Journal of Spatial Analysis

---

 Quantitative assessment of landslide sedimentation in the ILAM dam Basin

Information on the accurate volume of landslides and sedimentation in landslides is a research necessity, with the assumption that the bulk of sediment accumulated in the ILAM Dam (located between , E and , N) is related to the surface landslides of the basin. Although the role of landslides in erosion, sediment transport and sedimentation of slippery basins is confirmed and different experts understand and determine the relationship between the fluctuation of slopes and the fluctuation system in many respects more important than other areas. Because according to the results they can assess the widespread environmental changes, but comprehensive research on the scale of catchment basins has done very little (Harvey 2002). So far, the study of wet landscapes in Iran has been more sensitive to the factors, their sensitivity and their hazards, and there has been no study on the sedimentation of landslides.

Data and Method
First, using a geomorphologic system methodology with topographic maps of 1: 50000, geological map of 1: 100000, aerial photography1: 20000, Landsat TM1988 ETM2002,2013 satellite imagery, and Google Earth in the GIS environment in the following sub-basins and landslide events at the following levels The basin was drawn. The discharge data of the water and sediment flow of three hydrometric stations GOLGOL,CHAVIZ and MALEKSHAHI Station were provided from the waters of the ILAM province. Two models of estimated MPSIAC and EPM models have been used to estimate soil erosion and subsoil sedimentation. The Moran spatial correlation model was used to introduce the spatial pattern of landslides, and the fuzzy logic model was used to determine the relationship between the dependent landslide to the independent variables and the potential risk of landslide hazard in the basin. In order to elucidate the quantitative results of landslide sedimentation, empirical models of estimation of sediment erosion, hydrological model of discharge curve and sediment, observational statistics of sediment during statistical period, landfall time occurrence in compliance with the hydrometric station sediment peak during the statistical period of computation Estimated a small amount of sedimentation of the landslides of the ILAM dam basin.
Result and Discussion
The spatial correlation model of Moran showed that the data have spatial correlation and cluster pattern. The average total sediment production in the MPSIAC model in the GOLGOL basin was estimated to be 13.3 tons per hectare per year under the CHAVIZ basin of 10.3 tons per hectare for one year and 4.00 tons per hectare in the sub-basin MALEKSHAHI. Using hydrological model of discharge-sediment curve, the mean sediment was calculated during the statistical period at the hydrometric station of the sub-basin of GOLGOL 18.8 ton per hectare, the station CHAVIZ 10.4 tons and the station MALEKSHAHI 0.9 tons of sediment per hectare per year was calculated. According to the results of the research methodology, the observation of the sediment in the two stations of GOLGOL and CHAVIZ compared to estimated sediment is related to the events occurring in these two sub-basins.
The data of 16 active landslides were recorded. Under the GOLGOL basin, 9 landslide events occurred, and in the CHAVIZ basin, 7 landslide events, the time of landfall occurrence recorded with sedimentary peaks, the length of the statistical period, the precipitate in the sub-basins was almost synchronized. Average relationship between suspended period of the statistical period - average of the peak delayed flight time of the statistical period coinciding with the occurrence of landslide = the amount of suspended load of landfall occurrence in the basin.
The amount of suspended land slip under the GOLGOL 75088.19 - 315.85=74772.34
Landing slope under the Chavez Basin 19907.30 - 20.24=19887
The area of the sub-basin is about 29,000 hectares and the active landslide area is about 100 hectares. According to the calculations, 77772.34 tons of suspended sediment is a sedimentary passage passing at the GOLGOL hydrometric station, which shows with a coefficient of 1.4 times the suspended sediment load of approximately 104681 tons of landslide sedimentation in this sub-basin, which shows the amount of sediment yield 100 hectares of landslide, so each landslide hectare had an average of 1046. 81 tons of sediment deposited at the GOLGOL hydrometric station. The area under the Chavez Basin is about 14000 hectares and the active landslide area of this sub-basin is about 65 hectares. According to the data of the discharge data, the sedimentation of the Chavez hydrometric station is 19887 tons of suspended sediment load, which shows a 1.4 equivalent of 27842 tons of landslide sedimentation in this sub-basin, equivalent to a slope of 428.33 tons per hectare.
Conclusion
According to the calculations, it is concluded that in the sub-basin of flowering GOLGOL, 37.35% is equivalent to 4.9 tons per hectare per year, the increase of sediment is related to landslide events. As a result, 28.2 tons of sediment per hectare were introduced in one year Dam reaches ILAM. The results showed that in the CHAVIZ sub basin, 38.2 percent is equivalent to 4.6 tons per hectare per year for the increase of sediment related to landslide events. As a result, an amount of 14.5 tons of sediment per hectare has entered ILAM dam in one year. In the sub-basin MALEKSHAHI, there was no increase in sediment during the period without active landslide occurrence. A total of 1237314 tons of landslide deposition have entered the ILAM Dam. To control this threat, the appropriate action by the executive office for sustainable development should be applied.

