JWSS - Journal of Water and Soil Science

Agricultural water management studies require accurate information on actual evapotranspiration. This information must have sufficient spatial detail to allow analysis on the farm or basin level. The methods used to estimate evapotranspiration are grouped into two main groups, which include direct methods and indirect or computational methods. Basics of the indirect methods are based on the relationship between meteorological parameters, which impedes the use of these data with a lack or impairment. On the other hand, this information is a point specific to meteorological stations, and their regional estimates are another problem of uncertainty of their own. To this end, the use of remote sensing technology can be a suitable approach to address these constraints. Real evapotranspiration can be estimated by satellite imagery that has short and long wavelengths and is estimated using surface energy equations. Examples of such algorithms include SEBAL, METRIC, SEBS. Among the above mentioned algorithms, SEBAL and SEBS have been used. Among the factors of superiority of the SEBAL and SEBS algorithms, in comparison with other remote sensing algorithms, is a satellite imagery analysis algorithm based on physical principles and uses satellite simulation and requires minimum meteorological information from ground measurements or air models. 

Snow cover is one of the important climatic elements based on which climate change may have a special effect. In general, climate change may be reflected in different climatic elements. Therefore, it is very important to study and measure changes in snow level as one of the important sources of water supply. Ardebil and Sarein cities are located at 48° 18׳ east longitude and 38° 15׳ north latitude. In this study, Sentinel-2 optical satellite was used to monitor the snow cover surface in 2018, and NDVI, S3, NWDI, NDSI, Cloud mask indices were applied to detect snow-covered surfaces using ArcGIS and Snap software. Next, to validate the snow maps extracted from the images, it was compared with the snow data in terrestrial stations using linear regression in MATLAB software and to evaluate the accuracy of the model statistical indices including RMSE, MSE, BIAS, CORR were used. The present study showed that according to Ardabil city climatic conditions, maximum-snow covered area in January with an area of 356.52 km2 and minimum snow-covered area in March with an area of 96.10 km2. The highest snow cover is observed in the high slope areas in the western slopes (Sabalan Mountain Heights) and the lowest snow cover is observed in the lower eastern slopes. The results of linear regression with generalization coefficient are 85% and the results of statistical indices of error are equal to MSE: 0.086, BASAS: 0.165, CORR: 0.924, and RMSE: 0.03. Correlation relationships between terrestrial data and estimated snow maps showed a high degree of correlation. This result is statistically significant at the 99% level. The use of optical images in estimating snow levels is very cost-effective due to the size of the areas and the high cost of installing snowmobiles. The results obtained in the present study indicated that traditional radar images with high spatial resolution and good correlation with terrestrial data can be a good alternative to snowmobiling ground stations at high altitudes or in passable areas.

change, accurately predicting wheat production is essential for developing precision agriculture. Remote sensing enables the indirect prediction of crop production before harvest. This research investigates the application of the random forest method and support vector regression for simulating wheat production across ten selected farms in Qazvin Plain from 2019 to 2020, employing NDVI, MSAVI, and EVI vegetation indices. Sentinel 2 satellite data was utilized for the vegetation indices. Production data for the ten wheat fields was obtained from the Agricultural Jihad Organization of Qazvin Province. Evaluation of support vector regression and random forest to assess both the observed and simulated wheat production data was conducted using R2, MBE, RMSE, and MAE statistics. To explore the simulation of wheat production using vegetation indices, seven methods were defined: methods 1 to 3 examine each index separately; methods 4 to 6 focus on binary combinations of the indices; and method 7 considers the combined effects of all three indices. The support vector regression model provided good estimates of wheat production in all methods, except methods one and four, in the test phase, with a coefficient of determination of more than 0.98 and a low RMSE. The random forest model showed significant results in all methods except methods two and six during the test phase, achieving a 95% probability (P-value=0.00) with a coefficient of determination greater than 0.8. Overall, this research highlights the importance and potential of machine learning techniques for timely crop production prediction as a strong foundation for regional food security.