fall frost is one of the unfavorite climate influences that cause reduction in crop products. The purpose of this paper is to analysis the frequency occurrence of first frost day in several Khorasan’s synoptic stations as study area. Screening and initial basic tests such as randomness homogenity, independence, etc. were done. Eight distribution function, namely Normal, Gumbel type 1, Gamma 2 parameters, Log normal 2 and 3 parameters, Generalized Pareto and Pearson type III were fitted to data with five probability distributions methods (Ordinary Moments, Maximum Likelihood method, Modified Moments, Probability Weighted Moment and Maximum Entropy). Goodness of fit test was Kolmogorove-Smirnov test. PWM and ModM methods revealed relatively superior results compared to the rest of methods. Generalized Pareto distribution had the best fitted to data (60% of cases compared to the other functions). The results significantly indicated that the occurrence of first frost on the first day of process is in place. The first frost in the period of 2 years at all stations, not occur earlier than Aban. The 100-year return period event does not occur earlier than first of Mehr. There is no significant relationship between first frost in the period of 2 years with other factors such as altitude, latitude, longitude, temperature and precipitation as well.
