Flame Heat Sources as a Sustainable Method of Protecting Fruit Orchards against Frost in Poland

Abstract

One of the key factors affecting the yield of fruit orchards is their protection against spring frosts. This paper presents the methods currently used (like anti-frost nets, chemical agents, etc.) and proposes an alternative method of raising the temperature around fruit trees using flame heat sources with a suitable heating power. Flame heat sources in the form of metal containers filled with paraffin were utilised in this experiment. The effect of the flame heat sources on the relative increase in the temperature of a wooden surface (simulating trees located in Rzeszow, Poland) depending on the distance from the heat source was measured and estimated based on mathematical models. Based on the results of experimental work, an optimal placement of flame heat sources (1.58 m from each tree) was proposed for orchards of semi-dwarf fruit trees. The proposed modifications of the flame heat sources to incorporate infrared radiation converters resulted in a greater increase in the temperature of the wooden surface. The analysis of the available economic and environmental data showed that the proposed method may be a competitive alternative to the standard method of protection. If the flame heat sources are filled with combustible materials derived from renewable sources, such as those produced from fats, the overall CO₂ emission balance for this solution will be significantly more favourable. The use of flame heat sources does not necessitate the introduction of permanent changes in the environment, as is required in cases when irrigation infrastructure is installed.

1. Introduction

Horticultural production, mainly fruit growing, is one of the main branches of agricultural activity in Poland. In 2020, according to the statistical data available, the area of fruit tree cultivation was approximately 350,000 hectares, and the revenue from the sale of the fruit produced (4.7 million tons) constituted a significant part of the country’s GDP [1]. The largest share is formed of the production of apples, which exceeded 4.2 million tons in 2022. Pear harvests that year were estimated at 80 thousand tons, while 132 thousand tons of plums were harvested. The production of fruit in cherry orchards was estimated at over 183 thousand tons, and cherry harvests increased by 30% to the level of 77 thousand tons. The total production of peaches, apricots, and walnuts was estimated at almost 22 thousand tons. The export value of fruits (1.06 million tons) was estimated at 402 million euros that year [2]. The factors determining the size and quality of the fruit yield are the use of appropriate agrotechnical treatments, mainly fertilisers, and the provision of protection against diseases and pests [3,4]. The weather anomalies observed in recent years in Central and Eastern Europe, which are a consequence of negative, progressive climate change, cause additional problems with maintaining the profitability of production. Particularly dangerous from the point of view of fruit production is the occurrence of hail in the summer and spring frosts when flower buds are developing [5]. Low temperatures during this period can cause a significant decrease or even a lack of yield from fruit trees [6,7].

A major problem with horticulture at latitudes where Poland and other countries with similar climates are located is the occurrence of large temperature drops in the spring. During this period, the start of vegetation growth increases the water content in individual tissues, especially in cell sap. The recorded temperature phenomena during this period in recent years reached as low as −8 °C (March 2023) [8]. When the temperature suddenly drops below critical values (this temperature depends on the plant species and varietal characteristics), water crystals form, increasing its volume and causing the destruction of plant cell structures and tissues. This phenomenon is particularly detrimental during the formation of flower buds in fruit trees, which directly translates into a decrease in yield and economic return [9]. The most sensitive stage occurs from the pink (or white) bud phase to the end of flowering. For apple trees, temperatures ranging from −2 °C to −4 °C during this period can lead to the loss of 10% to 90% of blossoms, consequently reducing the final fruit yield. The most effective method of protecting orchards from spring frosts is overhead sprinkling, but this entails a high cost in installing equipment and the use of significant amounts of water [9]. Other methods of protecting fruit trees from frost, albeit with significantly less economic significance, include anti-frost misting, protective sprays, covering with agricultural fleece, wind machines, and hanging anti-frost nets. However, these methods, especially the use of agricultural fleece and anti-frost nets, are expensive and challenging to apply, particularly in orchard locations prone to intense winds [10,11,12]. On the other hand, orchard protection through the application of chemical agents that lower the freezing point of water, including micronutrient preparations containing various types of salts, forms another approach. These preparations ensure proper plant function and increase the density of juices by increasing the concentration of nutrients in the plant. It should be noted that sprays of this kind should be applied before the onset of frosts, preferably at temperatures around 10 °C, to allow penetration into the plant’s cellular structures. However, this is not always feasible, particularly due to the unpredictability and variability of meteorological conditions during the winter period [10]. Additionally, it is important to acknowledge that this method burdens the environment with chemical substances that do not naturally occur in it. However, the preventive solutions applied to counteract the negative effects of significant temperature drops during this period in the spring mainly involve the use of over-crown watering of trees with water [13]. This method utilises the high latent heat of the water phase transition, which prevents the formation of water crystals in the cell space of plant material leading to the destruction of developing flower buds in this period [14]. The main barrier to the widespread use of this solution is the necessity of installing expensive technical solutions and access to and the cost of appropriately treated water [9]. The use of this method for protecting trees from spring frosts is additionally limited by legal regulations giving priority access to water for utility purposes. Moreover, the observed global water shortages in recent years exacerbate this problem and generate the necessity to search for alternative methods of protecting orchards from frosts. In the work of Drepper et al. [15] and Rieger [16], alternative methods are described for protecting blossoming orchards from late spring frosts, with a particular focus on strategies applicable to all species of fruit trees and vines cultivated in temperate climatic zones. The data also take into account changes in the climate observed in recent years, recorded up to 2021 [8]. It was indicated that systems utilising different heat sources, especially in closed production systems, demonstrate the greatest effectiveness. On the other hand, other authors suggest that the most effective method for protecting orchards when the temperature falls below approximately −2.5 °C is irrigation. It should be noted that such conditions are present in the research location [6,17].

