Numerical Analysis of Bifacial Photovoltaic Systems Under Different Snow Climatic Conditions

Abstract

The reflective property (albedo) of the ground plays an important role in the performance of bifacial photovoltaic modules. Snow, as a natural light-colored surface, reflects most of the light that falls on it. However, snow does not have a fixed albedo value. Therefore, it is essential to investigate the high albedo provided by snow in bifacial panels, which are becoming increasingly common. The albedo value of snow is influenced by numerous factors, including the precipitation characteristics of the snow, its depth, and the time since the previous snowfall. This study aims to investigate the impact of snow cover and the number of days with snow cover on the energy production of bifacial panels. An innovative dynamic albedo model integrating the snow type, depth, and duration was developed to advance bifacial PV system performance analysis under various snow and climate scenarios. PVsyst simulations were conducted to analyze the annual energy yield of bifacial photovoltaic panels in Erzurum Province under various snow conditions and accumulation levels. Furthermore, the variation in the number of days with snow cover according to different climatic regions and its effect on the energy production were evaluated for seven different provinces located in seven different regions of Turkey.

1. Introduction

Bifacial photovoltaic panels are able to generate electricity on both sides (front and rear). As production costs decrease, the use of these panels becomes more widespread because they are more efficient than monofacial panels [1,2]. However, the performance of these panels is influenced by several factors, the most important of which is the reflectivity of the surface. The reflectivity of the ground (albedo) directly impacts the radiation that reaches the rear surface of the panel. As a result, the albedo directly affects the energy production of the back-surface. While light-colored surfaces with high albedo, such as aluminum and snow, have been shown to positively impact performance, dark surfaces with high light absorption, including asphalt and heavily soiled galvanized surfaces, have been observed to negatively affect performance. For this reason, favoring bright, light-colored, smooth surfaces whenever possible will increase efficiency by increasing energy production. Research on the albedo effect of snow, one of the natural reflective surfaces with these properties, is crucial for maximizing energy generation potential.

A limited number of studies have investigated the effect of snow and snow characteristics on the performance of bifacial photovoltaic panels. Hayibo et al. investigated energy yield and snow losses for monofacial and bifacial solar photovoltaic systems under U.S. conditions. The findings revealed that bifacial systems exhibited superior performance in winter conditions, generating a bifacial gain of approximately 20%. The study reported an exponential relationship between albedo and the overall contribution percentage of the bifacial system [3]. In another study, Frimannslund et al. investigated the effect of design parameters such as tilt, azimuth, ground clearance, etc., on the energy yield of a solar power plant due to accumulated snow drifts. They simulated the performance of mono and bifacial panels using the Pvsyst program under Norwegian conditions with an Arctic climate. The study’s findings suggest that increasing the distance between the gap and the ground improves the performance of bifacial panels [4]. Another study by Frimannslund et al. focused on solar power plants in the polar region. In this study, monofacial and bifacial panels were utilized, and it was concluded that low temperatures enhance module performance. The study found that bifacial panels exhibited an average annual bifacial gain of 14.7%, highlighting the potential for enhanced performance in these conditions [5]. Russel et al. investigated the snow depth in the location of Canada. They reported an average daily energy loss of 0.034% for every 1 cm of snow due to snow accumulation [6]. Riise et al. examined the performance of a bifacial solar plant in Norway and determined that snow-related losses were 0.6% per year [7]. Molin et al., in their study in Sweden, reported that bifacial modules provide an annual efficiency increase of 48% compared to monofacial modules [8]. Singh and Jones underscored the beneficial impact of snow accumulation in bifacial panels, asserting that the presence of snow within the perimeter and beneath the panel can enhance the ground albedo by up to 75% [9]. A majority of the studies have generally been conducted near the polar regions, where snow cover is present for long periods of time. In addition to the positive effect of snow on energy production, factors such as snow pollution and snowdrifts have been demonstrated to modify energy production. This study is particularly significant, as it is among the first to evaluate bifacial panel performance in different climates across Turkey. It uses PVsyst simulations to analyze the effects of different snow types and thicknesses on photovoltaic generation. This is different from other evaluations that only consider limited conditions, as in [3]. Unlike many previous studies, which were conducted at a single site, this study quantitatively reveals generation differences by considering the number of snow days and albedo parameters in seven different geographical regions.

Numerous studies have been conducted on the performance of bifacial photovoltaic systems. However, the majority of these studies only consider the effect of snow cover on power generation in general terms using fixed albedo values and do not include the joint modeling of variables such as the snow type, snow cover duration, snow thickness, etc. For instance, d’Alessandro et al. [10] developed an analytical model of albedo reflectance; however, they did not take into account dynamic factors such as snow type and duration. In a similar manner, Ghafiri et al. [11] evaluated the performance of bifacial PV systems in Sherbrooke, Canada, but did not model the physical properties of snow in detail. Riedel-Lyngskær et al. [12] performed spectral albedo measurements on various surface types (i.e., grass, dry grass, gravel, and snow) but did not model the effects of snow cover duration, type, and depth on performance. Asgharzadeh et al. [13] employed a fixed snow albedo value (~81%) but did not incorporate the dynamic characteristics of snow (e.g., type, depth, duration of cover). Alam et al. [14] examined the performance of bifacial photovoltaic (PV) systems on various soil types; however, the snow effect and temporal variability were not incorporated into the analysis. Riise et al. [7] evaluated the snow effect over a short period in Norway and reported it as insignificant, with a low loss rate of 0.6%. The dynamic snow parameters were not included in the model. Hayibo et al. [3] compared the performance of monofacial and bifacial systems in snowy environments and analyzed the snow loss using albedo measurements and image processing methods. However, the physical parameters, such as the snow type, thickness, and duration of cover, were not systematically modeled, and the study was limited to a single site in Michigan.

