فهرست مطالب محمدرضا محمدپور پنچاه

Weather forecasting and monitoring systems based on numerical weather forecasting models have been increasingly used to manage issues related to meteorology and agriculture. Using more accurate daily average wind speed (10m) and relative humidity forecasts can be helpful in this regard. But systematic and random errors in the model affect the accuracy of forecasts. In this study, the model errors during the 5 and 14 days training period in the same climate areas on the points of the network where the observations are available were calculated. Then the errors were generalized on all points of the network using the cokriging interpolation method. This preserves the model forecasts for other points of the network and only error values are applied to them. To better evaluate the model, the spatial and temporal distribution of daily average wind speed (10m) and relative humidity forecast errors were also investigated over Iran. Observed daily wind speed and relative humidity data from 560 meteorological stations for the period 1/11/2019 to 1/2/2021 were used to evaluate the WRF model performance. The WRF model was run daily at 12UTC, with a forecast time of 120 hours, and first 12 hours of each run was consider as the model spin-up time and was not used in errors calculation. In order to correct wind speed and relative humidity forecast errors for next three days (forecasts of 36, 60 and 84 hours), the forecasts for each day in the period of 11/1/ 2019 to 1/2/2021, was extracted from the model outputs. In order to evaluate the error correction method, the skill score index was used. The validation results of the error correction method showed that the absolute mean error value, correlation coefficient and RMSE improved after the error correction compared to results that were before the error correction, which showed that the error correction method can be used for other network points that did not contain observational data. In general after correction, the RMSE for wind speed and relative humidity forecasts could decrease by 13% and 18%, and the skill score could increase to a maximum of 160% and 308%, respectively. Value of correlation coefficient, after correcting the model error, was significantly increased, compared to the raw model output. In general skill score for the raw wind speed and relative humidity forecast for more than 50% of the days was more than -0.5 and -0.3, but after corrections were  increased to 0.2, 0.4 respectively. Without exception, all climatic regions after error correction have higher skill scores than before error correction, so that the model skill score for most climatic regions after error correction was reached above zero for more than 75% of the days. The results showed that errors of the model in different months, places and climatic zones did not have a uniform distribution. In general, the model underestimated the wind speed and overestimated the relative humidity in most areas. In general, the lowest skill scores for relative humidity forecasts occurred in the colder months of November to February in most climatic zones. The 14-day error correction method did not improve the modeling skill score much compared to the 5-day error correction method, and they acted almost similarly. Knowing the spatial and temporal distribution of model forecast error can be helpful for researchers to have an overview of the areas (and months) where the model forecast error can be high or low.

Weather forecasting and monitoring systems based on numerical weather forecasting models have been increasingly used to manage issues related to meteorology and agriculture. Using more accurate minimum and maximum temperature forecasts can be helpful in this regard. But systematic and random errors in the model affect the accuracy of forecasts. In this study, the model errors during the 5 and 14 days training period in the same climate areas on the points of the network where the observations are available are calculated.Then the errors are generalized on all points of the network using the cokriging interpolation method. This, preserves the model forecasts for other points of the network and only error values are applied to them. To better evaluate the model, the spatial and temporal distribution of the maximum and minimum temperature forecast errors are also investigated in the country. Observed daily maximum and minimum temperatures data from 560 meteorological stations for the period 1/11/2019 to 1/2/2021 are used to evaluate the WRF model. The WRF model is run daily at 12UTC, with a forecast time of 120 hours. And first 12 hours of each run is consider as the model spin-up and is not used in errors calculation. In order to correct the maximum and minimum temperature forecast errors for next three days (forecasts of 36, 60 and 84 hours), the forecasts for each day in the period of 11/1/ 2019 to 1/2/2021, is extracted from the model outputs. In order to evaluate the error correction method, the