فصلنامه سنجش از دور و GIS ایران
سال چهاردهم شماره 1 (پیاپی 53، بهار 1401)

Map of croplands is one of the information layers required in the efficient management of these lands. Having such maps makes it possible to monitor agricultural fields during the growing season continuously. In this study, a solution to produce map of Shahrekord’s agricultural lands in two agricultural and non-agricultural classes is presented using the time series of different extracted indices from Sentinel-2 images. Since the use of large data sources is one of the obstacles to the development of methods based on the time series of satellite images, the Google Earth engine processing platform has been used in this study. The proposed method is based on integrating supervised pixel-based classification results with segmentation results. First, training data of supervised classification is provided in a rigorous refining process without the need of collected data from field surveys or interpretation of high-resolution satellite images. Then, by calculating the separability of the two target classes in the time series of each index, the optimal indices are selected. Finally, by combining the results of segmentation and classification methods based on the votes obtained from the classification results, agricultural or non-agricultural class is assigned to each of the image segments. In addition to incorporating spatial information including edges and spatial proximity, this method has been able to improve the noise and porous results of pixel-based classification and has increased the overall accuracy of the final map from 90.7% to 96.05%. Also, user accuracy of both agricultural and non-agricultural classes show an improvement of 3.27 and 7.97%, respectively.

Land surface temperature is considered a key parameter in the physic processes of land surface at all scales of local to global. In this study, the relationship between land surface temperature with vegetation and soil surface moisture in land uses of Zahak plain of Sistan area was investigated. In order to, Landsat TM (1987), TM (2001) and OLI (2018) satellite imagery were used. After the preprocessing and image processing steps, the extraction of land use maps was performed based on the monitored classification method and through maximum probability algorithm for a period of 30 years. Also, land surface temperature was evaluated statistically by separate window method and the relationship between land surface temperature with vegetation and soil moisture. The results showed that the accuracy of classification by maximum probability method through geomorphic facts data, TM and OLI images in terms of kappa coefficient of 0.89, 0.95 and 0.84, respectively, based on the overall accuracy of 91.8, 96.45 and 87.89% was obtained. During 1987, 2001 and 2018, average of the land surface temperature indices were 38.13, 45.73 and 41.14 ° C, the normalized difference vegetation index was -0.11, -0.13 and -0.16, and the normalized difference moisture index was estimated 0.64, 0.63 and 0.58. The relationship between land surface temperature and normalized difference of vegetation index was no correlative. The correlation between land surface temperature and the normalized difference of humidity index was also inverted and negative. Plant regeneration and growth was decreased owing to factors including hydrological drought and Climatic conditions due to reduced rainfall, rising air temperature and Dust storms. Therefore, due to the lack of suitable vegetation, vegetation is not effective in reducing the surface temperature of the study area.

Estimation of the production potential of a crop is a function of climatic conditions, crop genetic potentials and various other environmental and managerial factors. Assessing the ability of regions to realize the genetic potential of crops is one of the important points of macro-planning in agriculture. Considering the position of Qazvin province in the production of Maize and the importance of cultivating this crop, estimating the yield of this strategic product as accurately as possible is very necessary. In this regard, by studying an 11-year statistical period, Maize yield was estimated with the new crop model AquaCrop-GIS. The zoning of key product indicators was simulated through the model in the province. By examining the results of these parameters, it finds that Qazvin and Moallem Kelayeh study stations with higher reference evapotranspiration rates have higher water productivity. Then, with the help of the computational yield, components of water footprints, and total water footprint of the crop was estimated within the study stations. By examining the regression equations in each station, it was found that the relationship between blue water footprint and crop yield compared to other water footprint components for all stations has a higher coefficient of determination (R2 = 0.43, R2 = 0.51, R2 = 0.43, R2 = 0.77 and R2 = 0.79 for Qazvin, Avaj, Moallem Kelayeh, Takestan and Buin Zahra stations, respectively) and level of significance. In general, the coefficient of determination of these relationships in Buin Zahra station with R2 = 0.88, R2 = 0.79, R2 = 0.56 and, R2 = 0.53, respectively, for green, blue, gray, and total water footprints compared to other stations were rated higher. This reduction in yield at the station had a significant effect on increasing the total water footprint of the crop.

The study of snow cover as one of the most important sources of freshwater supply is of great importance. Due to the mountainous conditions of Iran, it is not possible to measure the area of snow cover. Accordingly, the use of satellite imagery to identify snow storage is of great importance. In this study, the spatio-temporal changes of Iran snow cover for the cold period of the year were evaluated using the snow cover product of MODIS Terra satellite during the period of 2003-2018. The trend and slope of the snow cover were investigated using Man-Kendall non-parametric tests and the Sen's slope estimator and change-point of snow cover using Buishand test. The results showed that in January, the highest amount of snow cover is 16.6 percent, and the lowest amount of snow cover was computed in October, which is less than 1 percent. The main center of Iran's snow cover in the cold period of the year in the highlands is above 4000 meters. The snow cover trend is negative in all studied months and the maximum decrease in snow cover was calculated in January and the change-point was calculated in 2008 January, which is statistically significant at the level of 0.05. The significant decrease in snow cover during the cold period of the year which is a major threat to Iran's water resources.