An alternative way to increase the temperature around fruit trees is to use flame heat sources fuelled by a flammable material of appropriate heating power like paraffin or more sustainable derivatives of fats or fatty alcohols. The aim of this paper is to present the energy and environmental aspects of using flame heat sources (about the size of a large candle), which can be an alternative to conventional methods of protecting fruit trees from frost.

2. Materials and Methods

2.1. Research Material

Flame heat sources were constructed from metal containers with an appropriate diameter and height (

Table 1. Basic parameters of the flame heat sources used.

	Flame Heat Source Types
	A	B	C	D
Mass of paraffin [g]	3200	1800	2100	3600
Diameter [cm]	29.5	15.5	15.5	19
Height [cm]	21.5	15.5	18	24
Burning time [h]	72	40	47	81
CO2 emissions [g]	9900	5566	6497	11,137

), which were filled with paraffin wax, with a candle wick placed in the centre. The paraffin wax used in the experiment had a freezing point of 56–58 °C, an oil content of 1%, and an ignition temperature of approximately 80 °C. It was white in colour and odourless. However, for practical applications, it is more reasonable to use combustible materials derived from renewable sources, such as derivatives of fats or fatty alcohols.

In the second variant of the experiment, an emitter of infrared radiation was used, which took the form of an absorber/emitter of energy from the fumes, made of aluminium foil placed around the flame heat source (Figure 1). This element took the form of a rectangular prism (30 × 30 × 1 m) without a base. The flame heat sources proposed in the method for protecting orchards from late spring frosts varied in terms of geometric dimensions, directly affecting their operating time, as well as the manner and amount of heat emitted, especially the infrared radiation emitted by the material enclosing the heat source.

Properties of the aluminium elements:

o

Density: 2.7 g cm⁻³;

o

Aluminium foil specific weight: 6.35 µm foil weighs 17.2 g m⁻²;

o

Melting point: 660 °C;

o

Electrical conductivity: 64.94% IACS (IACS: International Annealed Copper Standard);

o

Electrical resistivity: 26.5 nΩm;

o

Thermal conductivity: 235 W m K⁻¹;

o

Thickness: Foil is defined as measuring less than 0.2 mm (<200 µm).