In this study, meteorological and regional variables were evaluated in a PVsyst simulation environment. These variables included the snow type (fresh, old, wet, or melting), snow thickness (in centimeters), and the number of days with snow cover. The albedo values obtained were dynamically calculated on a monthly and regional basis. Consequently, in contrast to the frequently cited fixed albedo approach, a time- and region-dependent dynamic albedo model that is sensitive to climatic conditions was developed in this study. Through this model, complex production patterns formed by the combination of snow type, snow duration, and geographical effect were analyzed.

Another contribution of this study is the systematic demonstration of the effects of different snow climates on bifacial PV performance through comparative analyses of provinces representing seven different climatic regions of Turkey, rather than focusing on a single location. In this respect, this study not only enables the development of regional PV system design strategies in the Turkish context, but also provides a generalizable methodological framework for cities around the world with similar climate characteristics, such as Quebec City (Canada), Duluth (USA), Lviv (Ukraine), Sibiu (Romania), Sion (Switzerland), and Lillehammer (Norway).

In summary, this study makes significant and original contributions to the extant literature in terms of (i) modeling the snow-induced albedo effect in regional and seasonal contexts, (ii) adopting a dynamic albedo approach instead of a constant one, (iii) conducting multi-regional and multi-parameter scenario analyses, and (iv) presenting a generalizable approach for countries with climatic diversity, such as Turkey.

This study is significant in the context of a comprehensive evaluation of snow properties, as it integrates the analysis of snow precipitation type and snow thickness. Moreover, the absence of research evaluating the impact of snow and its characteristics on the electricity generation of bifacial panels in Turkey’s meteorological context underscores the significance of this study. Consequently, this study’s findings are crucial for expanding the utilization of bifacial panels and assessing their applicability, particularly in eastern regions of the country where the impact of snow climate is evident.

This article is mainly composed of three parts. In the initial part, the effect of the snowfall type on the energy production of the bifacial panel is investigated using the PVsyst program under the conditions of Erzurum province. The subsequent part investigates the effect of the snowfall amount, which is also based on the location of Erzurum province. The third part deals with the effect of the number of snow days on albedo for different climate zones in Turkey. Finally, a comprehensive discussion of the results and conclusions is provided.

2. Materials and Methods

Albedo is a parameter that defines the reflectivity of a surface, and it ranges from 0 to 1. The closer it is to 1, the more the radiation falling on the surface is reflected. Snow, a natural surface, possesses a high albedo value due to its ability to reflect nearly all of the light it receives. The albedo value of fresh snow can reach values between 0.8 and 0.95, but factors such as the snow type, depth, aging, and pollution can affect this value [15,16,17]. There are numerous models, including JULES, SNICAR, HTESSEL, and Noah-MP, which were developed to consider various properties in order to determine the relationship between the snow and albedo value [15,18,19,20]. These models are based on the optical properties of snow (e.g., the reflection and absorption of radiation at different wavelengths, sizes, shapes, and structures of snow crystals, etc.), physical properties (e.g., the density, water content, snowpack thickness, aging and albedo reduction, etc.), and surface roughness. Among the numerous models proposed for snow albedo, the following models can be used to determine how albedo changes with time, temperature, and surface variables.

According to the JULES model, which delineates the dynamics of snow albedo in relation to the depth, temperature, and the progression of snow aging, the albedo is calculated as follows [18,19]:

$${\alpha }_{sw}={\alpha }_{cds}+k_1·\left({\alpha }_0-{\alpha }_{cds}\right)·\left(T_s-T_c\right)$$

(1)

where ${\alpha }_{sw}$ is the snow albedo; ${\alpha }_0$ is the snow-free albedo;${\alpha }_{cds}$ is the cold deep snow albedo;$k_1$ is the snow-aging parameter; $T_s$ is the surface temperature; and $T_c$ is the snow albedo threshold temperature [18,20].

Another model is the HTESSEL model, which shows that albedo decreases exponentially with time, and according to this model, albedo is calculated as follows [18,20]:

$${{\alpha }_{sw}}^{t+1}=\left\{\begin{array}{c}{{\alpha }_{sw}}^t-{\tau }_a∆t/{\tau }_1, M_{sw}=0\\ \left({{\alpha }_{sw}}^t-{\alpha }_{min}\right)·exp\left({\tau }_f∆t/{\tau }_1\right)+{\alpha }_{min}, M_{sw}>0\end{array}\right\}$$

(2)

where ${{\alpha }_{sw}}^{t+1}$ is the snow albedo at the next time step; ${{\alpha }_{sw}}^t$ is the snow albedo at the current time step;${\tau }_a$ is the linear coefficient for the decrease in the albedo of non-melting snow (${\tau }_a$ = 0.008); $∆t$ is the model time step; ${\tau }_1$ is the length of day (${\tau }_1$ = 86,400); $M_{sw}$ is the melting snow amount; ${\alpha }_{min}$ is the minimum albedo of exposed snow (${\alpha }_{min}$ = 0.5); and ${\tau }_f$ is the coefficient for exponential decrease in the snow density and melting snow albedo [18,20].

Turkey is a country with high energy potential [21,22]. The solar energy potential map of Turkey, as illustrated in Figure 1, facilitates the analysis of sunshine duration and monthly irradiation data on a detailed province-by-province basis of Erzurum, which is located in the eastern region of Turkey and exhibits a strong potential for solar energy investment, particularly in the southern portion of the province, where the annual irradiation approaches 1700 kWh/m².

This study examines the energy production of bifacial panels as a dependent variable of snowfall type and snow cover thickness in Erzurum province, Turkey. Then, 7 different provinces are selected from 7 geographical regions of Turkey, and the effect of the number of days with snow cover is analyzed.