skill score index is used. The validation results of the error correction method shows that the absolute mean error value, correlation coefficient and RMSE are improved after the error correction compared to results that were before the error correction. This shows that the error correction method can be used for other network points that do not contain observational data. The results shows that the RMSE of the raw model maximum (minimum) temperatures forecasts for next three days is approximately 6 degrees Celsius (5 degrees Celsius), which after error correction reaches 2 degrees Celsius (4 degrees Celsius). Also the value of correlation coefficient, after correcting for the model error, has a significant increase compared to the raw model output. The average skill score for the raw minimum and maximum temperature forecast for more than 50% of the days is more than -1 and -1.9, respectively, but after correction, the model skill scores become closer to one and for more than 75 percentage of days that reach above zero. Without exception, all climatic regions after error correction have a higher skill score than before error correction, so that the model skill score for most climatic regions after error correction reaches above zero for more than 75% of the days. Before error correction, the warm semi-humid zone has the lowest average skill score for forecasting maximum and minimum temperatures among climatic zones, but after error correction it reaches the highest value among other zones. In general, for areas with hot and dry climates, the raw output skill score for predicting the minimum temperature in July, August, and September is minimized. The 14-day error correction method did not improve the modeling skill score much compared to the 5-day error correction method, and they acted almost similarly. In areas with high elevation gradient, the model error increases. In general, model underestimates the maximum and minimum temperatures in most areas. Knowing the spatial and temporal distribution of model forecast error can be helpful for researchers to have an overview of the areas (and months) where the model forecast error is high.

Evapotranspiration as one of the main components of the hydrological cycle, has a significant role in proper irrigation planning and water resources management. In this case, estimating evapotranspiration is limited due to a lack of data and a deficiency of meteorological stations. Therefore, today numerical models such as WRF are a powerful tool for generating and predicting meteorological quantities (wind speed, humidity, etc.) that are needed to estimate evapotranspiration. So far, no research has been conducted to investigate the effect of different schemes of the WRF model on the estimate of rice evapotranspiration. The purpose of this study is to evaluate the efficiency of the WRF model and obtain the result for estimating evaporation for rice plant in the central plain of Guilan.

Evapotranspiration rates vary from 2.7 to 8.5 mm per day. The average ET during three different periods of plant growth, including the initial, middle, and final periods, is estimated to be 4.63, 5.97, and 5.98 mm per day, respectively. The three configurations 1, 2, and 4 are mainly overestimated in predicting evapotranspiration of rice plants, and the computational values are estimated to be higher than the values measured by the lysimeter. The results show that the highest amount of RMSE occurred in configuration No. 4 at 8.47 and the lowest rate occurred in configuration No. 3 at 1.26. Summary of results shows that configuration No. 3 in all four criteria mentioned has performed better than other configurations to predict daily evapotranspiration of rice. The results showed that the non-local schema used in the model, simulates better than the local schemas for the daily evapotranspiration of the rice plant. Findings show that in the local YSU schema, the accuracy of predictions is significantly increased and is only 0.64 mm on average less than the estimated lysimetric data.

The results showed that using appropriate schemas in the surface layer and boundary layer of the WRF model, affects on accuracy of evapotranspiration predictions. The results of this study showed that, this model by using the YSU non-local boundary layer scheme can accurately predict the evapotranspiration rates of the rice plant for the next day and this is due to the higher ability of this schema in predicting the parameters affecting evapotranspiration (including temperature and wind). Therefore, the WRF model can be implemented by using GFS forecast data for the next few days and by applying the FAO-Penman-Monteith equations to the model outputs, the values of potential evapotranspiration for different regions of the country can be calculated. Since evapotranspiration is directly related to atmospheric thermodynamic processes, so using other different atmospheric physics schemas (not considered in this study) can produce different results.