2.2. Measurement of the Range of Thermal Radiation from a Heat Source

Wooden elements simulating fruit trees were made as rectangular blocks with a height of approximately 1.5 m and a base surface area of 10 cm². The experiment was carried out in Rzeszow, Poland (GPS 50.002332, 22.031396), on 28–30 November 2023 in the morning. The blocks were placed at distances of 0.5 m, 1 m, 1.5 m, 2 m, 2.5 m, 3 m, and 3.5 m from the paraffin-fuelled flame heat source, forming an arc. A control test was carried out which consisted of a standard element simulating a fruit tree located at least 25 m away from any of the heat sources utilised, which ensured that any influence of flame heat sources on its temperature could be excluded. Measurements of the temperature of the control sample were consistent with the ambient temperature. After this, the flame heat sources (type A, B, C, and D) were lit and the surrounding temperature of each element simulating a tree was measured after 30 min using a thermal camera (Figure 2). Temperature measurements were taken at a distance of 10 cm from the wooden elements at heights of 40 cm and 90 cm. Temperature measurements were taken in four repetitions at 1 h intervals. The flame heat source experiment was carried out at an air temperature above 0 in atmospheric conditions where intense sunlight was not present. Such conditions were modelled to simulate field conditions, taking into account naturally occurring fluctuations in meteorological parameters. The temperature values obtained were averaged, and the differences between the temperature values of the test samples at distances of 0.5 m, 1 m, 1.5 m, 2 m, 2.5 m, 3 m, and 3.5 m from the heat source and the temperature value of the control sample were determined for two measurement heights.

2.3. Measurement of the Range of Influence of a Flame Heat Source Enclosed in an Infrared Radiator

An infrared radiator with a height of 1 m and a diameter of 40 cm, capturing the heated convective mixture of combustion gases and air, was installed in a flame heat source powered by paraffin of type (A). The radiator receives heat from the gas mixture and emits infrared radiation from the elements made of aluminium foil. The remaining experimental conditions were identical to those described in Section 2.1. Temperature measurements were taken four times at 1 h intervals.

2.4. Mathematical Model for Predicting the Range of Influence of a Flame Heat Source

Statistica 13.1 software (TIBCO Software Inc., Hillview Avenue, Palo Alto, CA, USA) was used to predict the range of influence of paraffin-powered flame heat sources using analysis of variance (ANOVA) and regression modelling. The measurement results were approximated using a polynomial of degree n, which allowed a mathematical description of the relationship between the observed temperature increase on the element being heated and the distance from the heat source to be prepared.

3. Results and Discussion

3.1. Range of Thermal Interaction of Flame Heat Sources

The proposed experimental arrangements allowed the range of thermal interaction of flame heat sources distributed on elements with properties similar to those of fruit trees (heat absorption capacity, heat conductivity coefficient, etc.) to be examined. A wooden model with regular geometry was used to simulate a tree, which allowed reliable measurements to be obtained. Based on the experiment conducted, an increase was observed in the surface temperature of elements simulating fruit trees at distances of 0.5–3.5 m from the heat source for each type of flame heat source investigated (Figure 3). The analysis of the experimental results showed that the increase in temperature of the simulating elements showed a significant dependence on the distance from the heat source (p < 0.05 (

Table 2. Two regression model (quadratic or linear and logarithmic) equations and Pearson’s linear correlation coefficients.

Type of Flame Heat Source	Height of Measurement Point [cm]	Pearson’s Linear Correlation Coefficient rx:y, p *	The Model Equation: y—Temperature Increase [°C] x—Distance from the Heat Source [m]
		quadratic model
A	40	−0.98. p < 0.05	y = 2.08 − 0.346x
	90	−0.96. p < 0.05	y = 1.384 − 0.156x
B	40	−0.96. p < 0.05	y = 1.66 − 0.730x + 0.079x2
	90	−0.94. p < 0.05	y = 1.729 − 0.627x + 0.089x2
C	40	−0.91. p < 0.05	y = 4.852 − 2.892x + 0.482x2
	90	−0.89. p < 0.05	y = 3.128 − 1.946x + 0.345x2
D	40	−0.86. p < 0.05	y = 4.707 − 2.922x + 0.521x2
	90	−0.94. p < 0.05	y = 1.94 − 0.7063x + 0.092x2
		logarithmic model
A	40	−0.98. p < 0.05	y = −0.55ln(x) + 1.6772
	90	−0.85. p < 0.05	y = −0.22ln(x) + 1.1848
B	40	−0.95. p < 0.05	y = −0.643ln(x) + 0.9362
	90	−0.96. p < 0.05	y = −0.436ln(x) + 1.15
C	40	−0.97. p < 0.05	y = −1.644ln(x) + 2.3434
	90	−0.97. p < 0.05	y = −1.11ln(x) + 1.4443
D	40	−0.96. p < 0.05	y = −1.476ln(x) + 2.2437
	90	−0.98. p < 0.05	y = −0.542ln(x) + 1.2719

*—Significant correlation at p < 0.05 (α = 0.05).

