Electrical resistivity tomography (ERT) measurements on date palm stems to support irrigation scheduling

Abstract

Commercial cultivation of date palms has high economic significance, particularly in arid and hyper-arid regions. In Israel, the commercial cultivation of date palms in the hyper-arid Arava and Jordan valleys (from the Sea of Galilee, along the Jordan valley to the Red Sea) has grown substantially in the past two decades. As such, achieving commercially viable yield of dates amid water scarcity and enforced restrictions on water use requires optimising the amount and timing of irrigation. To that end, better understanding of the temporal dynamics of the water use of date palm is of great importance. Traditionally, this can be achieved through sap flow estimates obtained using heat dissipation probes. However, such measurements lack the ability to provide information on the spatial distribution of sap flow, which is important considering that date palms transport water in the entire stem cross-section. Electrical resistivity tomography (ERT), a long established and widely used geophysical method for the imaging and characterisation of sub-surface structures may allow for obtaining improved estimates of transpiration from sap flow measurements and the assessment of stem water dynamics. The aim of this thesis is thus to explore the potential of combining sap flow and electrical measurements to obtain a better understanding of the transpiration and the spatiotemporal dynamics of water flow and storage in date palm trunks. In a first step, a suitable ERT set-up and data acquisition strategies were developed starting from ERT measurements on a sand column as a proof-of-concept, where ERT was used to monitor the progress of a tracer solution through a sand column. The results showed that the adopted strategy for ERT measurement acquisition was successful in capturing the development and breakthrough of the introduced tracer solution, not only in terms of the average real part of the electrical conductivity at each electrode plane, but also in terms of the spatial distribution of the real part of the electrical conductivity. The outcome of the flow experiment on a sand column was used as a starting point to further develop the ERT set-up and data acquisition strategies for measuring on date palm stems, which were then tested on a date palm stem segment. For this, flow was induced through a stem segment that was obtained from a felled date palm tree. ERT measurements were continuously obtained throughout a cycle of flow and no-flow periods. The results showed that the mean bulk electrical conductivity varied strongly due to changes in the flow conditions. In addition, it was found that the electrical conductivity of the outflow was much higher than that of the inflow, which indicates the release of stored salt from the stem segment. Analysis of the spatial distribution of the electrical conductivity suggested that flow mainly occurred in a limited part of the cross-sectional area of the stem. Extending from the flow experiment on a date palm stem segment, the electrical properties were further investigated using a multi-step outflow experiment on a smaller stem fragment. The results showed a slow water release from the date palm stem segment with increasing applied pressure, which suggests that the water is tightly bound in the stem as in a clay soil. The real part of the bulk electrical conductivity of the stem segment showed a declining pattern, which is generally in agreement with the decreasing water content in the stem segment. The real part of the electrical conductivity decreased with the saturation of the stem segment in a pattern that closely resembled that of a clay soil. From the laboratory experiments, it was concluded that ERT is a promising tool to investigate the spatial variability of water flow in date palm stems. In a second step, two field experiments involving irrigation deprivation were conducted on living date palms. In both experiments, ERT measurements were obtained at high temporal resolution following the established set-up and data acquisition strategies from the experiments in the laboratory. Sap flow estimates were obtained from heat dissipation probes in conjunction with continuous ERT measurements. The first experiment involved a juvenile date palm growing in a lysimeter, while the second experiment involved a mature date palm. The combined sap flow and ERT measurements were continuously obtained for a baseline period, followed by an induced water stress period. The monitoring was continued during the recovery period following the restoration of the irrigation treatment for a few days. It was found that the diurnal variations in the electrical properties of the date palm stem during the baseline period were consistent with the expected diurnal variation in transpiration and stem water storage. The findings also made clear that capturing those diurnal variations would not be possible without following a measurement acquisition strategy utilising a high temporal resolution. During the irrigation deprivation period, the juvenile date palm showed a strong response to water stress, as observed in the temporal dynamics of the average real part of the electrical resistivity over the plane as well as the average real part of the transfer impedances. The mature palm, on the other hand, showed no clear signs of water stress during the irrigation deprivation period. This finding was consistent with the results obtained on soil water content, which showed that the loss in water was superficial (above 60 cm depth). Furthermore, the observed spatial variation in the real part of the electrical resistivity showed that some regions of the stem cross-sectional area were more active than others, and that the distribution of areas with high and low electrical resistivity was dynamically changing throughout the day and night. This observation suggests an on-going redistribution of the stem water content and continuous changes in stem water storage. It was further observed that the areas with high electrical resistivity noticeably increased following the suspension of the irrigation treatment. A closer examination of this finding indicated that this was more evident for the juvenile date palm than for the mature date palm. The obtained sap flow estimates and ERT data showed no signs of quick recovery in case of the juvenile date palm after the irrigation was restored. However, subsequent monitoring showed that the juvenile date palm did eventually recover after a few weeks. The results further showed that the mature date palm did not experience significant water stress and by extension recovery. A long-term monitoring on a bi-monthly basis on the mature date palm showed a temporally stable spatial distribution of the real part of the electrical resistivity for measurements made at the same time for a period of several months. This finding provides further confidence in the established ERT set-up and data acquisition strategies as an approach to support irrigation scheduling for date palms and detecting early signs of water stress. In conclusion, the work presented in this thesis provides an important contribution to the establishment of a combined approach in which sap flow estimates obtained from heat dissipation probes and continuous ERT measurements are jointly used to obtain improved monitoring and interpretation of water flow and storage in date palm stems. It was shown that the proposed set-ups, ERT data acquisition protocols and inversion strategies allow for reliably obtaining representative ERT measurements on date palm stems that enabled monitoring of spatiotemporal variability in water flow and storage in date palm stems. The proposed approach also allows for improved estimation of daily variation in transpiration requirements, and detecting of early signs of water stress, which can make a valuable contribution to the planning and scheduling of irrigation treatments that account for transpiration requirements of date palms while adhering to imposed limitation in water use.