فصلنامه سنجش از دور و GIS ایران
سال شانزدهم شماره 1 (پیاپی 61، بهار 1403)

Climate change is considered one of the most significant challenges facing humanity. This phenomenon has had notable impacts on agricultural production in most parts of the world, especially in arid and semi-arid regions. The average temperature has also increased in many areas over the past few decades. Today, various studies utilize remote sensing indices as one of the modern methods for identifying climate change. One of the important remote sensing indices is the phenology characteristics of vegetation cover, which has demonstrated a promising ability in identifying and estimating vegetation cover in recent studies.
The five-day time series of the Normalized Difference Vegetation Index (NDVI) from NOAA-AVHRR images and plant phenology parameters were utilized to examine changes in vegetation cover in rangeland and rainfed agricultural lands of the Urmia Lake basin over the years 1984-2013. Temperature and precipitation data were obtained from meteorological stations in the Urmia Lake basin and were used in comparison with satellite image results.
The results of the time series analysis over the thirty year period in the Urmia Lake basin revealed that the start of the growing season in Oshnavieh, Saqqez, and Sarab areas in 2013 commenced earlier than in 1984, while in the Maragheh area, it started later. The end of the growing season in Oshnavieh, Saqqez, and Takab ended earlier, and based on the peak growth parameter, vegetation in these counties reached its maximum value earlier. The results also showed the length of the growing season in Oshnavieh, Maragheh, and Saqqez counties has shortened, respectively.
The results obtained from satellite images and climate data indicated that the  observed changes in phenology parameters are location-dependent. Cold nights and warm days at the start of the growing season decreasing and increasing, respectively. However, warm days increased at the end of the growing season. These changes have increased the slope of the plant growth phenology curve during the senescence period, ultimately reducing the length of the growing season.

Remote sensing provides a powerful data source for the mapping of urban areas and the monitoring of urban dynamics on a range of scales. Among the variues types of  remote sensing data, images captured at night offer an effective means ofmonitoring human activities on a global scale. The distinctive features and capabilities of these images permit the separation of urban areas and other human activities, the main feature of which is the use of light at night by accurately measuring the location, from the background without light. Via providing uninterrupted and continuous monitoring from the night world perspective, these images provide a valuable source of information about human activities over time from the past to the present. The time series analysis of this data is highly valuable for discovering, estimating and monitoring social and economic dynamics in countries, especially sub-regions where there are no official statistics. With recent developments in night-time data satellite sensors and new research conducted in this field, this study aims to review the advances in night-time sensors, introduce the existing data and products, review and express the advantages and disadvantages of each one, and review the methods and solutions presented in previous research for solving the existing problems and limitations in order to improve these images.

The main objective of this research is to introduce and review the general charactristics of night-time light data, discussing their advantages, challenges, and methods for addressing these challenges. The majority of studies on DMSP night light images focus on two spatial and temporal dimensions. In the spatial dimension, inherent deficiencies of this dataset are observed, such as saturated numerical values in central urban areas and flourishing effects in suburban and rural areas. In the temporal dimension, the lack of calibration in the processor, necessitates the implementation of additional processes on annual products of stable DMSP night light data in order to examine urban dynamics. The existing methods for correcting spatial problems are divided into tow categories: spectral and non-spectral. Similarly, methods for addressing temporal issues are divided into two categories: annual calibration of night light data and adjustment of temporal patterns. NPP-VIIRS monthly images encompass various features including fixed light values such as city lights and transportation routes, as well as noise values such as gas flames, biomass burning, and background noise. Therefore, preprocessing is necessary before utilizing this data. Furthermore, the positioning accuracy of Loujia_01 data is lower than its spatial resolution, resulting in image displacement of up to 650 meters in some areas. Geometric correction is applied to rectify this issue, and various correction methods have been investigated.