)). A gradual decrease in the rate of temperature increase was noted with the increasing distance of the surface simulating a fruit tree from the heat source for the measuring heights examined. The greatest temperature increase was noted for flame heat sources of type C and D. At a height of 40 cm, an increase in temperature was recorded ranging from 3.6 °C (distance 0.5 m) to 0.46 °C (distance 3.0 m) for flame heat source type C. A similar relationship was observed for flame heat source type D, for which the temperature increase ranged from 3.5 °C (distance 0.5 m) to 0.66 °C (distance 3.5 m). Temperature measurements of the elements simulating trees at a height of 90 cm were also taken, but the ranges of temperature change were lower and ran from 2.4 to 0.2 °C (flame heat sources of type C) and 1.57 to 0.64 °C (flame heat sources of type D). These changes were caused by the relative increase in distance from the heat source compared to measurement points that were lower in height. It should be noted that the measurements were taken on the planes of the elements simulating a fruit tree which were oriented toward the heat source. The proposed experimental model allowed the range of interaction of the flame heat sources applied in one plane to be determined. These parameters should be considered as the radius of the minimal interaction hemisphere. In the vertical directions, where fruit tree canopies are located, the interaction may be greater than the minimum due to the occurrence of convective flow of the exhaust gases. The proposed methodological approach guarantees that an increase in temperature is observed at specified distances from the flame heat sources. Measurements taken on other planes were similar to the temperature recorded for the control sample. This indicates a mechanism of heat transport through radiation [18]. The interaction is directional and only heats the surface oriented toward the radiation source, preferably perpendicular to it [19]. It should be noted that the driving force of the heat exchange process is the temperature difference, which is independent of the absolute temperatures at which it is recorded. This difference was calculated between the control sample (ambient temperature) and the study element. Additional factors affecting the heat exchange process may include atmospheric conditions such as air humidity, wind strength, etc. However, these factors were kept constant during the experiments [20].

The application of a mathematical description in the form of quadratic functions (Table 2) gave the best fit which is indicated by the coefficient r. It should also be noted that the mathematical description provides a solution with physical significance only within the proposed range of arguments (the suggested measurement range provided in the graphs). The application of heat exchange theory [20] would require one to propose a complex system of equations that take into account all the heat exchange mechanisms observed in the proposed measurement setup. In our opinion, this would significantly complicate the calculations, and the proposed simplification and mathematical description are sufficient to provide a mathematical description of the effects obtained.

3.2. Range of Influence of the Flame Heat Source with an Infrared Radiator Feature

In addition to the energy radiated as electromagnetic radiation, the flame heat source also produces a certain amount of gases at an increased temperature which, due to convection, move in the vertical direction. This direction is highly unfavourable because this energy is not utilised to protect fruit trees from low temperatures [21]. The solution to this problem is to use elements that convert the thermal energy of the mixture of heated air and exhaust gases into infrared radiation [19]. To this end, an infrared radiator was installed which acts as an energy absorber made of aluminium foil placed around the flame heat source (Figure 1). This element was in the form of a rectangular prism without a base. The measured temperature of this element was approximately 35 °C and varied depending on atmospheric conditions. This modification affected the increases in surface temperature that were measured on the elements simulating fruit trees (Figure 4).

In a similar manner to the case of the influence of flame heat sources on elements simulating fruit trees in the orchard, the modified heating systems significantly affected the increase in the temperature of the surfaces of the simulating elements. However, it should be noted that the installation of infrared radiators that convert the heat of exhaust gases resulted in a greater increase in surface temperature of the elements simulating fruit trees, reaching up to 2.9 °C (measured at a height of 40 cm) and 2.05 °C (measured at a height of 90 cm) (Figure 2). The statistical analysis showed a significant influence on temperature increase of both the distance of the elements simulating fruit trees from the heat source and the height of measurement (p < 0.05) (

Table 3. Two regression models based on equations (quadratic or linear and logarithmic) and Pearson’s linear correlation coefficients.

