Estimation of actual evapotranspiration in pistachio orchards using SEBAL algorithm in three irrigation system

Background and Objective:

Over the past 100 years, the country has lost about 90 percent of its per capita renewable water. About 90% of the country's renewable water resources are allocated to the agricultural sector. With the increase in the area of pistachio orchards and the increase in demand for water on the one hand and the limited water resources in the region, on the other hand, the imbalance between supply and demand for water is sharply increasing. In this regard, the most important step to prevent water loss is the uniform distribution of water on the field, optimal at each stage of growth. About 99% of the water absorbed by the plant is used for evapotranspiration. Therefore, studying this phenomenon can play an important role in determining the water needs of plants. It is difficult to measure the actual evapotranspiration outside the laboratory. Many experimental methods have been developed to estimate actual and potential evapotranspiration using meteorological and climatic data. But most of these methods are only able to estimate potential evapotranspiration and do not estimate the actual amount of it.  In contrast, remote sensing methods have been developed that are a good solution for estimating the actual evapotranspiration. Satellite imagery with global coverage and repetitive Acquisition has made it possible to monitor evapotranspiration at the field level and during plant growth. Various studies have been conducted to estimate the actual evapotranspiration of agricultural areas using satellite images, which indicate the acceptable accuracy of these methods. However, most of this research is related to agricultural fields and no significant research has been done to estimate evapotranspiration at the orchards. Vegetation at the farms is uniform and homogeneous compared to orchards, so the estimation of vegetation index, which is one of the inputs of the SEBAL model in orchards is more difficult than agricultural fields, which can affect the final accuracy. Therefore, the main purpose of this study is to estimate the amount of evapotranspiration in the pistachio orchard using the SEBAL algorithm and evaluate the accuracy of estimation. Also, this research has been

Materials and Methods :

The present research has been carried out in pistachio orchards in Zarandieh city of Markazi province. The gardens had three different irrigation systems including flood irrigation systems, surface, and subsurface drip irrigation systems. Actual evapotranspiration is estimated using water balance and SEBAL algorithm. Meteorological data from Imam Airport Synoptic Station and Landsat8 satellite imagery has been used to estimate evapotranspiration using the SEBAL algorithm. Actual evapotranspiration is estimated at satellite overpass times during the growing season. To select hot and cold pixels in the SEBAL algorithm, the semi-automatic method proposed by Oldmo is used, which minimizes user participation in the selection of hot and cold pixels. To evaluate the accuracy of evapotranspiration estimation, the information of soil moisture sensors in the orchard has been used. 28 sensors measure soil moisture in different parts of the orchard. Using the soil moisture values, the actual evapotranspiration was estimated using the water balance method and used as a reference value.

Results and Discussion:

 A comparison of the results of the SEBAL algorithm and water balance method showed that the SEBAl algorithm was able to estimate the actual evapotranspiration in different parts of the orchard with an RMS error of 0.57. In addition, the correlation between the values estimated by the two methods was equal to 0.82, which indicates the appropriate capability of the SEBAL algorithm in estimating evapotranspiration values. The correlation between the actual evapotranspiration estimated from the SEBAL model and the reference evapotranspiration is 0.76. In addition, in the research, changes in the evapotranspiration in different parts of the garden and also gardens with different irrigation systems including flood, surface, and subsurface drips have been investigated. The results show that the orchard with subsurface irrigation had the lowest average of evapotranspiration on different dates. Considering that evapotranspiration is equal to the sum of evaporation from the soil surface and transpiration from the plant, this decrease can be attributed to the decrease in evaporation from the soil surface. In addition, evapotranspiration heterogeneity can be observed in all parts of orchards with the same irrigation system on all dates. For example, in the orchard with a flood irrigation system, parts of the garden show low evapotranspiration, which can be due to the lack of smoothing of the surface and lack of proper moisture in these areas. Obviously, the same amount of moisture accumulates in other parts of the garden and is inaccessible through deep percolation. This uneven distribution is also observed in the garden with a surface drip irrigation system. For example, the middle part of the garden with surface drip irrigation always shows a higher amount of evapotranspiration, which can indicate the loss of water in this part, due to the miss-operation of the dripper. To evaluate the difference in evapotranspiration in different irrigation systems, the average, minimum, maximum, and standard deviation values of evapotranspiration in orchards related to three different irrigation systems have been calculated. The results showed that in all dates, the ranges and standard deviation of evapotranspiration in the flood irrigation system were higher than in other systems, which indicates the lack of uniform irrigation in the orchard. Also, on all dates, the average amount of evapotranspiration in the orchard with a surface drip irrigation system has been more than flood irrigation system. Vegetation in orchards with drip irrigation systems (surface and subsurface) was denser compared to the flood irrigation systems.

Conclusion:

 In this study, the actual evapotranspiration of pistachio orchards has been estimated using satellite imagery and the SEBAL algorithm. The results of the study indicate the appropriate accuracy of the SEBAL algorithm in estimating the actual evapotranspiration of the orchards. Compared with the water balance method, the correlation coefficient was 0.82 and the root means the square error was 0.57. In addition, comparing the moisture situation in different parts of the orchard and in orchards with different irrigation systems has shown that by estimating the actual evapotranspiration using satellite imagery, appropriate information can be obtained on how to distribute moisture in the garden. This information provides valuable information on the optimal management of water resources and increases irrigation efficiency. Other results of this research include the significant difference between surface and subsurface drip irrigation methods. The results show that using subsurface irrigation methods can effectively reduce irrigation water loss due to evaporation from the soil surface. The results show that in areas where there is no access to information from soil moisture sensors or direct measurements of evapotranspiration, the use of the SEBAL algorithm and remote sensing methods can provide appropriate information for optimal water management.