In Erzurum, particularly during the winter months, temperatures often fall below zero, and substantial snowfall is a common occurrence. Erzurum province was selected as the study site due to its prolonged periods of cold and snow-covered days. The data presented in

Table 1. Climatic data of Erzurum province [24].

	Jan.	Feb.	March	Apr.	May	June	July	August	Sept.	Oct.	Nov.	Dec.	Annual
Average Temp. (°C)	−9	−7.6	−2.3	5.4	10.7	14.9	19.2	19.5	14.8	8.2	1.2	−5.8	5.8
Average Sunshine Duration (hours)	3.4	4.5	5.2	6.4	7.9	10.2	11.2	10.7	9.1	6.9	4.9	3.2	7.0
Average Number of Rainy Days	11.23	10.98	12.54	13.7	16.04	10.98	6.75	5.2	5.14	9.44	9.19	10.59	121.8
Average Monthly Total Precipitation (mm)	21.5	25.4	35.5	54.6	73.1	48.5	28.6	17.6	24.1	46.7	33.3	22	430.9
Maximum Temp. (°C)	10.4	10.6	21.4	26.5	29.6	32.3	35.6	36.5	33.3	27	20.7	14	36.5
Minimum Temp. (°C)	−36	−37	−33.2	−22.4	−7.1	−5.6	−1.8	−1.1	−6.8	−14.1	−34.3	−37.2	−37.2

, derived from MGM [24] and encompassing measurements from 1929 to 2024, include monthly insolation duration, rainy days, and total precipitation. As shown in Table 1, the minimum number of sunshine hours is observed in the winter months (December: 3.2 h; January: 3.4 h), while the average annual sunshine duration is 7 h. In provinces where the average temperature is below 0 °C during the winter months, extremely low temperatures (between −36 °C and −37 °C) have been recorded, particularly in January, February, and December. These cold weather conditions indicate that precipitation in the winter months falls mainly in the form of snow. A further analysis of the precipitation data reveals that the number of days with precipitation in the winter months is approximately one-third of each month. This is evident in the following figures: December: 10.59 days, January: 11.23 days, and February: 10.98 days. Furthermore, the precipitation levels exceed 20 mm on these days. On 23 February 2004, the snow height reached its maximum, with a recorded value of 110 cm.

Because of the high solar energy potential of Erzurum, the long-term cold climatic conditions, and the heavy snowfall during the winter months, the use of bifacial panels is a good option. While the more efficient operation of the panels at low temperatures contributes positively to the energy production, the high albedo value of the ground due to snow in winter allows for higher energy production compared to monofacial panels.

The latitude shown in Figure 2 is the location of the investigation, 39.90° N, the longitude is 41.26° E, and the altitude is 1918 m in the Yakutiye district of Erzurum province. The climatic data of the selected site are shown in

Table 2. Annual solar energy and meteorological data for selected locations.

Month	Global Horz. Irradiation (kWh/m2/month)	Diffuse Horz. Irradiation (kWh/m2/month)	Temp. (°C)	Wind Speed (m/s)	Linke Turbidity	Relative Humidity (%)
January	79.1	23.2	−11.1	2.2	2.677	84.2
February	93.1	35.8	−8.9	2.39	3.034	84.9
March	149.2	44.6	−1.2	3.4	3.529	74.3
April	186.3	48.3	5.5	3.6	4.223	66.6
May	208.4	50.5	11.3	3.3	3.659	65.2
June	237.6	60.5	15	3.2	3.169	57.1
July	230	57.2	19.7	3.7	3.226	51
August	215.9	52.4	19.3	3.89	3.211	44.8
September	174.2	42.3	14.4	3.1	2.86	49.1
October	124.7	35	8	2.9	3.204	65.6
November	83.6	25.8	0	2.59	2.763	74.4
December	68.2	23	−7.5	2.2	2.638	83.1
Year	1850.3	520.9	5.4	3	3.183	66.7

.

The climate data presented for the Yakutiye district allow for a detailed analysis of the region’s solar radiation, temperature, wind speed, atmospheric turbidity, and relative humidity. The total annual global horizontal irradiation has been measured at 1850.3 kWh/m², while the diffuse horizontal irradiation is 520.9 kWh/m². In terms of solar radiation, the most productive months are June (237.6 kWh/m²) and July (230.0 kWh/m²), whereas the lowest values are observed in December (68.2 kWh/m²) and January (79.1 kWh/m²). This indicates that the spring and summer months are more suitable for solar energy systems, while the lower radiation levels in winter may negatively impact energy production.

The temperature variations in the region clearly reflect the influence of a harsh continental climate. The annual average temperature is 5.4 °C, with the lowest recorded in January (−11.1 °C) and the highest in July (19.7 °C). Particularly in spring and winter, significantly low temperatures are observed. Cold but sunny days can enable solar panels to operate at peak performance. For example, while a temperature of 19.7 °C in July is relatively favorable for solar panels, the lower temperatures and high irradiation levels in April and May contribute to increased system efficiency.

Regarding wind speed, the annual average wind speed in Yakutiye is 3.0 m/s. The highest wind speeds are recorded in August (3.89 m/s) and March (3.40 m/s), while the lowest values are observed in January (2.2 m/s) and December (2.2 m/s).

The Linke Turbidity (LT) index, which represents the level of atmospheric turbidity, has been calculated at an average of 3.183. The highest LT value is observed in April (4.223), while the lowest is recorded in December (2.638). This suggests that during the spring months, solar radiation is more scattered in the atmosphere, resulting in increased diffuse radiation received by solar panels. These findings indicate that both direct and diffuse radiation should be considered when designing solar panel systems in the region.