Oman Sea and its coastlines have an important role in the international trade, coastal management and marine industries. Large weather instability and intense wind occur in Oman Sea due to tropical cyclones. The wind field simulated by atmospheric models can be used in ocean model for wave prediction. The main purpose of this research is to investigate applicability of WRF mesoscale model version 3-7-1 in surface wind simulation using various boundary and initial conditions over Oman Sea. for this aim, three data sets including Era-Interim reanalysis data, FNL and GFS analysis data have been used. Simulated wind at the coasts of Oman has been evaluated using observational data measured at synoptic stations in Iran and Oman and also data measured by buoy at Gheshm Island. Evaluation of simulated offshore wind has been done using data from National Climatic Data Center Blended Sea Winds with 0.25 degree horizontal resolution and 6-hourly time step. Moreover, SST data from NCEP dataset with 0.083 degree in horizontal resolution have been used as WRF input data. Model outputs have been improved based on nudging technique. In this research, WRF model has been run using three 3-, 9- and 27-km nests, that the smaller one covers Oman Sea and some portions of the Persian Gulf. The model has been run for a time of 60 hour with 12 hour spin-up period for June 2009. Finally, fifteen “2-day re-started” simulations were performed to complete one month simulations. Results show that all three simulations overestimate wind speed at the considered coast area and the largest error belong to simulations that used Era-Interim dataset and the smallest error occurred in simulations that used FNL dataset. Comparison of the three datasets (analysis and reanalysis ones) with observational data indicated that using GFS dataset provided more accurate data due to its higher resolution. Moreover, ECMWF datasets underestimated them, while simulations using ECMWF them data as initialization and boundary conditions overestimated the winds. Bias-averaged values over the offshore areas demonstrated that using GFS and FNL datasets leads to underestimation, while using Era-Interim dataset resulted in overestimation in of predicted winds. Histogram of wind speed reveals that maximum error occurred for low wind speed for all three datasets (wind speed smaller than 3 m/s). In the mid-range (wind speed between 3-12 m/s), the model has an appropriate performance for simulating wind speed. Using GFS and FNL underestimates wind speed larger than 12 m/s, while using Era-Interim data overestimates that. Simulations using GFS and FNL have little discrepancy for various wind speeds, due to same model in producing these datasets. While results obtained from Era-Interim differ significantly with those from GFS and FNL datasets. Using FNL dataset produced the least error in wind direction. Since both GFS and FNL datasets are produced in NCEP with the same data assimilation techniques and forecast systems, the significant difference between these two datasets refers to the number of used observational data in producing analysis dataset (more observational datasets have been used in producing FNL dataset, comparing with those used in producing GFS dataset). Therefore, it can be concluded that dense grid of observational data in producing analysis dataset has an important role in mesoscale simulations. As a conclusion, using FNL dataset an input of WRF model led to the best performance in simulation of wind speed and wind direction for coasts and offshore part of Oman Sea.

Fog is among the most important weather hazards from the aviation perspective. This phenomenon can substantially lead to horizontal visibility reduction. Therefore, accurate prediction is essential for flight safety and easing air traffic. Fog consists of a weather condition in which water drops and ice crystals reduce the horizontal visibility to less than 1000 meters. Various methods are suggested for fog forecasting. Numerical and statistical methods, experimental approaches, and very short range fog forecasting are some of the most common methods. Experimental methods are commonly used for first guess in forecasting centers. Saunders technique is one of the forecasting methods for radiation fogs using radio sounds data. Although this technique goes back to many years ago, it is being used in many parts of the world, including UK Met Office, and is recommended by World Meteorological Organization.
Present study tries to evaluate the performance of two experimental methods using real data after studying synoptic condition of fog occurrence in two selected airports. The validity of them is then measured with the real occurrence in a number of case studies of fog occurrence for the selected airports using the bias technique in order to choose the more appropriate method. In the next step, the more appropriate method is administered using the numerical prediction model output and is again evaluated with the bias technique. In both these methods, an index called fog point temperature has been used, and the fog occurrence has been determined by calculating this temperature and comparing it with the minimum temperature. The selected airports are Mehrabad Airport, Tehran and Shahid Hasheminezhad Airport, Mashhad, which have been chosen because of high flight traffic (in Tehran) and high fog occurrence (in Mashhad). Experimental methods examined in this study are Saunders and Prichars-Crodack techniques, which 25 case studies in selected Airports tried to offer the best results for first guess of fog occurrence. The accuracy of these relations was evaluated comparing real conditions using Bias technique. After choosing the more appropriate method, a similar process has been carried out using numerical prediction model of WRF for the next 12 hours.
