نشریه اکولوژی سرزمین
پیاپی 1 (زمستان 1401)

Iran is considered one of the most arid and semi-arid regions in the world, where wheat accounts for about 65-70% of the cultivated land for the main crops. To predict the impact of climate change on the yield of fall wheat (blue) in Jiroft, the Aquacrop crop model was used to simulate fall wheat yield in this study. Then, 17 general atmospheric circulation models from the CMIP5 model suite were used under the RCP4.5 and RCP8.5 radiative forcing scenarios to show changes in climate variables from 2020 to 2098. The results of these models were fine-tuned using the MarkSimGCM climate generator. Evaluation of the accuracy of the AquaCrop model in simulating wheat yield during the baseline period (1990-2016) compared to the observed yield shows the optimal accuracy of this model with R2, MAE, RMSE, and MAPE coefficients of 0.7, 203, 237, and 11.43 percent.The results of predicting wheat yield in the future period show that it will increase under both scenarios, that the average yield increase in the future period compared to the base period is 75 kg/ha for RCP4.5 and 91 kg/ha for RCP8.5. The reason for the yield increase in this product is the reduction of minimum and maximum temperatures in the future period, especially at the end of wheat growth, and the reduction of heat stress.

In this study, the effectiveness of fuzzy models in modelling inner-city highway accidents is evaluated using the Mashhad highway accident data as a case study. For modelling based on fuzzy logic, the variables related to traffic flow and road geometry are used as input variables. Fuzzy modelling involves four stages: Fuzzification of input and output variables, generation of rules, combination and collection of diagrams and de-fuzzification. To fuzzify the variables in the scatter plot, the concept of statistical quantiles is used to assign linguistic terms such as low, medium or high.Based on fuzzy logic, this paper presents two models for predicting the number of financial and fatal accidents on inner-city highways. By comparing the accident numbers estimated by the models with the observed numbers, the efficiency and accuracy of the models can be evaluated by the correlation coefficient R^2. In order to create rules for fuzzy modelling, the results of previous studies were used to identify the factors that influence the occurrence of financial and fatal accidents on inner-city motorways. By demonstrating the effectiveness of the fuzzy logic models created in predicting the number of accidents, the results of these studies can be confirmed.

It is said that climate change will have a negative impact on the tourism industry, since climate is one of the fundamental variables of tourism. In order to investigate the impact of climate change on the future situation (2020-2050) of tourism climate in Lorestan province, two scenarios of radiation forcing RCP2.6 and RCP8.5 and tourism climate index TCI were used in this study. For this purpose, the data of the two scenarios were obtained from ESGF portal. After the exponential scaling of the data using the LARS -WG6 microrotation model, their reliability compared to the ground stations was checked monthly in the base period, then to check the future tourism climate, the monthly average of the parameters was calculated and called to the TCI index. The non-parametric Mann-Kendall test was used to examine the changes in the climatic parameters used in the TCI index. The results showed that qualitative changes of TCI index occur in the next period in Borujerd and Khorramabad stations in 7 months and in Borujerd station in 8 months of the year compared to the base period. In the next period, under both scenarios, we will generally see a decrease in the quality of tourism climate in Aligoders station, a qualitative increase in Borujerd station and the contrast of TCI decrease and improvement in both scenarios in Borujerd station. In the parameters of temperature, sunshine hours and wind speed, there is a positive increase in both scenarios, but it is not significant. In the parameters of precipitation and humidity, a decrease in precipitation in winter is often observed, which from this point of view is considered a positive factor for the tourism climate of the stations, while an increase in precipitation and humidity in summer, especially in RCP2/6, is considered a negative factor for the future tourism climate of the stations. Considering the increase in temperature and sunshine hours in the coming period and the abundance of water resources in the province, investment in water recreation infrastructure and the use of water-related tourism activities will not only utilize the capacity of travelers from neighboring provinces such as Khuzestan in hot seasons, but can also be combined with temperature increases. and sunshine hours to increase the attractiveness of travel and use in this part of tourism activities.