A general comparison of the data sets reveals that, despite the existing problems and limitations, the DMSP stable night light data outperforms other night light datasets due to its longer time series, which spans from 1992 to 2013. This extended temporal coverage makes it a valuable resource for research on urban dynamics and estimating the overall growth trend of cities. On the other hand, NPP-VIIRS offers advantages and is sensitive to faint light sources. However, its passage time at 1:30 in the morning, when many lights are turned off, limits its utility for urban studies. Consequently, it may not be the optimal choice for exclusively investigating urban areas exclusively. Nevertheless, the NPP-VIIRS data is more useful in research related to economic activities. Furthermore, the sensor's lack of sensitivity to blue light emitted by LEDs impacts its ability to accurately quantify artificial light emissions from the ground.

The objective of this study was to introduce types of remote sensing night light data and their analysis. In short, current research in the field of correcting spatial saturation and blooming problems is divided into two categories: spectral and non-spectral. Non-spectral methods typically rely solely on night light data, although they may also incorporate non-remote sensing data. Spectral methods often employ spectral indices that are related to vegetation and ground surface temperature. Currently, correcting DMSP images from the temporal dimension can be achieved through inter-data calibration, specifically via the fixed reference regions or reference pixels method. One of the most reliable methods in this field is the reference area method. Following the conclusion of the DMSP-OLS mission, the VIIRS was introduced. In contrast to the annual data of this satellite, the monthly data requires correction due to the presence of background noise, and stray lights. A reviews of existing studies indicates that the majority of methods aim to remove noise using specific frameworks although with differing assumptions. Finally, considering the current challenges and limitations of night light satellites, several recommendations for future progress and development in this field are put forth. Further investigation could be conducted into the integration of DMSP-OLS data with NPP-VIIRS data or higher resolution Loujia-01 data, with the objective of developing a longer time series for future research on urban dynamics.

  As one of the most essential needs of living beings, clean air quality has been threatened by natural and human activities. In recent years, dust storms have been increasing spatially and temporally, causing numerous damages to social, economic, and environmental health for the residents of the southern and southwestern regions of Iran. In the present study, MODIS sensor data were used to investigate dust storms and detect horizontal optical depth.

The  advantages of MODIS sensor data include high spectral and temporal resolution. Additionally, meteorological station data were collected based on the study period. After preprocessing the data and preparing field observations, the necessary features for modeling were extracted using the differential method between selected bands of each MODIS sensor image, along with features extracted from ground-based meteorological station sensors. After further investigations and evaluations and using the viewpoints of meteorological experts, 36 differential features from various MODIS image bands and six features from ground-based meteorological station data, totaling 42 features, were extracted. Subsequently, using feature selection techniques, the best features were identified. A novel method named ML-Based GMDH, which improves the GMDH neural network by altering partial functions with machine learning models, was employed to detect dust concentration and horizontal optical depth. To achieve optimal accuracy, the hyper-parameters of this model were heuristically tuned using the TLBO optimization algorithm. Additionally, machine learning methods such as Basic GMDH, SVM, MLP, MLR, RF, and their ensemble models were implemented to compare with the main approach. According to the results, the TLBO-tuned ML-Based GMDH method provided superior accuracy in detecting dust concentration compared to the aforementioned machine-learning methods.

The SVM-PSO method was selected as the best method in the feature selection phase, the RF method was chosen as the best method among basic classification methods, and the Ensemble SVM and Ensemble RF methods were selected as the best methods in the ensemble and classification phase. It was also observed that using the ensemble approach led to a desirable improvement in horizontal optical depth classification. In the second approach, a method titled ML-Based GMDH, which improves the GMDH neural network by altering partial functions with machine learning algorithms, was used for estimating dust concentration. Additionally, to achieve suitable accuracy, the hyper-parameters of this model were finely tuned using the TLBO optimization algorithm. The results showed that this method provided appropriate accuracy in estimating dust concentration and horizontal optical depth, out performing the best-selected methods from the first approach

The COVID-19 epidemic is considered a geographical phenomenon, and its impact on decision-making and daily life is significant. Geographic information systems (GIS) and spatial techniques play crucial roles in analyzing the spread of COVID-19 globally. Studies using spatial analysis have highlighted the importance of social and health variables in infection and mortality rates, despite existing uncertainties about the effects of meteorological variables. Given Iran's climatic diversity, it is valuable to identify the key spatial factors influencing COVID-19. Therefore, this study aims to model and determine the factors affecting the COVID-19 epidemic based on available data.