---

Occurrence conditions for severe snow blizzard in the west-north of Iran
 
Abstract
The blizzard incident is one of the climatic hazards that occurs due to the combination of other climatic factors such as temperature (below zero), snow and wind (at 15 m/s). In this research, the conditions of blizzard in Northwest of Iran are carried out using statistical methods. By analyzing all the meteorological codes of the blizzard (36, 37, 38 and 39) during the statistical period from 1987 to 2016 for 11 synoptic stations of the study area, codes with severe blizzard (37, 39) were selected. Then, using the geopotential height, wind and Leveling temperature of 500 and 850 hpa, obtained from the NCEP/NCAR open source database, the synoptic patterns of blizzard incident analyzed. Statistical analysis of the relationship between the effects of geographic factors on severe blizzard has shown that the factor of height has the greatest effect on intensity, increase and incident spatial differences of this phenomenon. The study of the synoptic patterns of the incident of the blizzard phenomenon showed that five main patterns play a role in creating it in the region. The synoptic patterns of development include the formation of a low cut-off center, a long landing passage from Iran, the formation of a relatively deep and drawn Mediterranean East, The rectangular system is a rex-shaped system and is an umbilical bundle system. Among the patterns obtained, the patterns that were bundled were, the most important role in the survival and transfer of flows associated with cold, and other patterns, despite the frequency they had, were periodically of severity and weakness.
Keywords: Blizzard; North West; Wind speed; Temperatures below zero; Synoptic patterns
Human life is always affected by climatic phenomena, especially the hazards of the two variables of temperature and wind. One of the most important simultaneous phenomena of these two variables is the blizzard, which is caused by heavy snow, stormy winds, and very low temperatures. This climate risk can cause damage to various areas of horticulture, agriculture, urbanization, transportation, and so on. This phenomenon is present in regions such as Canada and North America with a cold weather wave that results from turbulence in the winters and damages the lawns in these areas. There are plenty of local storm in the polar regions and it lasts for a few days. For example, the wind in the Adelie land in the Antarctic is so severe that the area is known as the storm Land. In Iran, the most significant blizzard occurred in mid-February 1350, resulting in the deaths of more than 4,000 people across the country. In this research, considering the characteristics of the blizzard phenomenon at the time of occurrence (severity, continuity, expansion, and time of occurrence), the study has been conducted to determine the statistical synoptic patterns in the northwest region.
In this research, the studied area is northwest of Iran, which includes 6 provinces (Ardebil, West Azarbaijan, East Azarbaijan, Zanjan, Kurdistan, Kermanshah, Hamedan). In order to study, the days with the blizzard phenomenon in the form of 3 hours and in codes of this phenomenon (36, 37, 38 and 39), were obtained from the establishment of the stations studied by 2016. In the following, for precise examination, stations with 30 years of statistic from 1987 to 2016 were identified and the statistical (frequency, daily, monthly and annual frequency) codes 39 and 37 were studied. Finally, the relationship between blizzard with the latitude and elevation in the studied stations was determined. To assess the statistical results, the correlation coefficients (R) and coefficient of determination (R²) were used.
In the second part, the identification of synoptic patterns was done by Principal Component Analysis in MATLAB software and ocular method. The criterion for identifying synoptic patterns, the days where codes 37 and 39 are more than 1 time (3 hours) within 24 hours or two days behind each other at the stations studied. In order to determine the patterns, at first, the average geopotential data of the 500-hpa level from 1987 to 2016 were obtained from a range of 10-70 degrees north latitude and 0-80 degrees east longitude with a spatial resolution of 2.5 * 2.5 from the NCEP / NCAR data.
Statistical analyzes on the relationship between the effect of geographic factors on severe blizzard showed that the factor of height had the greatest effect on the intensity, magnitude and spatial differences of this phenomenon. In sum, the most important factor in the occurrence of this phenomenon is due to atmospheric conditions and synoptic patterns of the region. In this study, the most frequent occurrence of codes 37 and 39 in all stations studied was at Sardasht station and Khalkhal station, respectively. Also, the statistical study of the frequency of the annual and monthly occurrence of each code showed that code 39 in 1990 and code 37 in the years 1989 and 1990, as well as in January, had the highest frequency of each of the two codes.
Investigating the patterns of the occurrence of the blizzard phenomenon showed that five main patterns have contributed to its creation in the region, the first pattern due to the formation of a low cut-off center, which, with the cold weather in Central and Eastern Europe, has reduced the temperature in the northwest. The second pattern is due to the high landing passage from Iran, which has crossed the descent from a cold and cold weather zone from Europe to Iran. The third pattern is the location of the studied area in the relatively moderate, dragged, eastern Mediterranean wavelength, causing cold weather to fall to the northwest. The fourth pattern, with the formation of a Rex-type blockade on the Mediterranean, has led to the transfer of cold air from Eastern Europe, Kazakhstan, and high latitudes to Iran. The fifth pattern, with the formation of a blockade, has caused cold weather in northern Europe and Central Asia to enter the country from the north, causing a drop in temperature in the region.
Among the known patterns, the patterns that were blocked (pattern 4 and 5) played the most important role in the survival and transfer of cold fluxes and even drawn to lower latitudes. Other patterns, despite frequent periods, provide conditions for the occurrence of this phenomenon and, unlike the blocking patterns, have had severity and weakness.
 
