I. INTRODUCTION

Tree health has traditionally been performed based on human visual observations ^[1]. In Fig. 1, the tree on the left is a healthy tree with high physiological activity, and the tree on the right is an image of an unhealthy tree. Because it relies on human observation, it was necessary to develop equipment that could measure it quantitatively.

Fig. 1. Visual differences depending on tree health.

The first work to quantify and assess tree physiological activity using electrical resistance was carried out by Shigo in 1977 ^[2]. Shigo attempted to quantify tree physiological activity by measuring the electrical resistance. The Shigometer^TM is still used to quantitatively study tree's physiological activity ^[3]. Shigometer^TM measures resistance one-time by sticking two needles into a tree. Because the two-point measurement method is used, in principle, an error occurs equal to the contact resistance formed between the needle and the wood. To overcome the shortcomings of the 2-contact measurement method, the 4-contact measurement method must be used.

In this paper, we introduce a tree physiological activity monitoring system (TPAM) using a 4-contact resistance measurement method. The developed TPAM is mounted on a tree and continuously measures the resistance of the tree, transmitting data wirelessly. It was applied to the resistance measurement of vegetables and actual trees, and the errors of 2-contact measurement and 4-contact measurement were compared. The results of continuously measuring the ambient temperature, light intensity, and resistance of real wood for 24 hours are presented.

Section II describes the developed TPAM system, section III describes the application results of the TPAM system, and section IV describes the conclusion.

II. PHYSIOLOGICAL ACTIVITY MEASUREMENT SYSTEM

To reduce contact resistance errors that cause errors in Shigometer^TM, a 4-point measurement method should be adopted in the measurement system. The 4-point measurement method is a well-established method that is widely used to measure sheet resistance in the semiconductor field ^[4].

Fig. 2 illustrates the 2-point measurement method and 4-point measurement method. The left side of the image presents the electrical model of a 2-point probe and a 4-point probe, while the right side of the image illustrates the equivalent circuit for 2-point resistance measurement and 4-point resistance measurement. R_c is the contact resistance generated at the interface between the probe and the tree, and R_t is the resistance value that we actually want to know.

Fig. 2(a) shows the principle of 2-point measurement. Constant current is provided by the current source, it induces voltage across the resistance components due to the current flow. Using the induced voltage, resistance values can be calculated through mathematical expressions. The resistance for a 2-point measurement can be expressed as R_2-point = R_t + R_c1 + R_c2. In this expression, the value we want to measure is R_t, but R_2-point includes an unnecessary element R_c. This becomes an error factor in measuring tree physiological activity using Shigometer^TM.

Fig. 2(b) shows the principle of 4-point measurement. The measurement principle is similar to using induced voltage, just like 2-point measurements, but the difference with 2-point measurements is the use of 4-point probes. Two of the probes are connected to the voltmeter to detect voltage, and the remaining two probes are connected to the current source. The characteristic of the voltmeter is that it has very high resistance, which means that no current flows in the direction of the voltmeter. Therefore, there is no induced voltage across R_c connected to the voltmeter. This means that R_c is effectively removed from the measurement. The resistance for a 2-point measurement can be expressed as R_4-point = R_t. As a result, the effective removal of contact resistance R_c can be achieved through 4-point measurement.

Fig. 2. Resistance measurement methods of (a) 2-point measurement; (b) 4-point measurement.

Fig. 3 shows the block diagram of the TPAM system, which consists of two parts: on-tree system and the hub. TPAM system employs a 4-point probe to measure electrical resistance continuously. Data is wirelessly transmitted using Zigbee, and the system is equipped with solar panels for energy harvesting. The probe is in the form of a gold-plated needle on the surface of copper and has a size of 1 mm in diameter and 33 mm in length.

To avoid potential light obstruction by leaves, light sensors and a solar panel are installed on the tree canopy. The solar panel generates power from sunlight, charging the battery, while the optical sensor measures ambient light intensity. Four gold-coated probes on the tree trunk measure the tree's electrical resistance.

Fig. 3. Block diagram of TPAM system.

The Microprocessor unit (MCU), responsible for measuring tree resistance using AFE4300, sensing ambient information, transmitting data to Zigbee module. AFE4300 delivers current and measures voltage for resistance sensing. The sensors record ambient temperature and ambient light intensity, and a Zigbee module wirelessly transmits temperature, light intensity, and resistance data to the hub. The MCU controls the devices which is AFE4300, Zigbee module, and sensors. The MCU, battery, AFE4300 and Zigbee module are housed in a robust enclosure to protect them from outdoor weather conditions.

In the hub, the data transmitted by the Zigbee module connected with MCU is received and plotted in real-time, making data visible to the user. There is a feature that sends notifications to user's mobile, when data reception is interrupted, to help uninterrupted monitoring.

Fig. 4 shows TPAM evaluation results using known resistance values. In the resistance range of 1 k${\Omega}$ to 15 k${\Omega}$, TPAM exhibited 2% full-scale nonlinearity.

