Improving the Irrigation Scheduling of Drip Irrigation System Using Field Measurements and Crop Modeling

Introduction

Agriculture is the largest consumer of fresh water in the whole world. Currently, almost 11% of the Earth’s total land surface is arable, which is expected to reach 13% by 2050. Roughly 17% of these lands are subject to modern form of irrigation management, which constitutes about 30-40% of the gross agricultural output. Reducing water consumption and increase water productivity in agriculture, requires a correct understanding of the biological response of crop to water. The ever-increasing growth of the population and the limitation of fresh water resources have led irrigation experts to new and efficient approaches in making decision to increase water productivity. The lack of scheduling in irrigation or their incompatibility with weather conditions, soil, irrigation system, agricultural restrictions and different phenological stages of crop, has caused severe losses in irrigated fields. Crop modeling and field data measurement can help irrigation experts improve irrigation scheduling in field and reduce water losses. Coupling in-situ measurement and crop modeling during the growing season is one of the useful solutions to improve irrigation scheduling in different farm conditions. In this paper, AquaCrop was calibrated for maize (Zea mays L.) in research farm of Ferdowsi University of Mashhad with comprehensive dataset. Variations of soil water content at different depths and also different growth indices was monitored during one growing season. The main novelty of this research is the targeted use of AquaCrop software to reduce water consumption by knowing the phenologically sensitive stages.

Materials and Methods

In-situ and in-lab measurements along with plant modeling, have been the two main parts of this research. First, the input files of the AquaCrop software were prepared and calibrated for maize during one growing season. The outputs of software including variations of moisture, biomass produced during the growing season and final yield were compared with the values measured in the field. After ensuring the accuracy of calibrated software, an improved irrigation scheduling was investigated. Subsequently, by means of crop modeling, the sensitive intervals of the crop to soil water stress as well as the thresholds of yield reduction in different stages of growing season were determined with the aim of improving irrigation scheduling in maize field. Biomass reduction, dry yield production, and the changes in water use efficiency during the growing season were investigated according to different amounts of moisture reduction in root zone, and irrigation scheduling was fulfilled by applying stress to less sensitive stages. The software was first run in Net Irrigation Requirement mode. Then, different amounts of drought stress were applied to each of the growth stages of the crop, and the other stages were kept constant in non-stressed condition. Dry yield and biomass reduction as well as water productivity changes were obtained for each stage. According to the threshold values and the amount of yield reduction at each stage, it is possible to fine-tune the time and amount of applying stress to crop. Furthermore, according to the moisture profile of the root zone, the drained water was significantly reduced in field. In this research, in order to increase accuracy in moisture measurement and prevent errors, an equation was developed to convert the mV output of sensor to volumetric moisture. At each moisture measurement, the PR2 sensor reports a number in mV for each depth, which is converted to volumetric moisture using device's conversion equation. This equation can vary according to field and soil condition. To develop the conversion equation, six same pots with a height of 40 cm and diameter of 30 cm were filled with the desired field soil, and the access tubes were placed horizontally inside them. The soil inside these pots was completely saturated and exposed to air for drying. During the drying period of the pots, the soil moisture was measured regularly using PR2 sensor and at the same time by weight method. After these measurements, calibration curves for PR2 sensor were obtained using the alpha mixing method for different depths.

Results and Discussion

The Pearson correlation coefficient for measured and simulated moisture by software, was 0.84 and the root mean square error was 12 mm. Also, the value of Pearson correlation coefficient for measured and simulated values of biomass was equal to 0.99 and mean square root of error was 1.3 ton/ha. The results showed that the vegetative stage of crop was sensitive to drought stress and caused a significant yield reduction at the end of the growing season. The stage of germination and flowering were less sensitive, such that the decrease of moisture up to 12.3% compared to the Net Irrigation Requirement mode would not change the final yield of the crop. The improvement made in the field resulted in no change in the amount of biomass and dry yield (0.38% increase in biomass and 0.52% increase in dry yield), at the same time 26.6% decrease in depth of irrigation water, 85.6% decrease In drainage and increasing the efficiency of evapotranspirated water, it has increased from 4.66% to 4.67% kg/m3 compared to condition before modify. Before modify, the farm was managed in traditional way.

Conclusion

More research in the field of irrigation scheduling and providing comprehensive instructions to farmers in different weather conditions and different irrigation systems, taking into account different management approaches in allocating water to different parts of the farm along with investigating the effects of saline water irrigation, can be topics for the next research should be for experts in this field.