Site-specific irrigation: Improvement of application map and a dynamic steering of modified centre pivot irrigation system

Abstract

Introduction: A management concept for sustainable utilization and the efficient use of agricultural inputs is known as “Precision Agriculture” (PA). The PA concept, when applied to irrigation management is known as Precision Irrigation (PI). In PI, the need for irrigation may differ between zones of a particular field due to the spatial variation of soil properties or the cropping of different plants on the same field. Spatial variation of total available water content (TAWC) as a primary factor causes spatial variation of irrigation depth and frequency within fields. While moving irrigation systems apply water at constant rates, some areas of the field may receive too much water and others not enough. In this regard, precision irrigation (PI) is capable of applying water in the right place in the right amount at the right time using the right irrigation system. Therefore the key objectives of the present study were a) Delineation of irrigation management zones (IMZs) using sensor-based soil electrical conductivity (ECa) measurement with the aid of EM38 and VERIS 3100, b) Developing and evaluating a precision mobile drip irrigation (PMDI) and c) Evaluating wireless EnviroSCAN sensors and AMBAV-models to measure the soil moisture content.
Materials and methods: EC25 data (ECa in 25° C) were collected using EM38 and VERIS 3100 at field capacity on a 16.6 ha non-saline field in the FAL, Braunschweig, Germany. ECa data were obtained in 1-s intervals corresponding to a 2 to 3 m data spacing on transects spaced approximately 4 to 6 m apart. An ArcView (ESRI) software program was used to create the EC25 and TAWC maps after the readings were interpolated using a spherical kriging model. 29 calibration points taken at a depth of 0 - 60 cm depth were located using DGPS based on the ECa spatial variability pattern and with the objective of covering the whole range of ECa values present to determine the best sensor-based method to monitor TAWC. The second span of the centre pivot irrigation machine (CP) was modified to PMDI and controlled for variable-rate water application with a pulsing technique by installing solenoid valves (SV), programmable logic control (PLC) and using a Siplast drop tube instead of sprinklers. One quarter of the study field was controlled by the EnviroSCAN soil moisture sensor and another quarter was controlled by the AMBAV-model to determine irrigation depth. In addition, the hydraulic performance of the Siplast drop tube was evaluated in the laboratory by collecting discharge rates at different pressure of 50, 100, 150 and 200 kPa.
Results and discussion: This study showed that, while qualitatively similar, EC25 data obtained with different commercial sensors were quantitatively different because of different depth-weighted response functions. The highest coefficients of determination (R2) were generally found between EM38_h and EM38_v (R2 = 0.55). In this study, a better value of R2 between TAWC and the VERIS 3100 readings was found. The R2 value from VERIS 3100-sh data for TAWC estimation was maximally (0.77) and matched the TAWC data quite well, whereas R2 values to EM38-h and EM38-v data were low and apparently could not adequately reflect the spatial variability of the TAWC due to the higher influence of the EM38 on deeper layers. Six IMZs (IMZ1: 99 to 105, IMZ2: 105 to 116, IMZ3: 116 to 127, IMZ4: 127 to 138, IMZ5: 138 to 149 and IMZ6: 149 to 152 mm/60 cm) were identified based on fuzzy-k-means unsupervised classification as an optimum number of IMZs within the study field. It was concluded that under conventional uniform irrigation, IMZ1 and IMZ2 were over-irrigated, whereas IMZ4, IMZ5 and IMZ6 were under-irrigated. The developed concept of pulse irrigation was a feasible and a viable technique. Water application was directly proportional to the fraction of time the valve was opened as the system was capable of controlling fifteen banks of fifteen nozzles. There were no apparent problems with the pulsing water delivery system where the field tests were conducted. CP speed and the pulsing technique used to deliver variable amounts of irrigation had little adverse effect on system uniformity and the nozzle flow rate. Uniformity coefficients were reduced by decreasing the pulsing level and increasing CP speed. The control unit was able to monitor wireless soil moisture sensors via radio telemetry and communication from the EnviroSCAN sensors to the central ISM modem, which worked as expected. Although the EnviroSCAN soil moisture sensor was found to be delicate and intricate to use and calibrate, soil moisture data were easily sent from the control unit and received by the mobile phone and then transferred to an Excel table on a computer using easy and suitable “Kurznachricht Pro 2.2” software to calculate irrigation depth. The results suggest that EnviroSCAN sensors are able to follow the general trends successfully as soil water content measured by sampling changed during the growing season, but are not a reliable sensor to repeat moisture conditions on sandy soils (at greater depths than 40 cm ) despite its soil-specific calibration. Meanwhile, an AMBAV model as a cheap and reliable alternative instead of the expensive EnviroSCAN sensor was capable of determining and simulating soil moisture in the root zone of grass crops. Drip irrigation design should be based on reliable data sets, but not on data supplied by the manufacturer. The laboratory experiments showed that the effect of operating pressure on the discharge of Siplast emitters was highly significant and the emitter discharge was strongly influenced by the operating pressure, while some deviation from the design flow rate claimed by the manufacturer occurred. CV values were classified as good, on the basis of the ISO standard. Based on the laboratory experiments, it was found that the in-line Siplast emitter has high emission uniformity and a low coefficient of variation. In spite of high emission uniformity and a low coefficient of variation of the Siplast drop tube, it must consist of hard and inflexible material. To have a shorter drip tube installed on CP, using an in-line drop tube lateral with higher emitter discharge at low operation pressure and less emitter distance is proposed. The economic analysis of this study showed that although capital requirement per hectare under PMDI is about € 338 and € 250 more than for drip irrigation in Germany and Iran, respectively, it causes perceptibly less annual fixed cost than drip irrigation (111 and 128 [€/(ha x year)] cheaper than drip irrigation in Germany and Iran, respectively). Although PMDI causes more annual fixed expenses than CP irrigation, it has less total irrigation cost per hectare and year than CP and drip irrigation and has the potential benefit to increase yield quantity, quality and farming benefit. The results showed as an important policy implication that PMDI is not necessarily a water saving technology and it does not necessarily involve a reduction in total water use, but that it can optimize water consumption. Given a reduction of energy and water consumption of 70 % and 25 %, respectively, achieved by the PMDI as compared with the CP, results showed that about 575, 378, 462 and 588 kWh energy per hectare can be saved by PMDI in comparison with the conventional CP irrigation of lettuce, sugar beet, potato and strawberry.
Conclusion: Sensor-based ECa measurement at F.C. in non-saline soil can be used as a cheap, rapid and non-destructive alternative to delineate IMZ instead of using soil sampling and aerial photography methods. Field studies using larger irrigation systems and fields with different soil types, topographic or crop characteristics are recommended to validate the precision irrigation concept and to realize and ensure a positive net economic return to the producer. With due attention to the success of PI in the early stages and developments in industrial technology in the coming years, the extra costs of industrial accessories could be minimised.