Hail Detector and Forecaster ArtAr-HDF

This article describes a new and low-cost microwave passive sensor for hail prediction (forecasting) and detection developed in Armenia, which can be used to implement fully autonomous and automatically functioning hail protection of locally limited or large agricultural and urban areas in order to prevent, suppress or catch hail in traps. The article also presents the results of measurements of the intrinsic emission characteristics of water and ice, rain and hail clouds, carried out in laboratory and field conditions in the Ku-band of radio frequencies. The results obtained showed that the intrinsic emission of a hail cloud in the Ku-band of radio frequencies differs significantly from the intrinsic emission of a rain cloud. The presented results show that indeed the radar is not very suitable for the timely detection and determination of hail with a high probability, which is very important for the timely starting up of anti-hail protection means. On the contrary, radiometers (passive microwave sensors) can become an effective sensing tool for timely detection and recognition of hail with a high probability of long-range approaches up to ~12 - 15 km.


Introduction
Along with the existing difference between world population growth rates and food production, a continuous increase in environmental pollution and environmental degradation, diseases such as COVID-19 or more serious can cause hunger on the planet. Therefore, in order to partially overcome the inevitable famine on the planet it is necessary primarily to solve the problem of lossless preservation of the crop. The main obstacles to solving the problem of lossless preservation of the crop are natural disasters such as droughts, floods, storms (strong winds) and tornadoes, earthquakes and volcanic eruptions, early or late frosts, bad weather (continuous rains) during sowing and harvesting, hail and rainfall, etc. Many of these disasters are not subject to man at all and cannot be controlled by him. And only hail and rain in some cases can be controlled. The only problem is the effectiveness and cost of such management.
A detailed description of the flaws of existing methods and stations of anti-hail protection and the advantages of new, recently developed methods and networks of anti-hail protection and their application for anti-hail protection of restricted or vast agricultural and urban areas are presented in [1] [2] [3] [4]. Despite the effectiveness of these methods which aroused wide interest and discussions among interested users and specialists, their use is temporarily delayed due to the lack of a prototype of hail detector-forecaster. Thus, the development of a cheap hail detector-forecaster and its field testing is an urgent and important task for the agriculture, the economy of the state and nature.
In this article a new and cheap sensor for predicting (forecasting) and early detecting hail at a distance up to 12 -15 km is described, and some experimental results of its application are presented. The developed detector-forecaster can be successfully used for implementation of fully autonomous and automatically functioning anti-hail protection of locally limited or vast agricultural and urban areas in order to prevent, to suppress or to capture hail in trap areas.

Hail Detector-Forecaster ArtAr-HDF
ArtAr-HDF was developed on the basis of Ku-band (~10 GHz) satellite dish antenna with LNB (low noise block downconverter). The block diagram of Ar-tAr-HDF entirely corresponds to the block diagram of the hail detector-alerter described in [5]- [10] in detail. It comprises a parabolic antenna (a dish), a radiometric (microwave) receiver (including LNB), a compensation device, a multi-level thresholder, a warner and a power supply. ArtAr-HDF can operate at any polarization of observation, vertical or horizontal. It is the cheapest hail detector-forecaster, since does not comprise a self-controlled and automatically functioning scanner, a remote warning device and other control means. Home-made work prototypes of ArtAr-HDF with 60-cm and 100-cm dish antennas and LNB set on manually controlled scanners are shown in Figures 1-3 Figures  4-6, respectively. ArtAr-HDF has two outputs, more sensitive and less sensitive, the signals of which can be recorded by a recorder, simultaneously ( Figure 6). The sensitivity of the radiometric receiver of ArtAr-HDF at 1s integration time is better than 0.5 K. The spatial resolution of ArtAr-HDF depends on the angle of observation ϑ and the altitude h of the cloud. It can be approximately estimated by the following equation:

and in
where d is a diameter of the cross section (footprint) of the antenna's main beam at the altitude h, D is the dish's diameter, λ is the radio wavelength and      In some cases, a dish is not necessary and ArtAr-HDF can only include a radiometric receiver. In such cases, in Equation (1) and in Equation (2), it is necessary to replace D by D f , where D f is a diameter of LNB feed horn which is approximately equal 4.5 -4.7 cm. For such studies, an external and stable microwave emission source is required, which would make it possible to assess the sensitivity of the radiometers and their performance. Only the sun could be such a source in field conditions.