When examining relative humidity levels, the annual average humidity is 66.7%. The highest humidity is recorded in January (84.2%) and February (84.9%), whereas the lowest humidity is observed in August (44.8%). During the winter months, high humidity levels combined with cold weather may lead to fog, frost, and heat losses, while in summer, the lower humidity levels create a more favorable environment for solar panel efficiency.

Cell temperature is a critical parameter that impacts PV module performance. Several models have been proposed to predict the module temperature, such as Sandia [25] and Faiman [26]. According to the Sandia model, which is a simple model for predicting the bifacial module temperature [27], you have:

$$T_{md}=T_{amp}+\left(G_f+G_r\right)·\left(e^{a+b.Ws}\right)$$

(3)

where $T_{md}$ is the module temperature (°C);$G_r$ refers to the rear irradiance (W/m²) that is received by the back-side of the bifacial photovoltaic module, and $G_f$ denotes the front irradiance (W/m²), which represents the solar radiation incident on the front surface of the panel. In bifacial systems, both the front and rear surfaces contribute to energy generation, making it essential to distinguish between these two irradiation components. Also, a and b are two empirical coefficients; and $T_{amp}$ and Ws are the ambient air temperature (°C) and wind speed (m/s), respectively. For the glass-cell-glass modules in open racking, values of a = −3.47 and b = −0.0594 are suggested [25,27].

Also, a model formulation was found in the literature by King et al., 2004 [25] and has been used in several studies for predicting the back-side temperature of PV modules under varying environmental conditions, as shown in Equation (4). It refers to the Sandia temperature model, an empirical approach developed to estimate the back-surface temperature of photovoltaic modules based on ambient temperature, solar irradiance, and wind speed. It differs from the more commonly cited NOCT-based or inverse-linear wind speed models, but it remains a valid empirical model for certain module configurations, especially under high irradiance and varying wind conditions. Equation (3) is based on a modified form of the Sandia module temperature model, particularly the version that expresses the module back-surface temperature $T_m$ as a function of incident irradiation $E$, wind speed $w_s$, and ambient temperature $T_a$, according to the following expression [25,28,29]:

$$T_m=E·\left(e^{a+b·w_s}\right)+T_a$$

(4)

Both models consist of an ambient temperature term and an irradiance-related additive component that is modulated by wind speed. Both use an exponential decay function with wind speed to reflect enhanced convective cooling as the wind speed increases. Also, the introduced models are empirical and require the calibration of coefficients a and b using site- or system-specific data.

When the variables are accepted as $G_f+G_r=1000$ W/m², a = −3.47, b = −0.0594, $Ws=1$ m/s, and e = 2.71, the module temperature as a function of ambient temperature can be expressed as follows:

$$T_{md}=T_{amp}+\left(1000\right)·\left({2.71}^{-3.52}\right)$$

(5)

As illustrated in Figure 3, the ambient temperature has a linear effect on the temperature of the bifacial module.

The following equation is a simple model for bifacial PV power [27,30]:

$$P_{md}=P_{md0 }\left(1+\gamma \left(T_c-T_0\right)\right)·\left(G_f+K_b.G_r\right)/E_0 ·\left(1+D_{obs}\right)$$

(6)

where

$P_{md}$ is the maximum power for a bifacial module (W);

$P_{md0}$ is the module power rating in STC conditions (W);

γ is the temperature coefficient for maximum power (1/°C);

$T_c$ is the cell temperature (°C);

$G_f$ is the front irradiance (W/m²);

$K_b$ is the bifaciality of the PV module’s $P_{md}$;

$G_r$ is the rear irradiance (W/m²);

$E_0$ is the reference irradiance (W/m²);

$D_{obs}$ is a derate (unitless) that is based on the amount of rear-side obstruction [27,30].

It is clear that the module temperature is a parameter that directly affects output power. In snowy climatic conditions, during periods of heavy snowfall, it is necessary to carry out periodic snow removal operations using snowplow machines at the plant site. This is considered a standard part of plant operation and maintenance procedures. Moreover, in regions with frequent and intense snowfall, the mechanical and structural requirements of the photovoltaic system must be designed in accordance with these climatic parameters. Structural elements such as columns, beams, and purlins should be dimensioned and reinforced based on the local snow and ice load data by using appropriate engineering calculations.

However, this study specifically focuses on a detailed technical analysis of the performance of bifacial photovoltaic systems under various snow climate conditions, emphasizing the impact of optical parameters such as albedo. Therefore, structural design aspects and maintenance cost evaluations are beyond the scope of this study.

From a cost–benefit perspective, operating a photovoltaic power plant in colder climates and at higher altitudes generally leads to increased energy production due to improved panel efficiency in low temperatures [31,32,33]. Solar cells exhibit a decline in efficiency, with a 0.5% decrease occurring for every 1 °C rise in temperature [33]. Additionally, the high albedo effect of snow-covered ground further enhances rear-side energy gains in bifacial modules, significantly boosting the overall system performance. Considering the typical economic lifetime of a solar power plant is around 30 years, installations in such regions offer a more advantageous investment in the long term, with greater energy output and improved return on investment. This phenomenon is cited in recent literature, with empirical studies demonstrating that lower ambient temperatures significantly enhance the electrical output of PV modules [32,34,35,36]. This means their output increases by ~0.4–0.5% for every 1 °C decrease in temperature below standard test conditions (25 °C).

The panels were positioned in the location specified at 35° tilt and 180° azimuth, as shown in Figure 4. The 35° tilt and 180° azimuth values used in this study were selected based on the local climatic conditions of Erzurum, which is characterized by cold winters and persistent snow cover. In fact, Bakirci determined the optimum tilt angle for Erzurum as 34.3° [37]. In regions like Erzurum, where snowfall is frequent and heavy, a 35° tilt helps prevent snow accumulation on the panel surface, thereby reducing energy losses and ensuring structural stability under snow loads. This tilt angle is also commonly used in field applications for its mechanical resilience in snowy environments. The 180° azimuth, corresponding to due south, is the standard orientation for maximizing the annual energy yield in Turkey. Thus, in this study, tilt and orientation were kept constant to isolate and evaluate the impact of the snow characteristics on bifacial PV performance. Future work will consider a more detailed optimization of the tilt and azimuth angles to enhance performance under varying environmental conditions.