Keywords: Blizzard; North West; Wind speed; Temperatures below zero; Synoptic patterns
 
 
 

---

---

---

One of the causes of environmental hazards is the change in the pattern of surface water flow in floodplains following the construction of flood Spreading networks. The purpose of this study is to prepare a zoning map of vulnerable areas of the flood Spreading station of Musian plain  in Ilam province after the implementation of the aquifer project in this plain. To prepare this map, five factors influencing the change in flow pattern including elevation, slope, flow direction, geological formations, and landuse change were examined. Then, in the GIS environment, each class of the mentioned factors was given a score of zero to 10 based on the range and the corresponding weight layers were created. Then, by combining the created weight layers, the vulnerability zoning map of the area was created based on 5 classes: very low, low, medium, high and very high. The results showed that the most important threat and danger factor is the concentration of waterways behind erosion-sensitive embankments. Also, the study area in terms of vulnerability includes three classes with medium risk, high and very high and covers 16, 62 and 22% of the area, respectively. Flood and upland Spreading areas, risk areas and lowland lands are the most vulnerable parts of the basin in terms of floods and sedimentary deposits.

---

.
 Introduction
Old trees are important and key elements of forest sites and are of great value in terms of forest management, reforestation, silviculture and ecology. Although old trees constitute a small percentage of forest trees, they account for a large share of forest carbon reserve and play a vital role in carbon storage. Understanding the how geographical and site distribution of these trees across the forest is essential to obtain information for forest restoration management. Therefore, this study was carried out to investigate the geographical and site characteristics of old trees of Wing nut, Ash, Hackberry, Sycamore, Elm, Olive, Cypress and Fig in Ilam province.
 
Materials and methods
After querying the villagers and local people and conducting numerous forest surveys, the old trees were identified and selected on the basis of the diameter of the breast. Then their geographical characteristics including city, district, village, geographical coordinates and site conditions including slope, aspect, altitude, soil depth, climate and proximity to water source were measured or recorded.