This study investigated the factors influencing the distribution of COVID-19 infection rates using global and local spatial regression methods. Seventy-three cities were selected, with data on COVID-19 infections available from March 10 to June 20, 2019. The factors considered were altitude, population density, average age, the ratio of the population over 55 years to the total population, and meteorological parameters, including humidity, temperature, pressure, and wind speed. Their relationships with the disease were analyzed using spatial statistics methods. Stepwise regression identified population density, air pressure, average age, and wind speed as significant predictors, and the occurrence of the disease was modeled using the Ordinary Least Squares (OLS) technique. Due to the unstable relationship between the independent and dependent variables, the Geographically Weighted Regression (GWR) technique was used. Principal Component Analysis (PCA) and SPSS software were employed to address spatial variability and multicollinearity.

The results showed that the OLS model was statistically significant, with variance values explained by the model being non-random. However, the explanatory variables had an inconsistent relationship with the dependent variable in both geographic and data spaces. The residual distribution deviated somewhat from normal, indicating model instability. Thus, the GWR technique was applied for modeling. PCA addressed multicollinearity (due to a cluster pattern in meteorological variables), reducing meteorological factors to one component, explaining nearly 70% of the variance. The model improved by consolidating the average age and the ratio of the population over 55 into one factor. Subsequently, population density, meteorological factors, and age demographics were utilized as predictive variables in the GWR model. A 10% increase in the adjusted R-squared of the GWR model (63%) demonstrated its relative improvement over the OLS model. Moran's spatial autocorrelation test indicated that, while the cluster pattern of residuals was less pronounced in the GWR model than in the OLS model, it remained significant at the 99% confidence level. Hot spot analysis at the 95% confidence level identified the western parts of Kurdistan province and the northern and western parts of Khuzestan province as hot spots (areas of significant underestimation), and the eastern parts of Hamadan province and the northern parts of Bushehr province as cold spots (areas of significant overestimation). Thus, at least one variable affecting disease occurrence was not considered. Other potential factors, such as cultural, health, and genetic variables, were not included due to unavailability or measurement difficulties.

This study highlights the importance of demographic and environmental factors in COVID-19 infection rates and provides a foundation for further research in the study area.

Drought conditions can vary from moderate to severe and have different durations, necessitating continuous and operational monitoring. The longer the drought persists, the more pronounced its impact on vegetation and water resources becomes, and the more severe the drought, the greater the limitation of services for humans and the alteration of natural systems. Habitat destruction for wildlife, reduced water quality, and reduced access to water resources could be consider as most effects of drought. Drought monitoring is essential for researchers, managers, and decision-makers to identify vulnerable areas, which can be used to reduce the consequences of drought.

In this study, an attempt has been made to investigate the vegetation drought situation in Iran by using Suomi NPP infrared sensor images obtained from the Earth Data website (earthdata.nasa.gov) and using (NDVI), (VCI), (TCI), and (VHI) indices. The study period, spanning from April 1st to July (the 13th to 26th week), was selected as it encompasses the typical drought duration in Iran. The Standard Precipitation Index (SPI) was calculated for Iran using daily precipitation data from 143 synoptic stations. Subsequently, the correlation coefficient was calculated between SPI and each of the indices (NDVI), (VCI), (TCI), and (VHI). In infrared images, M bands have a resolution of 750 meters, while I bands have a resolution of 375 meters.

Based on the rainfall data recorded in synoptic meteorological stations, there is minimal rainfall during the summer months (July, August, and September). Conversely, the majority of rainfall occurs during the autumn, winter, and spring seasons. Consequently, the water year in most regions of Iran commences approximately in the third decade of September and continues until the second and third decade of June annually. In this study area, the optimal temporal base for monitoring and estimating drought on the vegetation is from April 1st to June 30th. In this article, the effect of precipitation on vegetation conditions was investigated using the standardized precipitation index (SPI), derived from monthly precipitation data from synoptic meteorological stations. Iran experiences a dry season in summer, with August being the driest month of the year. The temporal and spatial changes in drought for each vegetation indicator are markedly different.