In this study, the selected 35° tilt and 180° azimuth angles are based on commonly applied field configurations in regions with frequent and heavy snowfall, such as Erzurum. A tilt angle of 35° is known to effectively prevent snow accumulation on the panel surface, thereby minimizing energy losses and maintaining panel efficiency during the winter months. The 180° azimuth ensures optimal annual solar irradiance by aligning the system directly south. Therefore, the chosen tilt and azimuth values are not derived from a theoretical optimization process but instead reflect practical and field-based engineering considerations.

Moreover, since the primary objective of this study is to comparatively analyze the effect of snow cover-related albedo variations on the performance of bifacial PV systems under different snow climate conditions, the tilt and azimuth parameters were kept constant across all simulations. This approach allows the isolated evaluation of albedo effects without interference from other variables. Thus, this study emphasizes a focused assessment of the optical impacts of snow conditions rather than a general design optimization analysis.

The system components consist of Eco Green Energy panels with a power of 500 W, and their I-V characteristics are shown in Figure 5. The inverter, which is another component, is the Layer brand, with a power of 4.00 kW (Nominal AC), which is preferred due to its high efficiency. The technical specifications of the inverter are delineated in

Table 3. Technical specifications of the inverter.

Parameter	Value
Manufacturer	Layer
Model	GC-204
Minimum MPP Voltage	100 V
Maximum MPP Voltage	430 V
Nominal MPP Voltage	400 V
Absolute Max. PV Voltage	500 V
Min. Voltage for Pnom	100 V
Maximum Input Current	21.1 A
Power Threshold	0 W
Grid Frequency	50 Hz/60 Hz
Grid Voltage	230 V
Nominal AC Power	4.00 kW
Maximum AC Power	4.20 kW
Nominal AC Current	17.40 A
Maximum AC Current	18.00 A
Nominal PV Power	4.50 kW
Maximum PV Current	21.00 A
Maximum Efficiency	96.00%

.

In the analyses performed with the software Pvsyst 8.0.6, four different snow types were accepted as the base. The types of snow and the ranges of albedo values, depending on the type of snow, are presented in

Table 4. Albedo values for snow conditions.

Snow Type	Albedo Lower Value	Albedo Upper Value
Wet snow	0.50	0.70
Old, dry snow	0.70	0.80
Fresh, dry snow	0.80	0.95
Melting ice/snow	0.25	0.80

, and the average of these ranges is used in this study [18]. The concern regarding the assumption of constant monthly albedo values is valid, as snow conditions indeed exhibit temporal variability on daily or even hourly scales. However, the aim of this study is to provide a comparative and regionally representative analysis of bifacial PV performance under typical snow conditions across different climatic zones. Given the limitations of available long-term and high-resolution snow albedo datasets, monthly averaged values were adopted as a practical approximation to reflect the dominant surface conditions during the winter months. Due to the technical limitations of PVsyst, monthly albedo inputs were preferred; in addition, a weighted average albedo was calculated for each month by taking into account the number of days with snow cover, which was obtained from the MGM. Consequently, more realistic values were obtained according to the rate of snow presence during the month. While this approach may not capture short-term fluctuations, it enables consistent and reproducible modeling across multiple scenarios. Future work may incorporate higher temporal resolution datasets or dynamic albedo models to refine the accuracy of simulations, especially for site-specific performance forecasting.

The system under consideration was generated according to these 4 different snow conditions. In the computation of the amount of energy to be generated by the panels throughout the year, it is assumed that there is snowfall during the winter months (December, January, and February) and not during the other months of the year. In other words, it is assumed that the ground is covered with snow during the winter months and that the ground is covered with grass during the other months of the year. The albedo value of grass is reported to range from 0.15 to 0.25 [38]. In the present study, 0.25 was adopted as the albedo value. Utilizing these assumptions, the monthly albedo values for each snow type are presented throughout the year in Figure 6.

3. Results and Discussion

3.1. Performance Analyses of Monofacial and Bifacial Photovoltaic Modules

The performance of monofacial and bifacial photovoltaic panels is a significant area of research. Consequently, a comparison of bifacial and monofacial systems was conducted under similar environmental conditions before investigating the snow scenarios. This comparison aims to elucidate the potential benefits of bifacial panels more distinctly. In the case of Erzurum province, all system parameters (panel power, orientation, tilt angle, inverter specifications, etc.) were maintained as constant, with the sole modification being the replacement of the panel type to monofacial. Simulations were then conducted. The characteristics of the monofacial and bifacial modules are illustrated in

Table 5. Characteristics of bifacial and monofacial modules.

Properties	Bifacial Panel	Monofacial Panel
Model	EGE-500W-108N(GM10R)	EGE-500W-108N(M10R)
Nominal power	500 Wp	500 Wp
Tecnology	Si-mono	Si-mono
Short-circuit current (Isc)	15.9 A	15.9 A
Open circuit (Voc)	39.30 V	39.30 V
Module length	1961 mm	1961 mm
Module width	1134 mm	1134 mm
Weight	28.50 kg	25.00 kg
Number cells in series	54·2	54·2

.

In

Table 6. Monthly energy generation and performance ratio of the bifacial and monofacial panels.