Results of synoptic evaluations show that high-pressure systems are a major factor in creating coldness in lower levels of the atmosphere. Evaluation of pressure field in this study doesnt show figures below 1020 hPa. Specific humidity values were 6-8 g/kg and 4-6 g/kg for 1000 and 925 hPa levels respectively. Winds are frequently northern or eastern and cold weather advection is seen in selected stations.
In Saunders technique, using radio sound data of 1200 UTC in 25 case studies for mentioned airports, the fog point is calculated. This temperature is then compared with next day's minimum temperature and if the difference is less than -2°C, fog occurrence would be ruled out. Saunders considers this method mostly useful for radiation fog. In Crodack-Prichars technique, which is performed by creating a regression association between temperature and dew point temperature, the fog point temperature is determined. Here again, fog is not formed If the temperature difference is less than -2°C.
After calculating fog point temperature using Saunders technique and comparing it with actual observation, it was found that among 25 cases, 15 fog observations were consistent with Saunders technique calculations. In the five cases of fog nonoccurrence, the results of this method were consistent with reality. So, Bias evaluation technique shows 75% for probability of detection.
The same process has been carried out for Crodack-Prichars technique. In this method, a linear relationship exists between temperature, due point temperature, and fog point temperature. Wind condition and cloudiness are also presented experimentally in the form of a table. For different amounts of these two factors, a numerical amount of 1.5 to -1.5 is added to the fog point temperature. Fog occurrence is determined by calculating fog point temperature using Crodack-Prichars technique and comparing it with the minimum temperature according to table 2. This evaluation showed that in 13 of 20 fog occurrence cases, the right answer were obtained, and 5 cases of fog nonoccurrence, were consistent with reality. Therefore, POD index was reduced to 65%. Based on the results, Saunders technique has been considered as the more appropriate method for initial guessing in fog forecasting in the airports under study. Now, the values of temperature and due point temperature were determined for the next 12 hours using WRF numerical prediction model, and Saunders technique was used again for predicting fog (using predicted data). The results of this evaluation were also investigated using Bias evaluation technique, which were not so agreeable, so that it was consistent with reality in 50% of cases. Hence, it seems that careful consideration of numerical prediction models output is needed.

Introduction One of the factors that affect the climate of arid and semi-arid areas is dust storm. Numerical models are new methods for evaluation of dust storms which can also be used for forecasting dust storms. The goal of this research was to utilize air mass back trajectory and mesoscale meteorological model analyses to investigate the wind patterns in Iran leading to synoptic-scale dust outbreaks impacting the west and center of this region. Back trajectories from a receptor site should assist in determining the air masses and dust transport patterns into the area of study to help better understand and forecast the formation and direction of motion of dust plumes. Synoptic-scale desert dust outbreaks are initiated by specific weather conditions that depend on the nature and strength of large-scale wind systems over the region. Weather patterns that lead to dust outbreaks can be simulated using computer models that support a wide range of simulations related to the long-range transport, dispersion, and deposition of aerosols. Also, identify the source of infection is the first step in the process of determining an effective strategy for controlling pollution. One way to find the sources of pollution is back trajectory this means that the back trajectories from the receiver site can be used to specify the source locations. A coupling of meteorological and trajectory models is common method in studies of dust storms. In this study, we evaluated one off intense dust event that affected most of Iran. We used two numerical models, Weather Research and Forecasting (WRF) model and HYbrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model. WRF model analyses were used to investigate meteorological conditions in the center of Iran and HYSPLIT back-trajectory analyses were used to investigate the wind patterns that led to dust outbreaks by evaluating air transport pathways reaching Iran during dusty days.