Based on the majority of years experiencing drought, the vegetation cover is expected to face mild or severe drought. This is demonstrated by a decrease in the values of each indicator. In years  that the vegetation was affected by drought, the values of the indices show a decrease in April, followed by an increase in June and July. This suggests the beginning of a severe drought. Based on the calculated SPI, it was determined that the area experiences low precipitation during the hot months, indicating a lower rate compared to other months.

Effective decision-making and management in the field of sustainable development of natural resources require access to accurate and up-to-date climatic information. This information enables the examination of the role of climate change in various issues and the formulation of effective management strategies accordingly. In this context, temperature as one of the most important climatic indicators plays a key role in environmental analyses and research. Given the fundamental role of temperature in various situations, access to precise and comprehensive temperature data is of high importance. Such data should be sufficiently detailed to provide a clear and complete picture of temperature patterns over time. Unfortunately, climatic data often faces problems such as statistical discontinuities and measurement errors, which can lead to incorrect decisions and inefficient planning. In this research, statistical methods have been employed to analyze existing temperature data and their statistical discontinuities. These methods include geographical coordinates (graphical), normal ratio, weighted correlation coefficient, and arithmetic mean, which are well-established and widely used in completing climatic data. Selecting the most appropriate method from among these can enhance the accuracy of temperature data estimation and play a pivotal role in decisions based on more comprehensive and reliable data. The objective of this research is to identify the optimal methodology for estimating data and addressing statistical discontinuities. This will assist researchers, managers, and policymakers in the field of sustainable development and in better understanding climatic conditions, enabling them to make more informed and effective decisions.

In order to address the statistical gap, a number of well-known and popular classical statistical methods were evaluated for estimating Iran temperature data. These included geographical coordinates, normal ratio, weighted correlation coefficient, and arithmetic mean. In order to determine the best method for completing missing information, data from 125 stations with complete information (without any missing data) over 21 years (2000 to 2020) were used. Given the extensive and time-consuming nature of the calculations, a random selection of 10% of the stations with an appropriate spatial distribution was employed to carry out the data filling operations. The information of the selected stations was removed at each stage separately and reconstructed based on their five nearest stations. To evaluate the aforementioned methods, statistical evaluation criteria such as R-squared (R2), root mean square error (RMSE), and mean absolute deviation (MAD) were used.

The results of the analysis of computational values through the normal ratio method were evaluated against observational values. It was found that all the stations under study exhibited a high correlation, indicating the acceptability of the normal ratio method to data estimation. The average values obtained from the evaluation of results indicate that the methods of normal ratio, weighted correlation coefficient, geographical coordinates, and arithmetic mean are prioritized in order, with RMSE values of 3.05, 3.28, 3.30, and 3.51 degrees Celsius, respectively. Consequently, the normal ratio method is the most suitable among the other studied methods and can be employed to address issues such as a lack of information, existing data errors, and also the expansion of the study period.

Among the methods reviewed, the normal ratio method is generally more acceptable and of higher quality than the other methods and is recommended for use in future research within similar study ranges. In subsequent ranks, the methods of geographical coordinates, weighted correlation, and arithmetic mean are placed, respectively. It is notable that, although the other methods are considered of secondary importance, they nevertheless demonstrate satisfactory efficacy in certain locations. Consequently, under varying circumstances, a range of methods may be employed to address data deficiencies, and the optimal approach should be selected and utilized in accordance with the specific study area.

One of the most important steps towards sustainable development is the protection of land integrity, so that a part of the land is changed annually for several reasons, and the withdrawal of such lands from the production path creates irreparable damages. Given the high intensity of land use change in Mazandaran province, including the Haraz plain, which represents a significant environmental issues at both macro and spatial scales, the monitoring and analysis of theese changes can be considered a valuable tool for for the management and planning of land use. Considering that Haraz plain has not been spared from the crisis of destructive land use changes in recent decades, the need to monitor, highlight and process these changes as one of the most important management factors in this region is confirmed.