	Bifacial Panel		Monofacial Panel
Month	Energy Generation (kWh)	Performance Ratio	Energy Generation (kWh)	Performance Ratio
January	131.0	1.823	44.7	0.622
February	139.9	1.187	79.6	0.676
March	269.7	0.993	188.5	0.694
April	466.9	0.967	389.5	0.807
May	614.2	0.936	544.3	0.830
June	730.9	0.920	660.2	0.831
July	680.5	0.909	610.2	0.816
August	545.2	0.918	465.9	0.785
September	340.9	0.933	259.0	0.709
October	165.4	1.061	91.0	0.584
November	119.0	1.492	49.7	0.623
December	124.9	1.915	39.9	0.612
Year	4328.5	0.983	3422.6	0.777

, the monthly energy production amounts and performance ratios of the monofacial and bifacial solar panels are presented. Despite the unchanging environmental conditions, such as irradiance, temperature, and wind, the bifacial panel exhibits a consistent increase in energy production on a monthly basis. The annual energy production of the monofacial module is 3422.6 kWh, whereas the annual energy production of the bifacial module is 4328.5 kWh. The annual performance ratios are 0.777 (monofacial) and 0.983 (bifacial), respectively. As demonstrated in Figure 7, the bifacial panel exhibits a distinct advantage in terms of energy production and performance ratio. This finding indicates that the bifacial panels generate approximately 26.4% more production, even in the absence of snow cover, which is consistent with the documented bifacial gain values ranging from 25 to 30% reported in the literature [39].

3.2. Performance Analysis of Snow Conditions

This section presents the findings resulting from the implemented adjustments. In the first case, the investigation focused on the impact of different snowfall types on annual energy production. To assess this effect, a scenario with no precipitation during the winter months was used as a reference condition. The analysis revealed that the annual energy production was 4328 kWh without snowfall (albedo = 0.25 in all months). The height of the panels above the ground is assumed to be 1 m, and the results obtained for different snow types are given in

Table 7. Annual energy values for different snow types.

Condition	Snow Type	Lowest Annual Energy Production (for Lower Albedo Value) (kWh/yr.)	Highest Annual Energy Production (for Upper Albedo Value (kWh/yr.)
0	No snow	4328 (reference)	4328 (reference)
1	Wet snow	4439	4526
2	Old, dry snow	4526	4570
3	Fresh, dry snow	4570	4635
4	Melting ice/snow	4328	4570

, under the assumption that the other variables remain constant. Additionally, these results can also be seen graphically in Figure 8. The highest energy production is observed in the third case, i.e., a fresh, dry snow precipitation type, with a value up to 4635 kWh, which is about 7% more energy production potential compared to the reference value of 4328 kWh.

In the second case, the effect of the snow depth on energy production was investigated, and it was determined that the amount of snowfall is another factor affecting the albedo value. The albedo value generally increases with snow thickness, likely due to the surface features (such as soil or grass, etc.) decreasing with thickness, and only the snow cover causes reflection. The albedo value ranges from 0.5 to 0.7 at a snow thickness of 1–4 cm (cm), approaches 0.8 at a thickness of 4–5 cm, and exceeds 0.8 at a thickness of 10 cm [15].

The average albedo values, depending on snow thickness, can be determined as follows:

• Approximately 0.5 when the snow thickness is 1–2 cm;
• Approximately 0.6 when the snow thickness is 2–3 cm;
• Approximately 0.7 when the snow thickness is 3–4 cm;
• Approximately 0.8 when the snow thickness is 4–5 cm;
• Above 0.85 when the snow thickness is more than 10 cm.

The annual energy generation values obtained according to snow thickness and related albedo values are presented in

Table 8. Annual energy values for different snow depths.

Condition	Snow Depth	Albedo Value	Annual Energy Production (kWh/yr.)
1	1–2 cm	0.5	4439
2	2–3 cm	0.6	4483
3	3–4 cm	0.7	4526
4	4–5 cm	0.8	4570
5	5–10 cm	0.82	4579
6	10 cm–upper	0.88	4605

and Figure 9. While the annual production was 4439 kWh in the first case of low snowfall (1–2 cm), it reached 4605 kWh, representing an increase of approximately 4% at 10 cm and above. As illustrated in Figure 9, the annual energy production exhibits a positive correlation with snow thickness, reaching its maximum value at 10 cm and above.

In the third case, the variation of the effect of the number of days with snow cover on albedo, according to different climate zones and its effect on energy production, was investigated. In Turkey, the number of days with snow cover generally increases as one moves toward the eastern part of the country. There are regions in Turkey where snowfall occurs only a few days a year, as well as regions where snow cover is seen for months. Figure 10 illustrates the annual average number of days with snow cover in Turkey between 1970 and 2023 [40]. The average number of days with snow cover across the country is 27.8 days.

In this study, long-term meteorological data from 1970 to 2023 were utilized to model the albedo-related contribution of snow cover to energy generation across Turkey’s seven geographical regions. These data reveal significant regional differences in snow cover duration, forming the basis for the assignment of albedo values used in simulations. The albedo values were calculated and modeled by taking into account region-specific physical characteristics such as the snow type, density, and surface aging. The simulations were performed using representative fixed albedo values that were selected within these defined ranges. This approach enabled a comparative analysis by isolating the effect of albedo on performance from other fixed parameters such as tilt and azimuth.

Albedo depends not only on the presence of snow, but also on its type, age, and level of contamination. The physical structure of snow in different regions can produce different albedo values for these reasons.

Table 9. Number of days with snow cover and albedo ranges for seven regions of Turkey [18,20,40].