Results and discussion
The results showed that in terms of geographically distribution, the identified old trees have located in Ilam, Mehran, Malekshahi, Badreh and Dehloran cities. Topographically, the old trees of Wing nut, Elm, Ash and Fig were located in the 0-10% slope class, Hackberry and sycamore in the 0-10% and 10-30% slope classes, olive in the 10-30% slope class and Cypress in the 40-70% slope class. The old trees of Wing nut, elm, Ash and Hackberry were located in the north aspect, fig, sycamore and Cypress in the south aspect and olive in the west and south aspects. The old trees of Wing nut, elm, Ash, Hackberry, Sycamore and Cypress were dispersed at altitude class of 1100–1250 m and olive and fig old trees were at altitude class of 1250–1400 m above sea level. Climatically, the old trees of Wing nut, elm, Ash and Hackberry were located in the very cold Mediterranean climate, Cypress trees and some sycamore trees in the cold Mediterranean climate, and fig, olive and some plantain trees were in the semiarid cold climate. In terms of access to water resources, old trees of Wing nut, elm, Ash, Sycamore, Hackberry and Fig were located on the bed or margin of river, old Cypress trees had no access to water resources and some olive trees were close to water resources. In terms of soil subsidence, old trees of Wing nut, elm, Hackberry, olive, and fig were mostly in soils with medium depths. Old ash and sycamore trees were present in shallow to medium depths and old cypress trees were present in shallower soils. Although the identified old trees were present in limited sites, their long-term and sustained presence in these sites indicates that sites conditions are favorable for their survival.

Conclusion
Therefore, it can be concluded that the presence of low slopes, suitable soil bed and access to water resources were desirable characteristics for stability and survival of the studied old trees in these sites. Due to the above-mentioned characteristics, ecologically similar sites can be found in the forests of the province and can be restored by seed of old and resistant trees.
Keywords: Site, Topography, Climate, Geography, Old trees, Ilam

 

---

Introducing the appropriate model of oak forest and drought relations

in Ilam province

Introduction
The forest ecosystems of the Zagros vegetation region have a very long history of exploitation in various
forms. The material of the Zagros vegetation region is Iranian oak. In recent years, a significant
proportion of oak forests have dried up or have experienced drought. Although the main cause of drought
in these forests has not been determined yet, in the preliminary studies, factors such as climate change,
increasing dust, increasing drought periods, pest infestation and disease, high user changes have been
cited as reasons for drought in the Zagros forests. (Hosseinzadeh and Pourhashemi, 1396). Iran's location
in the arid and semi-arid zone of the world (sub-tropical region) has often been associated with
fluctuations in climatic and atmospheric elements and under the influence of atmospheric currents,
synoptic patterns, irregularities in precipitation and temperature patterns (Rahmati, 2016; 1383).
Comparison of the effect of climatic variables on healthy masses and affected by the decline of oak in
Khorramabad city based on rainfall and temperature data using Pearson correlation coefficient, on annual
growth rings of oak trees Effectiveness of drought of oak trees from both series In general, healthy trees
have been more affected by monthly and seasonal temperatures and have shown the highest coefficient of
correlation with the temperature of the region (Naseri Karimvand et al., 2016). , And the Standardized
Rainfall Index (SPI) and the correlation between these two indicators in assessing and monitoring drought
in different areas of Isfahan province, the results showed that NDVI plant index can be a good alternative
to climatic indicators in drought assessment and monitoring) with the conclusion and colleagues, 2011:
79).
Data and Method
So from SPI and NDVI indicators and Moran index and statistical regression statistics and satellite
images of Modis and Landsat have used to analyze the relationship between dieback of Ilam forests and
happened drought in the region. The precipitation data of 93 rain gauge stations were analyzed during the
statistical period and according to the dry coefficients of SPI index, drought zoning layers of Ilam
province were prepared for two time series of 2000 to 2009 and 2010 to 2019. Greenery's raster layers
were prepared from Modis satellite imagery for the mentioned time series. The results of analysis of
Moran's statistical showed a significant correlation between the SPI index and the NDVI index in spatial
dimensions. By a simple random method, 143 points of oak dieback with dimension of 30 m 2 , which each
point was equivalent to a pixel-size, were recorded with a GPS device, and by simulating in satellite
imagery, the droplet layer of oak dieback was extracted.
Result and Discussion
What is debatable about the results of the implementation of methods for obtaining drought ranges and its
relationship with oak drought points or masses is that the results of the models show a statistically close
and direct relationship between drought and oak drought. . The general trend of oak drought and drought
in these two decades has been from the southeast to the northwest of Ilam province, with increasing
temperature and decrease in rainfall in the southern and eastern regions of the province and increasing
rainfall and decrease in temperature in the central and northwestern regions of Ilam province. The data of
the synoptic stations are consistent. Analysis of satellite imagery and declining greenery in the models
although the study was aimed at meteorological drought and precipitation fluctuations, but spatial
changes of arid points and masses in the province were adapted to field visits and human intervention,
especially in the southeast with agriculture. Under the rubble and the remnants of the dried trees, the ax
has been placed on the roots of these trees, and this trend is spreading in almost other parts of the arid
areas of the province. Therefore, due to the irregularity in the pattern of precipitation and temperature of
the research country (Rahmati, 2016; Zandi Army, 2004) and the effect of monthly and seasonal