In order to investigate land use changes, it is necessary to integrate a number of layers over a specific period of time. This research aims to investigate land use changes in Haraz Plain from 1980 to 2021. Therefore, Landsat data was employed to quantify the changes. By applying atmospheric, geometric, and radiometric corrections, image enhancement operations were performed and land use change maps were produced based on the supervised classification method, maximum likelihood algorithm and basis component analysis functions. The type of land use changes was determined from the difference function of the identification images and the accuracy of the maps using the overall accuracy test and the Kappa statistic.

The results showed that from 1980 to 1990, the area of forest lands decreased by 4 km2. The rangeland area also decreased from 450 to 436 km2. From 2000 to 2010, the area of forest land decreased from 272 to 270 km2 and rangeland decreased from 432 to 420 km2. Finally, between 2011 and 2021, the area of forest lands decreased by 9 km2 and the rangeland area decreased by 5 km2. The results indicated a reduction in the area of forest and rangeland, accompanied by an increase in the area of agricultural land and residential areas.

The results obtained and the defined goals allow us to conclude that the area in question underwent changes in terms of its use during the considered statistical period (1980-2021). These changes were noticeable. Therefore, human factors interventions play a pivotal role in land use changes. The results of this study can assist planners in identifying the factors influencing land use changes and in making appropriate management decisions in the future.

Land cover maps are essential elements in geographical analysis and spatial planning. The accuracy and effectiveness of these maps rely on three factors: Satellite imagery, classification algorithms and training samples. The quality of the training dataset significantly impacts the accuracy of classification results. This study aims to generate reliable training samples using the training sample migration method to monitor land cover changes in northwestern Iran from 2002 to 2022.

The study area covers 7653 square kilometers in northwestern Iran, situated between 44°35′59′′ to 46°01′25′′ longitude and 38°38′46′′ to 38°47′48′′ latitude. Data utilized in this research include satellite images and ground truth data, specifically Landsat images. The research methodology comprises five main steps. Initially, satellite images were obtained, followed by pre-processing steps involving radiometric and geometric corrections. Subsequently, training samples were prepared using high-resolution satellite images (Google Earth images) and ground surveys. The third step involved training sample migration, where spectral similarity between training samples from reference and target years was assessed using two parameters: Euclidean distance (ED) and spectral angle distance (SAD). After determining a suitable threshold, migrated training samples were distinguished from non-migrated samples. Evaluation of the accuracy of migrated training samples was conducted using reference data derived from Google Earth. In the fourth step, classification of satellite images from different years was performed using the migrated training samples. Finally, the accuracy of the classified images was assessed through the calculation of a confusion matrix in the fifth step.

The results indicate that a threshold of 0.9 to 1.1 is optimal for distinguishing migrated training samples from non-migrated training samples across different years. It can be observed that there is an inverse relationship between the accuracy of migrated training samples and the percentage of migrated training samples, with an increase in the percentage leading to decreased accuracy. Evaluation of the accuracy of migrated training samples based on each parameter (SAD and ED) reveals that migrated training samples based on the SAD parameter exhibit higher accuracy than those based on the ED parameter. Furthermore, the use of migrated samples based on both parameters has resulted in a 10.45% increase in accuracy compared to using the ED parameter alone, and a 5% increase compared to using the SAD parameter alone. Analysis of the percentage of migrated training samples in different land cover classes demonstrates that, on average, 80.6% of water class training samples, 75.4% of bare land class samples, 71.2% of built-up class samples, 64.6% of grassland class samples, 60.2% of cropland class samples, and 54.4% of wetland class samples were migrated from the reference year (2022) to each of the target years (2002, 2007, 2012, and 2017). The accuracy assessment of migrated training samples in different land cover classes also reveals that the water, built-up, bare land, grassland, cropland, and wetland classes had the highest accuracy in the migrated training samples, in that order. Analysis of land cover changes between 2002 and 2022 indicates a decrease in the area of bare land, water, and wetland classes from 2002 to 2022, while the area of the built-up class has increased during this period. Additionally, the grassland and cropland classes did not exhibit a consistent trend of change during this period, with their trends differing in different years. However, overall, the area of these two classes increased in 2022 compared to 2002. 

