| Summary: | IntroductionSoil salinization is a global environmental problem with serious economic, social and economic consequences. Measuring soil salinity includes the concentration of all salts dissolved in the soil, generally expressed in units of electrical conductivity (EC). Determining where, when, and how soil salinity occurs is essential to determining the sustainability of land use and development systems. Due to the ability to repeat and capture a wide range of remote sensing images, this technology will be useful for detecting changes even in short periods of time, and as an essential tool in monitoring soil salinity, it provides very valuable information on the size of the captured pixels. Flooding from this rainy season can have a large impact on soil salinity. The purpose of this research is to extract soil surface salinity with high spatial resolution and the effect of heavy rains and spatial analysis of the resulting anomalies in Fars Province using satellite image processing. Materials and Methods Fars Province is located in the south of the central region of Iran with an area of 122.799 square kilometers. The topography of the province consists of mountains and plains. In this province, eight million ha of land are suitable for agriculture and gardens, although only 1.6 million hectares have been used. The agricultural sector in Fars Province, which accounts for a major share of the national gross product, plays one of the most important roles in Iran's production, employment, and food security, so many of the province's agricultural products, such as cereals and citrus fruits, rank first to third in the country. Since our study was carried out in a wide area of the country, it was decided to use Google Earth Engine (GEE) as an open-source platform. Also, the Generalized Difference Vegetation Index (GDVI) prepared by Wu (2014) was used to analyze soil salinity. In order to evaluate the efficiency of the obtained model, R2 and RMSE indices were used. In order to verify the output of the field data collected from the Agricultural and Natural Resources Research Center of Fars Province, which was used as a ground sample for the evaluation By using spatial analysis in the form of geostatistics, spatial structures can be identified and spatial planning can be done. Results and Discussiom For the studied area, soil salinity was between 7.01 and 53.63 decisiemens/meter. The difference between the highest and the lowest soil salinity in the study area is approximately seven times, the highest value being in the east and south of the province in the cities of Niriz, Larestan, Lamard and Zarindasht. A significant point is the sharp increase in soil salinity in the bed of rivers leading to Bakhtegan and Tashk lakes. According to the available ground data, the accuracy of the map was checked, and the square root of the error and the correlation coefficient were calculated as 0.33 and 0.59, respectively. Then, the soil salinity map was extracted using the same algorithm in the period of Farudin 2018 due to the heavy rains that were associated with the arrival of numerous rain systems in Iran. Soil salinity was obtained between 6.35 and 47.9 and was classified into five classes. The results showed that changes have been made in the minimum and maximum values of salinity and soil salinity levels and soil salinity has decreased especially in the south of the province.Then soil salinity anomaly was obtained and spatially analyzed. The term soil salinity anomaly means deviation from the reference value or long-term average. The results showed that the amount of abnormality increased and decreased between 0.8 and -0.9 in the province. Areas with lower salinity have experienced a greater share of positive anomalies. The positive anomaly was mostly around Darab, Zanian and Babamonir in the northeast of Jahrom. The southern and eastern parts, including Lar, Ozer, Rastaq, Ahl and Lamard, which were in the medium salinity class, have suffered less salinity anomalies. In order to understand the cluster or scatter pattern of soil salinity changes, Moran's spatial autocorrelation coefficient was investigated. The results showed that the anomaly of salinity distribution in the rainy year has a cluster pattern. By examining the available maps, it can be said that the clusters of soil salinity anomalies are mostly located in the north of the province, Baba Monir and Zarian at higher altitudes of the province and to some extent in the south of the province around Darab and Jahrom. Also, a little clustering has occurred in terms of anomalies in the plains of the province; That is, the rains could not cause major changes in the soil salinity of the plains. ConclusionIn this research, the soil salinity map using GDVI in two time periods before and after the heavy rains of the water year 1398-1397, using the open source platform Google Earth Engine, extracting and changing the soil salinity classes and converting the salinity classes to each other, as well as the method of spatial clustering. The salinity anomaly was investigated. Soil salinity for the studied area was calculated between 7.01 and 53.63 decisiemens/meter with square error and correlation coefficient of 0.331 and 0.59, respectively. Soil salinity has changed between 6.35 and 47.9 after heavy rains. The most changes due to heavy rains are related to the low salinity layer with 19% and the least changes are related to the very saline layer with 0.3%. The amount of anomaly between 0.8 and 0.9 decisiemens per meter was increasing in the center of the province around Bakhtegan and Tashk lakes and the western highlands of the province and decreasing in the south and east of the province. Areas with lower salinity contribute more positive anomaly. The southern and eastern parts, which have high and very high salinity, undergo less changes. The results of this study showed that the use of remote sensing and satellite data in the Google Earth Engine cloud system and spatial analysis to prepare soil salinity maps in areas that have a large area and are affected by salinity changes, has great financial and time savings, and in Areas where sampling is not done or is associated with issues can be very efficient. Such researches can easily and quickly identify areas that are most exposed to increasing or decreasing soil salinity and can be used in environmental planning to implement preventive measures.
|
|---|