Testing and Trial of ArtAr-HDF
However, daily capture of the sun by a narrow beam of a satellite dish is a very difficult task due to the fact that it is constantly changing its position in the sky, and it is very difficult to track these changes by mechanical scanners. Therefore   LNB was directed to water or ice surface in trays under the angle of incidence ~12˚.
Initially, experiments were carried out with trays installed on wooden stand and filled with water or frozen ice (see Figure 8). Trays were always installed in the same place and probed (observed) by LNB from the same position under the same angles of incidence and azimuth. For calibration of data of measurements LNB sometimes was directed to the same place of the balcony dry ceiling (Gray Body) with an emissivity of ~0.95 (see Figure 2). At any changes in the course of the experiment, the time of the corresponding change and air temperature were recorded, and since, it was almost constant during the day, the air temperature can be used as the physical temperature of the environmental objects like ceiling, for instance.
In Figure   of the signal levels marked by 5 and 6 which is equal ~43.5 Kelvin can be used as a calibration level (signal) to calibrate all measured data. The results in Figure 9 show that the level of intrinsic emission of freezing water relatively to the level of unfrozen, cold water at 0˚C periodically changes. The difference between these emission levels, hereinafter referred to as thermal contrasts ΔT, periodically increases, even reaching a value of 254 K, and decreases, reaching even to a minimum value of 18 K. In Figure 9 also shown the values of these contrasts estimated at different points of the maximum and minimum levels of the curve 4. These ΔT contrasts were estimated based on a calibration level of 43.5 K and correspond to the points indicated in Figure 9 by the arrows. During further experimental studies, analysis of measured data and conditions of experiments it was found that periodical behavior of the change of thermal contrasts ΔT of freezing water was due to the fact that during the experiments the water in the tray did not completely freeze. Inside of the ice layer of the tray a cavity with cooled water was remained, enveloped in an ice shell of 1 -2 cm walls from all sides, and the volume of this cavity changed during the day. This was due to the fact that in 2020 the winter in the city of Yerevan was warm and there were no severe frosts that could completely freeze the water of ~4 cm in the tray. Since the weather forecast did not predict frost or cold snaps in the coming weeks for the city of Yerevan, in order to avoid problems with incomplete freezing of water and hardening of ice, it was decided to change something in the experimental procedure. Namely, the wooden stand was covered with a duralumin sheet 4 mm thick, and the plastic trays were replaced with duralumin trays 1.5 cm deep, into which water was poured about 8 -9 mm (see Figure 10). The tray-free right side of the duralumin sheet (see Figure 10) was used to measure the specular reflection of sky emission through the duralumin sheet. A thickness 8 -9 mm for a layer of water is more than enough for studying microwave emission of smooth water surface, since the depth of penetration of electromagnetic waves of Ku band of microwave into water is very small and is only a fraction of a millimeter. As for the layer of ice 8 -9 mm thick, this decision was made due to the fact that pellets of hail with a diameter of 8 -9 mm are already quite dangerous hail, and a further increase in their diameter will not significantly affect the emission characteristics of hail clouds. After all these preparations and changes, it was still managed to conduct an experiment during next couple of days with subzero air temperature and get an interesting result. Really, in Figure     were carried out strictly under a certain azimuth and did not change during the experiment or from experiment to experiment. However, in case if it was absolutely necessary antenna was quickly rotated from one azimuth to another that was preliminary fixed as observable direction in order to follow or catch interesting clouds. These azimuths were previously investigated in detail to ensure that they had the same background levels for clear and cloudy sky and that there is no radio interference from satellites or the environment from these directions.