Height of Measurement Point [cm]	Pearson’s Linear Correlation Coefficient: rx:y	The Model Equation: y—Temperature Increase [°C] x—Distance from the Heat Source [m]
	quadratic model
40	−0.98. p < 0.05	y = 2.867 − 1.824x + 0.293x2
90	−0.92. p <0.05	y = 2.87 − 1.82x + 0.293x2
	logarithmic model
40	−0.99. p < 0.05	y = −1.474ln(x) + 1.9844
90	−0.98. p <0.05	y = −1.096ln(x) + 1.2581

). Flue gases are carriers of a quantity of thermal energy. In industrial solutions, efforts are made to harness this energy to enhance the energy efficiency of the process. By contrast, heat recovery technologies that transfer heat to other forms of energy are labelled as active technologies. Typically, waste heat is recovered to produce cooling and domestic heating simultaneously. However, some heat recovery systems are designed to utilise waste heat to generate another type of energy [19]. The proposed solution with a modified flame heat source clearly indicates that when it comes to protecting orchards from freezing using flame heat sources, such an approach is also justified.

3.3. Simulation of the Distribution of Flame Heat Sources in an Orchard and Their Potential Impact on the Natural Environment

The simulation of the favourable distribution of flame heat sources in the orchard was carried out taking into account the spacing of semi-dwarf trees in rows, for which the average planting density is about 3000 trees per hectare (Figure 5). The standard spacing for such trees is 1 m × 3 m. It was assumed that the minimum heating effect should be 1 °C. The mathematical models adopted indicate that an increase in the temperature of the fruit tree surface by 1 °C can be achieved by placing flame heat sources at a distance of approximately 1.5 m. This is only an assumption, but often a 1 °C difference in temperature measured on the tree is enough to protect the orchard against frost. An analysis of the geometry of the arrangement of fruit trees in the orchard showed that placing flame heat sources in the centre of rectangles defined by four adjacent fruit trees (Figure 5) results in obtaining a distance of 1.58 m between the flame heat sources and each tree. Such a distribution is advantageous because it allows the surface of fruit trees to be heated from all sides, assuming that the system is supplemented with heating elements placed at a distance of 0.5 m from the trees in the outer rows. Such a distribution requires the use of approximately 1600 flame heat sources per hectare.

3.4. Environmental, Energy, and Economic Aspects of Using Flame Heat Sources in Orchards of Dwarf Fruit Trees

According to the Central Statistical Office of Poland, the area of fruit orchards in 2020 exceeded 355,700 hectares [1]. In order to protect orchards from spring frosts, which can cause very serious losses in fruit yield, it is necessary to install a costly overhead sprinkling system. It should be noted that the sprinkler installation can also be used to irrigate the orchard in further stages of vegetation. This method requires access to significant amounts of water, as its unit water consumption is approximately 45 m³ ha⁻¹. Assuming that the number of days in the spring period with temperatures that threaten fruit trees in the orchard (bud development stage) is 15 (12 h of protection per day), taking into account the area of fruit cultivation in Poland and the unit water demand for the overhead method of tree protection, water consumption could reach 2.9 × 10⁹ m³. Given the recent water shortages in Poland and globally, especially in the spring period, providing frost protection for the entire area of fruit cultivation is almost impossible, as the water demand for protection is only one order of magnitude lower than the water demand for household purposes in Poland during this period (12 h for 15 days), which amounts to 4.3 × 10¹⁰ m³ (38 million people, daily water consumption of 150 litres per person) [1]. Additionally, just like in the case of using flame heat sources, the use of water for protecting orchards from spring frosts generates CO₂ emissions associated with the use of chemical substances and the mechanical energy required for water treatment. This process is costly and generates emissions estimated at 790 ± 228 kg CO₂e d⁻¹ (on-site) for treating 2 × 10⁶ m³ while removing 10 NTUs (Nephelometric Turbidity Units) of pollutants [22,23]. Additionally, emissions related to the creation of the entire infrastructure, including the water supply installation with pumps, pipelines, and sprinkler networks, should be taken into account. The demand for these elements depends on various factors and is difficult to estimate. Considering that flame heat sources will be filled with combustible materials derived from renewable sources, such as those produced from fats, the overall CO₂ emission balance for this solution will be significantly more favourable. This is possible because fuels produced from renewable sources exhibit similar parameters (calorific value and physical characteristics). To achieve comparable results to flame heat sources filled with paraffin wax, it is necessary to exclude fuels produced from renewable sources with significantly different parameters. An example of this could be ethanol, which burns at a temperature of 20 °C, while fatty acid esters burn at a temperature 250 °C creating a much larger temperature gradient and enhancing the effectiveness of the solutions applied [24]. The use of flame heat sources does not necessitate the introduction of permanent changes to the environment, as is required in the case of installing irrigation infrastructure.