This paper discusses research evaluating the efficiency of neural network models in modeling inner-city highway accidents. As a case study, accident data from urban highways in Mashhad and variables related to traffic flow and road geometry are used as input variables for neural network modeling. Neural network modeling involves three steps: determining the neural network architecture, determining the transfer functions, training and errors, and creating neural network models.In this research, two neural network models were presented to estimate the number of financial and fatal accidents on inner-city highways. To evaluate the efficiency and accuracy of the models, the number of accidents estimated by the models was compared with the observed number and the value of R was used. On this basis, the presented models are suitable for estimating the number of financial and fatal accidents.

The advances in technology have made it an alternative to traditional methods of monitoring, identifying and analyzing oil spills. SAR sensors are commonly used to detect oil spots, but their use of images is limited due to high costs, low crop width and poor time resolution. The MODIS sensor on the Terra and Okua platforms is taking pictures every two times from anywhere on the planet, with images at 250, 500, and 1000 meters resolution, which is a free and valuable resource for detecting oil spills. In this study, in order to detect oil spills during the three oil spill events in the years 2007, 2015, and 2016, oil spills were detected using surface-level images of MODIS. In the pre-processing stage, MODIS images were converted to atmospheric reflections. In order to detect oil spots, an indicator of oil spill detection was used to study the effect of oil spots on the concentration of water chlorophyll and phytoplankton, floating algae, water color, blue opacity and The surface water temperature was used from the indices and algorithms of OC3, CMI, FAI and SABI, NDVI, TWI and SST4 algorithms respectively. The results show that the oil spots in the OSI index output correspond to the pixels with the highest DN in the image per year, as well as oil spots, which reduce the concentration of chlorophylls, floating algae, water color and water surface temperature, and increasing the opacity of the water. For all indices and algorithms, oil spots are consistent with pixels with a minimum value of DNs. Among all the used methods, only the CMI for phytoplankton does not show any significant changes in the range of oil spills; therefore, it is suggested that the methods and indices used in this study be used to detect oil spots.

Urban sprawl, in addition to land use changes, cause of changes in urban landscape pattern has been caused and this phenomenon has several effects on the structure and functioning of ecological systems changes. So, with considering to negative effects caused by the improper use of land and land use changes, Knowledge and understanding of the variability trend is essential for assessing of environmental impacts arising from the development to the planning and sustainable management of land. In this research, Landsat TM satellite images related to May 2001 and OLI sensors related to June 1398 have been used. Also, field visit information and Google Earth images and comprehensive 1: 2000 maps of Kermanshah city have been used. In this study, for geometric correction, topographic maps with a scale of 1: 50,000 prepared by the Army Geographical Organization were used. At this stage, geometric corrections were applied to the images and the OLI image sensor in 1398 was referenced using the vector image method. For this purpose, 24 ground control points with suitable distribution and at the intersection of roads, waterways, etc. were used, so that the mathematical model used to find unknown coefficients in the equation has less error. To convert the coordinates of the corrected image to the uncorrected image, the first degree function was used and to re-sample the pixel value of the uncorrected image, the nearest neighbor method was used and finally the OLI sensor with RMS error equal to 0.35 ground was referenced. Geometric correction of TM image in 2001 was done by image-by-image method. For this purpose, the image of 1398 after geometric correction was considered as a basis. First, the control points were selected, then the points that had a lot of errors were removed from the relevant table, and finally, by removing 7 ground control points, the TM image was corrected with 35 control points with an error of 0.32. In order to determine the changes in addition to the coordinates of the images, the dimensions of their pixels must be the same, which in this study is the size of all pixels is 28.5 meters. First, it was observed that the highest producer accuracy, above 98%, was related to residential and rangeland lands (for two periods). This indicates the high spectral resolution for these classes. Secondly, according to the results, it was observed that The lowest accuracy of the manufacturer has been for the garden class. This class has been classified with %48.82producer accuracy for the image of this area (for 2001). It was also observed that the highest user accuracy was related to forest (2001) and garden (2017) land uses, which are %100 classified. The lowest user accuracy has been for the green space floor. This class is categorized with % 92.62user accuracy for the image of this area. This could be due to the complexity or proximity of the high spectral similarity boundaries to other classes and pixels mixed in the training and experimental samples.