نشریه تحلیل فضایی مخاطرات محیطی، سال نهم، شماره 1، بهار 1401 2
temperature on the growth and decline of oak trees in the study (Naseri Karimvand et al., 2016) and other
related research and The flooding situation in the basins of Ilam province, the rainfall, the impermeability
of the soil and their erosion, and finally the lack of moisture in the months before the oak trees grow in
the soil and the increase in temperature in the dry season, which leads to reduced humidity and eventually
greenery. Variables affecting oak drying in linear regression are not responsive, but more accurate results
will be obtained in multivariate regression, although regression analyzes are spatially empty, and X and Y
represent a one-way, quantitative analysis based on the number of dried trees with pixel counts. Drought
range is measured which this defect in SPI method despite its spatial and statistical analysis using Moran
statistical index due to non-compliance in the coefficients of this index with the range of changes in
Moran statistical analysis in statistical analysis is a more appropriate explanatory coefficient than The
regression models showed but at a lower level than the NDVI method it placed. The advantage of NDVI
method with Moran statistical analysis is the relationship between pixel and pixel, ie in spatial analysis,
all pixels that have green changes have been analyzed in the same domain of spatial changes with oak
trees. High results and higher statistical explanation coefficient were obtained than other models.
Conclusion
Although linear regression between extracted oak dieback points with SPI and Moran statistical indicators
was significant, but the relationship between NDVI index and Moran statistic has the effect of
independent variable of drought trend in spatial and temporal dimensions on the dependent variable
process of oak drought with spatial analysis. And nonlinear regression has a more appropriate and
accurate statistical significance and explanation. So this method as desirable method has been introduced
for analyzing of drought and oak dieback.
Keywords: Ilam province, oak forest drought, drought, Moran index

---

Vulnerability assessment of Miangaran wetland ecosystem

To support the proper management of ecosystems, vulnerability analysis of ecosystems is very important. Vulnerability analysis of ecosystems provides information about weaknesses and capacity of the studied ecosystem for recovery after damage. Considering the degradation status of Miangaran wetland, vulnerability evaluation of this wetland is one of the most important management methods in the region. For this purpose, in this study, after identifying and evaluating the threatening factors of Miangaran wetland, these factors were scored using evaluation matrices. Then, the interaction between these values ​​and threatening factors was examined and the vulnerability of wetland values ​​was obtained by multiplying the scores of all studied factors. Finally, management solutions were presented to deal with the most important threatening factors. According to the results, the most vulnerability is to the hydrological and ecological values ​​of the wetland. The highest effects of threats on the ecological value are also on the birds of Miangaran wetland. The results of the evaluation of Miangaran Wetland show that this wetland has a high potential for ecosystem functions of the wetland. These functions have been neglected in the planning and managing of wetlands at the local, regional and national levels. As a result, ecosystem-based management is suggested as the best management approach. The management in these areas should take action to prevent the vulnerability of Miangaran wetland. Also, the vulnerability evaluation method used in this study can provide a good understanding of the relationship between wetland functions and the resulting services for the management of the ecosystem of Miangaran Wetland.
Key words: Miangaran wetland, ecosystem management, vulnerability assessment
 