Region	Sample Provinces	Snow Cover Duration (Day/Year)	General Albedo Value
Eastern Anatolia	Erzurum, Kars, Ağrı, Van	80–120 days	0.70–0.90
Central Anatolia	Ankara, Eskişehir, Sivas	30–60 days	0.50–0.75
Black Sea	Kastamonu, Gümüşhane, Bolu	10–40 days	0.40–0.65
Marmara	İstanbul, Bursa, Kocaeli	5–30 days	0.35–0.60
Aegean	Afyon, Uşak, Kütahya	5–20 days	0.30–0.55
Mediterranean	Isparta, Burdur, Kahramanmaraş	0–10 days	0.25–0.50
Southeastern Anatolia	Diyarbakır, Gaziantep, Şanlıurfa	0–5 days	0.20–0.45

shows the number of days with snow cover, albedo ranges, and sample provinces showing the characteristics of each region for the seven regions of the country.

The following characteristics can be said about the regions:

• Eastern Anatolia (Erzurum, Kars, Ağrı): 0.70–0.90

  o

  Cold and dry snow has a higher albedo value (0.85–0.95).

  o

  Since the snow cover stays for a long time, the albedo may decrease over time due to aging (0.70–0.80).

  o

  Since the region has less industry and air pollution, the purity of the snow may be higher.

• Central Anatolia (Ankara, Sivas, Eskişehir): 0.50–0.75

  o

  In Central Anatolia, melting and refreezing are more common in winter.

  o

  The aging process of snow accelerates, which decreases the albedo value.

  o

  Snow can be polluted more quickly due to industrial and traffic emissions, and the albedo can drop to 0.50.

• Black Sea (Kastamonu, Gümüşhane, Bolu): 0.40–0.65

  o

  Due to high humidity and temperature fluctuations in the Black Sea region, snow is usually wet and heavy.

  o

  Wet snow has a low albedo value (0.40–0.55).

  o

  Due to frequent precipitation in the region, the snow surface is usually not clean, which causes the albedo to remain low.

• Marmara (Istanbul, Bursa, Kocaeli): 0.35–0.60

  o

  Due to the impact of industry and urbanization, the snow darkens faster due to air pollution.

  o

  In and around Istanbul, snow cover is polluted within a short time, and albedo can decrease rapidly.

• Aegean and Mediterranean (Afyon, Isparta, Antalya): 0.25–0.50

  o

  Snow cover is short-lived and melts quickly, so it is not possible to keep the snow clean and shiny all the time.

  o

  Usually, when there is a thin layer of snow, the albedo is low due to the effect of the ground.

• Southeastern Anatolia (Diyarbakır, Gaziantep, Şanlıurfa): 0.20–0.45

  o

  Snowfall is very low and usually remains as a thin layer.

  o

  Since snow mixes with soil and pollution within a short time, albedo decreases rapidly.

In view of this information, a representative province from each region of the country was selected, and regional evaluations were made. Details of the selected provinces, their location, snowfall time, albedo values, and the amount of energy produced are presented in

Table 10. Variations in snow cover duration’s impact on albedo across different climate regions and its effect on energy production.

Region	Province	Site	Accepted Time Interval for Snow Cover	Accepted Albedo Value	Annual Energy Production (kWh/yr.)
Eastern Anatolia	Erzurum	Latitude: 39.90° N Longitude: 41.26° E Altitude: 1918 m	December, January, February, and March	0.9 (Dec, Jan, Feb, and Mar) 0.25 (Other 8 months)	4799
Central Anatolia	Eskişehir	Latitude: 39.77° N Longitude: 30.51° E Altitude: 789 m	December and January	0.75 (Dec, Jan) 0.25 (Other 10 months)	3144
Black Sea	Kastamonu	Latitude: 41.38° N Longitude: 33.78° E Altitude: 795 m	December and January: first 10 days	0.65 (Dec) 0.38 (Jan) 0.25 (Other 10 months)	2972
Marmara	Bursa	Latitude: 40.19° N Longitude: 29.06° E Altitude: 252 m	December	0.6 (Dec) 0.25 (Other 11 months)	3142
Aegean	Uşak	Latitude: 38.67° N Longitude: 29.40° E Altitude: 913 m	January: first 20 days	0.45 (Jan) 0.25 (Other 11 months)	3390
Mediterranean	Isparta	Latitude: 37.78° N Longitude: 30.57° E Altitude: 997 m	January: first 10 days	0.33 (Jan) 0.25 (Other 11 months)	3362
Southeastern Anatolia	Gaziantep	Latitude: 37.07° N Longitude: 37.38° E Altitude: 828 m	January: first 5 days	0.28 (Jan) 0.25 (Other 11 months)	3647

. The PVsyst simulation program was used to conduct this analysis.

In this study, one province from each of Turkey’s seven geographical regions was selected to represent regional diversity, and simulations were carried out using the PVsyst 8.0 software based on long-term meteorological data from these locations. While this approach is methodologically meaningful for identifying general regional trends, it is acknowledged that it has certain limitations in terms of statistical representativeness. Therefore, this study serves as a comparative regional analysis rather than a basis for comprehensive statistical generalization, and it also provides a reference for similar evaluations in other locations. In future studies, this approach can be expanded by including multiple provinces from each region and by analyzing intra-regional variations through more extensive modeling. In doing so, both regional representativeness and the statistical robustness of the conclusions regarding the performance of bifacial PV systems under different climatic conditions will be significantly improved.

Albedo values can be defined on a monthly basis within the PVsyst program. Consequently, in the months characterized by snow cover, it is imperative to recalculate the monthly albedo values for periods of less than 30 days, such as 5 and 10 days, taking into account the number of snowy days. To analyze the albedo effect more accurately, the albedo values were recalculated using Equation 7. The surface reflectivity was determined by the weighted average method, depending on the number of snowy days. On snowy days in the selected province, the ground was considered completely covered with snow, and the snow albedo was used. On days without snow, the surface was assumed to be covered with grass, and the monthly average albedo value was obtained by summing the contributions of these two components.

$$A_{new}=\left({\displaystyle \frac{D_{snow}·A_{sr}}{30}}\right)+\left({\displaystyle \frac{{( 30-D}_{snow})·A_g}{30}}\right)$$

(7)

where

$A_{new}$ is the new albedo value;

$D_{snow}$ is the number of snowy days;

$A_{sr}$ is the maximum albedo value of the region where the province is located;

$A_g$ is the grass albedo (0.25).