Fig. 4. System evaluation result. x-axis is the reference resistance, and y-axis is the resistance measured by the system.

III. APPLICATIONS

It is necessary to verify whether the 4-point measurement method is effective in plant physiology activity monitoring. So, we applied TPAM to vegetables and actual trees to see if 4-point measurements were effective.

Fig. 5 shows the experimental setup for resistance measuring vegetables (tomato, cucumber) using TPAM. The experiment involves both the 2-point and 4-point measurement methods, and the resistance values of the vegetables are measured repeatedly. When measuring the resistance of vegetables, the probe was intentionally removed and reinserted each time to deliberately vary the contact resistance. This method allows us to assess how effectively the contact resistance is eliminated using the 4-point measurement technique.

Fig. 5. Experimental setup for measuring vegetables. Probes are inserted into vegetables with 5-mm interval.

Fig. 6 shows the results of resistance measurements in vegetables. For each area of tomato and cucumber, the left box represents the 2-point resistance measurement results, and the right box represents the 4-point resistance measurement results. The symbol R_r represents the difference between the maximum and minimum resistance measurements. The symbol R_c is the difference between the average resistance value by the two-point measurement method and the average resistance value by the four-point measurement method. The R_c value represents the contact resistance value between the probe and the vegetable. Detailed figures are summarized in Table 1. The symbol E_r is the repetition error rate, which means the ratio of the distribution range of repeated measurements to the average resistance value measured by the four-point measurement method. The symbol E_c is the error rate of probe contact resistance and means the ratio of the difference between the average resistance value according to the measurement method to the average resistance value measured using the 4-point measurement method. The repeatability error (E_r) ranges from 202% to 223% for the two-point method and from 67% to 100% for the four-point method. It was found that the repeatability error of the 2-point method was at least 2.2 times and up to 3 times greater than that of the 4-point method. In the two-point measurement method, the error (E_c) due to contact resistance was found to range from 63% to 202%. As a result, the errors associated with the 2-point measurement method can be effectively mitigated by using the 4-point measurement method.

Fig. 6. Results of resistance measurements in vegetables.

Fig. 7 shows the TPAM system applied to a real Zelkova tree. The probes, four in total, are inserted into the cambium zone in tree trunk and coated with high-resistance epoxy to provide insulation and protect them from rain. The transmitter contains the measuring PCB and Zigbee module, which transmit the collected data to indoors. In the hub, the data is plotted in real-time, enabling users to observe changes in the data as they occur.

Fig. 7. A TPAM system applied to real Zelkova tree.

Fig. 8 shows the measurement results conducted on Zelkova branches. The results show a significant increase in errors due to contact resistance. It means that using the Shigometer^TM to measure the resistance of trees and assess tree physiological activity is not reliable. In the other words, Shigometer^TM using the two-point method are inappropriate for monitoring tree vitality.

Fig. 8. Results of impedance measurements in Zelkova branch.

Fig. 9 shows the results obtained from TPAM installed on Zelkova for one day. Resistance, ambient temperature, and light intensity were measured. In the results, the resistance values vary with the amount of ambient light and temperature. Intermittent measurements are not meaningful because resistance values vary by about 12% (100 ${\Omega}$) during the day. Therefore, when monitoring tree physiological activity, continuous measurements are necessary.

Fig. 9. Results obtained from TPAM installed on Zelkova for 24-hour.

In Table 3, the performance of the proposed TPAM is summarized and compared with previous work. The TPAM aims on continuous measurements with 4-point probing on real trees. Hence, wireless communication was adopted for on-tree installation in the design. TPAM measures the electrical resistance of the cambium layer, making it suitable for use on most dicotyledonous trees of a certain age or size. In reference ^[4], Miccoli adopted Bluetooth communication, but Zigbee was chosen for TPAM as it provides a longer communication range and is more suitable for connecting with a larger number of trees simultaneously.

IV. CONCLUSION

A TPAM system to quantitatively observe the physiological activity of trees was developed. The developed TPAM can harvest energy from sunlight, measure the tree's electrical resistance, temperature, and light intensity, and transmit data wirelessly using Zigbee. To improve the error of a Shigometer^TM equipped with a 2-point resistance measurement method, the TPAM system is equipped with a 4-point resistance measurement method. For the first time in the world, the resistance of a living tree (zelkova) was continuously monitored by applying a 4-point measurement method. As a result, it was confirmed that when the 4-point measurement method was applied instead of the previous 2-point measurement method, the repetition error was reduced by 8 times and the error due to contact resistance was reduced by 5 times. Through 24-hour observations, it was observed that the tree's resistance fluctuated by 12% depending on surrounding conditions. This means that one-time measurements cannot represent the degree of tree health and that continuous data collection is necessary. We began collecting long-term data for four seasons using the currently developed TPAM system.