---

Explaining the spatial patterns of drought intensities in Iran

Abstract
Recognition of spatial patterns of drought plays an important role in monitoring, predicting, confronting, reducing vulnerability, and increasing adaptation to this hazard. This study aims to identify the spatial distribution and analyze the spatial patterns of annual, seasonal, and monthly drought intensities in Iran. For this purpose, the European center Medium-Range Weather Forecast (ECMWF) data for the period 1979-2021 and the ZSI index were used to extract the drought intensities. To achieve the research goal and explain the spatial pattern of the frequency of drought intensities (Extreme, severe, moderate, and weak), spatial statistical methods such as global Moran’s I, Anselin local Moran’s Index, and hot spots were used. The results of the global Moran’s I showed that with increasing intensity, the spatial distribution of drought events has become clustered. The spatial distribution of the local Moran’s Index and hot spots also confirms this. Very clear contrast was observed in the local clusters of high (low) occurrence as well as hot (cold) spots of severe (Extreme) yearly droughts in the south, southeast, and east. In autumn, weak to Extreme droughts show a southeast-northwest pattern. But in spring and winter, the spatial pattern of drought is very strong as opposed to severe and moderate drought. Despite the relatively high variability of maximum positive spatial Autocorrelation of severe and Extreme monthly droughts, their spatial pattern is almost similar. The spatial clusters of severe and very severe droughts in the northwest, northeast, and especially on the Caspian coast, are a serious warning for the management of water resources, especially for precipitation-based activities, such as agriculture.
Introduction
Drought or lack of precipitation over some time is the most widespread natural hazard on the earth compared to its long-term average. This risk negatively affects various sectors such as hydropower generation, health, industry, tourism, agriculture, livestock, environment, and economy. To reduce these negative or destructive effects, it must be determined how often drought occurs during the period and in which areas it is most severe. Doing so requires determining the characteristics of the drought. These characteristics include area, intensity, duration, and frequency of drought. Discovering the geographical focus, recognizing the pattern governing the frequency of occurrence and temporal-spatial distribution as well as changes in the dynamics of this hazard facilitate an important role in drought monitoring, early warning, forecasting, and dealing with these potential hazards; this information can be used to create a drought plan by providing analysts and decision-makers with ideas about drought, helping to reduce the negative and vulnerable effects and ultimately make it easier to protect or replace for greater adaptation. Many researchers have been led by these approaches to the use of statistical analysis. Numerous studies have been conducted in the study of climatic phenomena such as drought with space statistics techniques in various regions, including China, India, South Korea, and even Iran. Part of the domestic research on spatial patterns of drought is without the use of spatial statistics and a limited number of others who have used these analyzes have only studied the overall intensity of drought and have not studied the spatial patterns of different drought intensities. The main purpose of this study is to identify the distribution and spatial patterns of drought intensities in Iran using spatial analysis functions of spatial statistics based on the frequency of drought intensities (Extreme, severe, moderate, and weak) with yearly, seasonal and monthly multi-scale approach. Therefore, this study will answer the questions: a) What is the spatial distribution of drought intensity data in Iran? And b) What is the variability of spatial patterns of Iranian droughts at different time scales?
Material &Method
ERA5 monthly precipitation data for a period of 43 years from 1979 to 2021 were used for this study. an array of dimensions of 78×59×504 of data were formed in MATLAB software in which 78×59 is the number of nodes with a spatial resolution of 0.25 degrees and 504 represents the month. After creating the database, the ZSI index was used to calculate the severity of drought in annual, seasonal, and monthly comparisons. Finally, to achieve the research goal and explain the spatial pattern governing the frequency of drought intensities (Extreme, severe, moderate, and weak), spatial statistical methods such as global Moran’s I, Anselin local Moran I and hot spots was used.
Discussion of Results
Due to its ecological conditions, geographical location, and location in an arid and semi-arid region of the world, Iran is among the most vulnerable countries due to natural hazards, including drought. It has experienced many severe droughts in the last century. The occurrence of drought and its effects is one of the major challenges of water resources management in this century. The results of the Global Moran’s Index for all three annual, seasonal, and monthly scales showed a highly clustered pattern of drought events in the country. Spatial clustering of the occurrence of severe and Extreme yearly droughts in the eastern, southeastern, and southern regions is also an interesting result. These conditions are due to low precipitation and high spatial variation coefficient in these areas. This contrast of spatial clusters of drought intensities indicates the relationship between drought and temporal-spatial anomalies of precipitation so that with increasing precipitation, spatial variability of precipitation decreases, and consequently spatial homogeneity of precipitation increases. severe and moderate-intensity spots in the south-southeast in autumn and spring can be affected by fluctuations in the beginning and end of the monsoon season in South Asia due to the high variability of atmospheric circulation at the beginning and end of precipitation in these areas. Some studies have also shown the relationship between precipitation in these areas and the monsoon behavior of South Asia. Extreme drought events in winter and spring have had a positive spatial correlation pattern in the southwest, west, and northwest. However, precipitation at this time of year is concentrated in these areas. Warm clusters or concentrations of very severe drought events in the northern strip of the country, especially in the Caspian region, can be due to the high variability of precipitation at the beginning of the annual precipitation season (late summer and early autumn).  Observations of these conditions in the northern strip indicate that an event with a high frequency of severe droughts, even in rainy areas, should not be unexpected. Spatial clusters of Extreme, severe, moderate, and weak drought every month using both local Moran and hot spots statistics show the fact that in Iran, the most severe droughts have occurred in the western, northwestern, and coastal areas of the Caspian Sea. However, the absence of severe droughts or spatial clusters has been the occurrence of low drought in the southeast and to some extent in the south. On a yearly scale, the south, southeast, and east have played a significant role in the spatial cluster of severe and extreme droughts. So that these areas of the country have had positive spatial solidarity. However, in these areas, negative spatial correlation prevailed in the autumn for severe drought. This may indicate an anomaly and a tendency to concentrate more precipitation in Iran, as well as many changes in seasonal and local precipitation regimes. According to the research results, a high incidence of severe and extreme drought on all three scales (monthly, seasonal and annual) even in the wettest climate of the country (northern Iran, especially the southern shores of the Caspian Sea) shows that High-intensity droughts can occur in all parts of the country, regardless of the weather conditions.
Keywords: Natural hazards, spatial patterns, Moran statistics, spatial autocorrelation, hot spots