Methodology In this study numerical models are used to investigate and forecast of dust storms. The Advanced Research version of WRF (ARW) was used to produce atmospheric fields at a high resolution over the study region. WRF model analyses were used to investigate meteorological conditions in region. WRF is a limited-area, non-hydrostatic, terrain-following sigma-coordinate model designed to simulate or predict mesoscale atmospheric circulation. HYSPLIT model was used to compute simple air parcel trajectories as well as dispersion and deposition simulations. 1 nested domains of 30 km horizontal resolutions and 28 vertical levels was defined in WRF model. The model was integrated continuously for 192 h starting from 00UTC of 30 Jun 2009. Initial and boundary conditions were adopted from National Centers for Environmental Prediction Final Analyses (NCEP FNL) data available at 1° horizontal resolution. Boundary conditions were updated at 6-h intervals during the period of model integration. Then WRF outputs converted to format of input data for HYSPLIT model and this model was run to investigate the source of dust storm using calculation of back trajectories from receptor site. The back trajectories provide the Lagrangian path of the air parcels in the chosen time scale, which will be useful to identify the source locations of the pollutant that fall in the track of the back trajectories. Also for reduce uncertainty of back trajectories, trajectories were calculated at different altitudes (100 m, 500 m and 1000 m).
Discussion On the sixth day of July, 2009, when the storms come to Tehran, horizontal visibility is less than 100 meters. Many cities and human activities were suspended. Shamal Wind is the prominent stream. From late May to early July, One alternate weather phenomena in the central region of the Middle East, is a summer wind that is of great significance in enhancing surface dust flux. The storm occurred on the sixth of July 2009, the present of a low-thermal pressure on Iran creates a strong pressure gradient in a semi-permanent high-pressure zone on Saudi Arabia. Synoptic conditions simulated with WRF showed that the formation mechanism of a summer shamal wind on the Tigris and Euphrates basin set on the Mediterranean and a frontal cyclone with remarkable high level forcing is made. With strengthen the System, A heavy postfrontal dust storm has formed that by passing Zagros mountain transport this dust to wide range. Convergence of location paths using the HYSPLT model, seen in the West Country and the Iran-Iraq border that indicate the possible locations of springs in the area behind this point. Behind the convergence point is located parts of Iraq and Syria, where is included arid and dry land and could be source of this dust storm.
Conclusion In this study, we try to test use of numerical models to investigate the conditions governing the formation of dust storms and also measure the performance of numerical models in forecasting of dust storms. For this purpose, one of the severe dust storms that affected most of Iran was selected and WRF model was used to simulate the atmospheric conditions governing it. WRF model output results forecasted Shamal wind that was main factor governing the formation of the storm. Then, back and forward trajectories of particles scattered on windy days (18-12 July 2009) by the HYSPLIT model indicate that wind in these days came from northern Syria and Iraq. Using the results obtained in this study it can be said that the coupling atmospheric forecasting model (WRF) of diffusion model (HYSPLIT) can be well evaluated Weather conditions and trajectories and forecast dust storms.