Future studies should consider using other satellite images (including Sentinel-2) for migrating training samples to evaluate the impact of different spectral bands and satellite images on the migration process. Furthermore, investigating the effectiveness of the training sample migration method for migrating training samples of other land covers could be a potential research topic for future studies.The study area is located in northwestern Iran with an area of 7653 square kilometers. The study area lies between 44°35′59′′ to 46°01′25′′ longitude and 38°38′46′′ to 38°47′48′′ latitude. This study used satellite images and ground truth data.The research methodology consists of five main steps. The initial step involved obtaining satellite images and performing pre-processing steps (radiometric and geometric correction). In the second step, training samples were collected using high-resolution satellite images (Google Earth images) and ground surveys. The third step involved the migration of training samples. To do this, the spectral similarity of the training samples from the reference and target years was first calculated using two parameters: Euclidean distance (ED) and spectral angle distance (SAD). Then, adopting the specified threshold, the migrated training samples were separated from the non-migrated samples. Furthermore, the accuracy of the migrated training samples was evaluated using reference data prepared from the Google Earth. In the fourth step, using the migrated training samples, the classification of satellite images in different years was performed. Finally, using the indices obtained from the confusion matrix, the accuracy of the classified images was evaluated.The results showed that the threshold of 0.9 to 1.1 is the optimal threshold for separating migrated training samples from non-migrated training samples in different years. We also found a reverse correlation between the accuracy of the migrated training samples and the percentage of the migrated training samples, with higher percentages resulting in lower accuracy.The accuracy assessment of the migrated training samples based on each parameter (SAD and ED) showed that the migrated training samples using the SAD parameter have higher accuracy than the migrated training samples using the ED parameter. Moreover, employing the migrated samples considering both parameters has increased the accuracy by 10.45% compared to using the ED parameter to migrate the training samples, and by 5% compared to using the SAD parameter to migrate the training samples.The analysis of the percentage of migrated training samples in different land cover classes showed that, on average, 80.6% of the training samples of the water class, 75.4% of the bare land, 71.2% of the built-up, 64.6% of the grassland, 60.2% of the cropland, and 54.4% of the wetland were migrated from the reference year (2022) to each of the target years (2002, 2007, 2012, and 2017). The accuracy assessment of the migrated training samples in different land cover classes also showed that the water, built-up, bare land, grassland, cropland, and wetland classes had the highest accuracy in the migrated training samples, in that order.The classification of satellite images was performed using Landsat images between 2002 and 2022. Accordingly, satellite images were classified into six different land cover classes. The accuracy assessment results showed that the overall accuracy of the classified images in 2022, 2017, 2012, 2007, and 2002 was 94.95%, 91.93%, 90.74%, 89.45%, and 88.94%, respectively. The accuracy assessment of different land cover classes based on two parameters, producer accuracy and user accuracy, showed that the water class has the highest producer and user accuracy among different classes (98.2% and 99.34%, respectively in 2022). In contrast, the wetland class had the lowest producer and user accuracy (90.1% and 91.25%, respectively in 2022).The analysis of land cover changes between 2002 and 2022 showed that the area of bare land, water, and wetland classes decreased from 2002 to 2022, while the area of built-up class increased during this period. Furthermore, the grassland and cropland classes did not exhibit a constant trend of change during this time period, and their trends varied by year. However, the area of these two classes increased in 2022 compared to 2002. The analysis of the changes in the area of built-up class throughout this period shows a significant increase in the area of this land cover, which has increased from 20.38 square kilometers in 2002 to 123.98 square kilometers in 2022.It is suggested that in future studies, other satellite images, like Sentinel-2, be used to migrate training samples in order to evaluate the effect of different spectral bands and satellite images on the migration of training samples. In addition, investigating the effectiveness of the training sample migration method in migration training samples of other land covers can be one of the research topics in future studies.