 

---

Climate is one of the determining factors in the quantity and quality of agricultural products, therefore, in this study, the relationship between precipitation and temperature (as explanatory variables) with rice yield in 40 cities and wheat yield in 30 cities (as dependent variables) was investigated in the Caspian coastal area during 2000-2017. Spatial statistical analyses were performed with using the Moran autocorrelation test and geographically weighted regression. Based on the results (Moran index, z = 0.4342121 for rice and z = 0.719571 for wheat, respectively), it was revealed that the spatial distribution pattern of rice and wheat yield had a cluster pattern. The results of the geographic weighted regression analysis showed that the temperature increase was more desirable than the precipitation increase so the increasing temperature could lead to yield increases. In the eastern parts of the study area, the positive effect of precipitation on rice yield (with 0.020 to 0.540 regression coefficients) was remarkable; the results also revealed a negative relationship between temperature and rice yield in the southeast and eastern parts and a positive effect on rice yield in other areas. Also, the effect of precipitation on wheat yield was negative in the west and central parts of the study area (with -0.481 to -0.871 regression coefficients). According to the results, a negative relationship was dominant between temperature and wheat yield in the east and southeastern parts of the study area and a positive relationship was detected in other areas. Finally, the results indicated that in the western and central parts, due to heavy rainfall and a low number of sunny hours, an increase in temperature is more favourable than an increase in rainfall. In the eastern and southeastern regions of the region, where the amount of precipitation is lower than the threshold required for rice and wheat, an increase in precipitation is more desirable.