One of the factors that affect the climate of arid and semi-arid areas is dust storm. Numerical models are new methods for evaluation of dust storms which can also be used for forecasting dust storms. Weather patterns that lead to dust outbreaks can be simulated using computer models that support a wide range of simulations related to the long-range transport, dispersion, and deposition of aerosols. Mesoscale atmospheric models are widely used to capture the complex flow and meteorological parameters essential in dust outbreaks (for example, Ginoux et al, 2001; Zender et al, 2003; Kim, 2008).But one of the main problems in the study of contaminates such as dust is to quantify the relationship between air quality and pollution sources. Identify the source of infection is the first step in the process of determining an effective strategy for controlling pollution. One way to find the sources of pollution is back trajectory this means that the back trajectories from the receiver site can be used to specify the source location (Petzold et. al. 2009). Today, a coupling of meteorological and trajectory models are common methods in studies of dust storms. Yazd province is one of the low rainfall areas in the center of Iran that is almost covered with desert and sandy plains. Consequently, this province is frequently faced with dust storm phenomena. In this study, owing to lack of numerical studies of dust storm in Iran, dust storms of Yazd province were analyzed using numerical models. We used two numerical models, Weather Research and Forecasting (WRF) model and HYbrid Single-Particle Lagrangian Integrated Trajectory (HYSPLIT) model. WRF model analyses were used to investigate meteorological conditions in the center of Iran and HYSPLIT back-trajectory analyses were used to investigate the wind patterns that led to dust outbreaks by evaluating air transport pathways reaching Yazd area during dusty days. Study area: Yazd Province is one of the 31 provinces of Iran. It is in the center of the country. The province of Yazd has one of the driest climates in Iran due to its location east of the Zagros Mountains. Yazd is the driest major province in Iran, with an average annual rainfall of only 60 millimeters (2.4 in), and also the hottest north of the Persian Gulf coast, with summer temperatures very frequently above 40 °C (104 °F) in blazing sunshine with no humidity. Due to this climate and geographical position, the province of Yazd have been subjected to dust storm phenomena and have been suffering from large damages. In this study numerical models are used to investigate and forecast of dust storms. At first, Dust storms in the period of 2000-2009 were studied and dust storms which reduced visibility to less than 1000 meters were extracted. Then one of the strongest dust storms on 29 May 2003 was analyzed in detail. The Advanced Research version of WRF (ARW) was used to produce atmospheric fields at a high resolution over the study region. HYSPLIT model was used to compute simple air parcel trajectories as well as dispersion and deposition simulations. 2 nested domains of 45 and 15 km horizontal resolutions and 28 vertical levels was defined in WRF model that first domain was used for synoptic analysis and second domain was used for analysis of convergence zones and convective motions in the center of Iran. The model was integrated continuously for 48 h starting from 00UTC of 28 May 2009. Initial and boundary conditions were adopted from National Centers for Environmental Prediction Final Analyses (NCEP FNL) data available at 1° horizontal resolution. Boundary conditions were updated at 6-h intervals during the period of model integration. Then WRF outputs converted to format of input data for HYSPLIT model and this model was run to investigate the source of dust storm using calculation of back trajectories from receptor site. The back trajectories provide the Lagrangian path of the air parcels in the chosen time scale, which will be useful to identify the source locations of the pollutant that fall in the track of the back trajectories. Also for reduce uncertainty of back trajectories, trajectories were calculated at different altitudes (10 m, 500 m and 1000 m). Results of this study indicated that 20 dusty dates with visibility less than 1000 meter were occurred in Yazd station. Most of these cases occurred in February to July and in most of these cases wind direction were west and northwest. Synoptic and dynamic analysis of dust storm on 29 May 2003 using WRF outputs shows that passing of cyclonic systems of high levels and surface heating create strong instability and high surface wind speed in the region that this higher surface wind speeds lead up dust and sand. These conditions are associated with deep mixed layers and convective conditions before storm and a maximum area of strong convergence of surface wind at the time of starting dust storm and dust storm has formed. Back trajectory analysis using HYSPLIT model indicated that Gavkhoni marsh in southern part of Isfahan province and arid lands around it are the sources of the considered dust storm. In this study it was shown that the use of numerical models and appropriate approach to assess and predict dust storms. The models were used in this study were Weather Research and Forecasting (WRF) model and HYSPLIT trajectory model. For the case that investigate in this study WRF model simulated the wind flow under the influence of an existing cyclonic system during the study period. The passage of the cyclone with cold air in the central regions of Iran simultaneously with the heating surface area has created severe instability and formed the dust storm. Further, the back trajectories obtained from HYSPLIT model predict winds come in the quadrant between north and west where the Gavkhoni dry salt marsh is located.