The Results of Field Testing of ArtAr-HDF
Further, some interesting results of field measurements of atmospheric events and phenomenon will be shown and discussed.  was extinguished. The maximum contrast is ~152.6 K and the last contrast just before the extinguishing of fire is ~89.5 K. The signal level labeled 4 corresponds to the sun emission with the maximum value ~155.6 K. The curve labeled 5 corresponds to hail-rain clouds with the first maximum contrast ~96.9 K and with the last contrast level ~76.3 K. And at last, the signal level labeled 6 corresponds to an emission of dry, Gray Body (a balcony ceiling) with an emissivity of ~0.95 at +19˚C. This contrast which is ~277.4 K was used for estimating all other recorded contrasts have shown in Figure 16. In Figure 17 a split chart of the record of light cloudy sky emission carried out uninterruptedly during ~20 hours in the test site of ECOSEV ROC Co. Ltd. The signal level labeled 1 corresponds to the background of light cloudy sky, and as it was mentioned above sky emission is decreasing during the day from the morning till the sunset ~12.5 K. After the sunset sky emission is increasing and close to next morning (~5:00) it gets its maximum value (~18.7 K) and remains steady till the end of the record. The signal level labeled 2 corresponds to the sun emission with the maximum value ~217.4 K. During the night there were thunderstorms and rain which are labeled 3 and 4 on the split chart of the record of Figure 17, respectively. The curve labeled 3 with the maximum contrast ~123.6 K more likely corresponds to hail clouds. The curve labeled 4 with the maximum contrast ~78.6 K corresponds to rain and rain clouds. So, the results of field trials, selectively represented in Fig-ures 14-17, show that ArtAr-HDF can detect and recognize hail at the distance up to ~12 -15 km with high probability and timely warn means of anti-hail protection about coming danger. However, for more persuasiveness of this conclusion, a direct experiment was needed, that is, not only recognition and detection of hail, but also receiving of a visual or informational confirmation about it. Unfortunately, this is a very difficult experimental task, as hail is a rare, uncontrolled and unmanaged atmospheric phenomenon, and it can be expected for years in one or two preliminary selected test areas. Therefore, the results of measurement represented below have significant value for further researches and applying. So, 11 July, 2020, closer to the night, in the west of the experimental site of ECOSERV ROC Co. Ltd., at a distance of ~12 km and at an altitude of ~5 km, a powerful hail cloud was recorded which stirred up whole network of anti-hail protection, and, all gas generators of this area began to work without interruption in the late evening.
The photo of the screen of the monitor with a split chart of transient recorder of the event of 11 July, 2020 is presented in Figure 18 (see the original in Figure 6). The record was suddenly interrupted, since, the electricity in the area has been cut off and the record was not saved in computer and therefore was lost. It's good that it was managed to take a picture of the screen right before the power outage.
In accordance with the results of the split chart of recorder represented in Figure 18, where both sensitive and insensitive outputs of ArtAr-HDF are recorded, the sun's maximum contrast is ~269 K. It means that the sun passing through the main beam of the antenna very close to the central axis. Rain clouds intrinsic emission can be ~25 K or 58 K or some more. The registered maximum value of the contrast of hail intrinsic emission is ~165 K. So, the difference between hail and rain clouds intrinsic emissions is more than 100 K.

Discussion of the Results
Thus, the results obtained during laboratory researches and presented in Figure   9 and Figure 11 showed that the nature of the behavior of the changes of intrinsic emission of water transforming (converting) into ice is steady and is repeated from experiment to experiment. Frozen and icy water becomes a powerful absorber of Ku-band radio waves. It means that frozen water droplets will reduce radar cross section of hail cloud, that is will reduce the power of backscattered radar signals. In opposite, dry and solidified (hardened) ice becomes a radio-transparent material for the same radio wave range. And, therefore, the hardened (solidified) and dried hail pellets will simply pass radio signals through themselves, thereby reducing radar cross section and reducing backscattered radar signal power. As for other ranges of radio waves, then with a decrease in the length of the radio wave (with an increase in frequency), the nature of the changes of intrinsic emissions of water and ice will probably remain the same.
And with an increase in the length of the radio wave (with a decrease in frequency), the nature of changes of intrinsic emissions of water and ice possibly can change significantly. However, these issues require a separate and thorough study, which goes beyond our interests. The results obtained during laboratory experiments show once more that, indeed, radars are not suitable for the timely detection and identification (classification) of hail with a high probability, as mentioned in [3]. In contrary, radiometers can become an effective means of sensing for timely detection and recognition of hail with a high probability on long-distance approaches up to 12 -15 km. This statement is also mentioned in [3].
The results of field trials selectively represented in Figures 14-17 show that ArtAr-HDF can detect and recognize hail at the distance of up to ~12 -15 km with high probability and timely warn means of anti-hail protection about coming danger. It is due the fact that intrinsic emission of hail cloud in Ku band of radio frequencies differs significantly from the intrinsic radiation of a rain cloud.
The difference between hail and rain clouds intrinsic emissions can be more than 100 K. Hence, ArtAr-HDF can be successfully used for creation of global network of anti-hail protection. In Figure 19 a third prototype of ArtAr-HDF is presented which may be used without a dish if to change something in feed horn of LNB. A power consumption of ArtAr-HDF will not be more than 2 -3 W. A power supply of ArtAr-HDF is DC 12 V or 24 V and AC ~110 -220 V.

Conclusion
Thus, the results of laboratory measurements of water and ice emission characteristics have shown that radar is not suitable for the timely detection and identification of hail with a high probability which is very important for timely starting up of means of anti-hail protection. In opposite, radiometers can become an effective means of sensing for timely detection and recognition of hail with a high probability on long-distance approaches up to ~12 km. The intrinsic emission of hail cloud in Ku band of radio frequencies differs significantly from the intrinsic radiation of rain cloud. It is developed a cheap hail detector-forecaster which can be successfully used for timely detection and prediction of hail at a distance up to 12 -15 km and for creation of a global network of anti-hail protection.