The cost of installing an overhead irrigation system for fruit trees in an orchard, calculated per hectare, was estimated based on installation costs and material expenses related to the purchase of sprinklers, a water tank, and a pump, at an average of ~EUR 30,000. Such an installation can irrigate 1 hectare of fruit orchard with a mass flow rate of 45 m³ h⁻¹. Considering the lifespan of intensive orchards with dwarf and semi-dwarf fruit trees (dominant in modern orchards) at ~15 years, the installation cost can be estimated at EUR 2000 per year. Meanwhile, the cost of water used for irrigating fruit trees with a mass flow rate of 45 m³ h⁻¹ and an average cost of EUR 2.5/m⁻³ amounts to EUR 112.5 h⁻¹. Taking into account that negative temperatures during the spring occur for ±15 h on average, the cost of water consumption is EUR 1687 per day. Assuming that the number of days with threatening temperatures for fruit trees in the orchard during the spring (bud development phase) is 15, the total cost (considering both the installation cost and water cost) is EUR 27,305 per year. In the case of the proposed method using flame heat sources, this cost is lower, amounting to ~EUR 20,000. The cost-effectiveness of this method takes into account the use of 1600 flame heat sources (for the heat source with the highest fuel content) per hectare, with a usage time assumed to be 15 h per day over 3 days. Assuming a similar condition to overhead irrigation, that negative temperatures in spring occur for a period of 15 days, the total number of flame heat sources required during this period is 8000 units. Assuming a unit cost of approximately EUR 2.5, the total cost of seasonal tree protection using flame heat sources is EUR 20,000. In addition to the lower costs of implementing the proposed method, the aspect of its application in the light of global shortages of drinking water and the need for its rational management should also be considered.

4. Conclusions

Meteorological conditions and statistical data analysis have shown that in the European latitudes where Poland is located, a significant problem in fruit production is the occurrence of large temperature drops in spring. This is one of the important factors determining the size of fruit yield, and therefore the profitability of production. So far, the most popular method of protecting fruit trees from spring frosts is to sprinkle them with water. The use of this method is determined by the availability of water sources and is associated with the need to create a permanent and expensive installation. Additionally, it has been estimated that the demand for water means that it would not be possible to apply this method to the entire area of fruit cultivation in Poland.

The research conducted has shown that an alternative method of protecting fruit trees in orchards can be the use of flame heat sources. Experimental work has enabled mathematical models to be developed that calculate the change in surface temperature heated as a function of distance from the heat source and the height of the measuring point. Based on the assumption that the temperature of the heated surface of the tree will increase by 1 °C, a scheme was proposed for the placement of flame heat sources in semi-dwarf fruit trees. It has been shown that the required number of heat sources is ~1600 pieces per hectare. Proposed modifications to the flame heat sources in the form of the addition of infrared radiation converters, which intensify the effects of infrared radiation, result in an increase in the temperature of the fruit tree surface measured at 40 cm from 2 to 2.9 °C at a distance of 0.5 m compared to unmodified flame heat sources. This modification can significantly reduce the number of heat sources required to be installed per unit area of orchards. In practice, it is advisable to utilise renewable sources of combustible materials obtained from fats, such as fatty acid esters or fatty alcohols, as this significantly reduces the carbon footprint of the proposed method. The analysis of the available data presented here clearly indicates that the proposed method may be a competitive alternative to the standard method of overhead tree spraying.