The aim of this study was to estimate soil properties using Landsat 8 satellite bands in part of farmlands of Qorveh-Dehgolan plain in western Iran. Soil sampling was conducted at a total number of 107 points from 0-15cm depth throughout the study area and their physicochemical properties were measured in the laboratory. In order to extract information from the Landsat 8 satellite image following application of the vegetation mask; DOS values ​​for bands 1-7 were extracted for the sampling points. Correlation Analysis, Stepwise Linear Regression and Principal Component Regression were used to determine the relationship between soil properties and digital value of Landsat 8 bands. Validation of Regression Analysis was evaluated using two parameters of Coefficient of Determination and Root Mean Square Error. The results showed that there was a positive and significant correlation between the digital value of most Landsat8 bands to the amounts of sand and free iron oxides in the soil but a negative and significant of it to amounts of clay and silt in the soil. There was no significant correlation between heavy metals concentration and digital value in visible and near infrared bands while Regression Analysis did not provide acceptable performance in estimating soil properties of the study area. According to the obtained results, it seems that Landsat 8 satellite images can be used to estimate the soil texture and the amount of free iron oxides across the study area.

Changes in crop growth at relatively short intervals, asymmetry of cultivation of similar crops, the spectral similarity between different crops at certain times of the growing season, and lack of ground data make classifying crops in satellite imagery a challenging task. Changing the amount of canopy and greenness during the growing season is one of the most prominent characteristics of vegetation, including agricultural products, which can be monitored by using time series of vegetation indices that have useful information about the sequence of phenological features of crops. The use of deep learning methods with the ability of learning sequential information obtained from these time series can be useful in crop mapping and reducing dependence on ground data. The LSTM network is one of the types of RNNs in sequential data analysis that has the ability to learn long-term sequences of time-series information. Therefore, in this study, after extracting the NDVI time-series of 9 different dates from Sentinel-2 satellite images for a region located in Moghan plain, with ground labeled data related to the type of crops cultivated, we trained a convolutional LSTM network. Then we used this trained network to classify agricultural products in another region of the plain as a test site, and achieved an overall accuracy of 82% and a kappa coefficient of 0.8. Increasing the number of ground samples and selecting the exact boundary of crops, can increase the efficiency of the method used.

Development in its general sense, industrial, technological and spatial progress, especially in developing countries, has led to adverse effects on the environment not only on a regional scale but also at different regional, national and sometimes global levels which has similarly affected the ecological security of the regions. In recent decades, more attention has been paid to the issue of environmental safety in the world, and various methods have been developed to evaluate it, but to date, most research on ecological safety has been done based on the pressure- Status -response model and fewer studies have been conducted based on approach landscape ecology models have dealt with this category. There is also little research focusing on dynamic changes in ecological security, in particular simulating and predicting the future development of environmental security. The purpose of this study is to monitor and predict the environmental security situation in the period 1991 to 2035 by combining the support vector machine algorithm, landform ecology model, Markov chain combination model, and automated cells for the Nazarabad county area of ​​the functions Alborz province. For this purpose, using the classification of Landsat satellite images in two 15-year time periods from 1991 to 2021, the trend of land use changes in the region in five land use classes; Construction lands, cultivated lands, wetlands, vegetation and barren lands were studied and CA-Markov model was used to prepare land use maps for 2035. MPS, CA, NP, PLAND, AWMSI, and PD metrics were calculated to quantify the landscape appearance patterns at the class level and LPI, CONTAG, and SHDI metrics were calculated at the landscape level. Then, the ecological safety index was modeled for the landscape metrics of the study area. The results indicate a decrease in integration and an increase in the number of spots in the cultivated land class and the development and expansion of man-made lands in these lands. On the other hand, we have witnessed the phenomenon of integration in the barren lands of the region. Therefore, the ecological security of the region during the study period affected by the above events was evaluated more intensively during the years 1991 to 2006  and more gently in the years 2006 to the forecast of 2035. It was suggested that more attention be paid to environmental considerations and principles of protection in regional development programs.For this purpose, using the Landsat satellite image classification in two fifteen-year periods between 1991, 2006, and 2021, the trend of land use change in the study area in five land use classes; Construction lands, cultivated lands, wetlands, vegetation cover, and bare lands were examined and to quantify the patterns of landscape appearance at the class level, MPS, CA, NP, PLAND, AWMSI, PD metrics and at the landscape level LPI, CONTAG and SHDI metrics were calculated. The CA-Markov chain model was used to prepare land use maps for 2035, and then the ecological security index was modeled for land use metrics in the study area. The results indicate a decrease in integration and an increase in the number of spots in the cultivated land class and the development and expansion of man-made lands in these lands. On the other hand, we have witnessed the phenomenon of integration in the barren lands of the region. Therefore, the ecological security of the region, during the period under review, affected by the above events, was assessed with more intensity during the years 1991 to 2006 and with a milder slope in the years 2006 to the forecast of 2035. Therefore, it seems necessary to pay more attention to environmental considerations and the principles of protection and protection in the development programs of the region.