For instance, the albedo value of Gaziantep in January is calculated as (0.45·5/30) + (0.25·25/30) = 0.28. These calculations were performed for each province, and the albedo values to be used throughout the year were applied, as demonstrated in Figure 11.

As illustrated in Figure 12, the annual energy production values of each region are presented. The maximum production was attained in Erzurum province, located in the Eastern Anatolia region, with a value of 4799 kWh. Conversely, the minimum production was recorded in Kastamonu province, located in the Black Sea region, with a value of 2972 kWh.

Table 11. Comparison of studies on snow and albedo impact on bifacial PV performance.

Study	Location	System Type	Snow/Albedo Parameter	Effect on Energy Yield	Contribution
Riise et al. [7]	Norway	Fixed Tilt 35° Bifacial	Albedo-enhancing membrane (40–60%), snow	Bifacial gain 17%	Combined effect of artificial albedo and snow contribution
Molin et al. [8]	Sweden	Fixed Tilt 40° Bifacial/Vertical E–W	Albedo (snow + ground), snow cover days	Bifacial gain 5–48%	Direct measurement of snow-induced albedo effects
Singh & Jones [9]	USA, Utah	Horizontal Bifacial vs. Monofacial	Albedo 75%, snow melting effect	Bifacial modules cleared snow 2–3 days faster → lower energy loss	Snow dynamics and albedo benefit emphasized
Ghafiri et al. [11]	Canada	Fixed Tilt 30° Bifacial	Snow depth ≥2 cm, albedo >0.9	Winter bifacial gain 28.4%	Snow and albedo impact analyzed
This Study	Turkey (7 regions)	Fixed Tilt Bifacial	Snow type, snow depth, snow cover duration	Fresh snow ~7% gain; ≥10 cm snow ~4% gain; eastern regions up to 60% higher annual yield	First systematic regional analysis of snow and albedo impacts in a mid-latitude country

summarizes the recent studies that examined the effects of snow and albedo on bifacial photovoltaic (PV) system performance. Previous studies have shown that an increase in surface albedo due to snow can significantly improve bifacial gain [7,8,9,11]. The findings of this study are consistent with these results and confirm that the effects of snow and albedo contribute positively to energy production in regions with significant snow cover. Additionally, this study is the first to systematically analyze the impact of snow type, depth, and duration of snow cover on energy production across different climate zones in a mid-latitude country like Turkey.

4. Conclusions

This study aims to evaluate the impact of snow cover and regional meteorological conditions on power generation by investigating the performance of bifacial panels under snowfall. The findings revealed that the impact of snowfall should be considered in a comprehensive manner.

In all simulation scenarios, the system size was normalized to 1.00 kW of installed PV capacity in order to enable direct comparison of energy yield results. Accordingly, all reported annual energy production values (e.g., 4799 kWh in Erzurum) should be interpreted as kWh per installed kW of PV (kWh/kW.year). This allows for performance benchmarking across different climatic zones and with other studies in the literature.

The primary objective of this study was to analyze in detail the optical effects of snow-related albedo on the energy production of bifacial PV systems. Accordingly, the analyses were conducted using the PVsyst simulation program, focusing on how variations in snow cover duration across different climatic regions influence albedo and, consequently, system performance. However, the mechanical stress caused by snow load on panel surfaces during winter is a critical issue in terms of both structural integrity and long-term performance. For this reason, mechanical and structural evaluations should be addressed in a separate study. Using structural analysis software such as ANSYS or SolidWorks, deformation and safety limits of the panel mounting systems under snow and ice loads can be examined. Additionally, based on the magnitude of such loads, the appropriate cross-sectional dimensions and thicknesses of structural elements such as columns, beams, and purlins should be analyzed individually to ensure the safe and efficient design of the support structure.

In general, the energy generation of bifacial panels is positively influenced by factors such as snow cover, snow thickness, and the number of days with snow cover. In contrast to single surface panels, the back-surface contribution can be maximized with a highly reflective surface, such as snow.

In the analysis for Erzurum province with the same system components, the annual energy production of bifacial and monofacial panels is 4328 kWh/year and 3423 kWh/year, respectively. This shows that even in the absence of snow, the bifacial system provides approximately 26.4% more production. Especially in regions with long snow cover, the contribution from the back-surface reflection makes this difference even more significant. This result clearly shows that bifacial panels are more advantageous than monofacial systems in winter conditions.

The type of snowfall is a determining factor in energy generation, with fresh, dry snow leading to higher energy production, and wet snow or old snow appearing to reduce energy production. In this case, it can be inferred that the decrease in a plain smooth surface due to wetting and the tendency of the snow to melt, as well as pollutants such as dust, are effective. The analysis shows that, depending on the type of precipitation, the potential for energy production can be increased by approximately 7%.

The amount of snowfall was also found to positively affect energy generation. It was observed that an increase of approximately 4% in energy production was associated with snowfall accumulations of 10 cm or more. Consequently, the implementation of bifacial panels in mountainous and high-altitude regions with substantial snowfall has the potential to enhance energy production. However, it is imperative to carefully adjust the distance between the panel and the ground to optimize this effect, particularly in cases where excessive snow accumulation is anticipated.

In this study, the panels were installed at a 35° tilt angle, which is widely accepted in the literature as suitable for snowy regions. This tilt serves as a critical design parameter that physically prevents snow from accumulating on the panel surface for extended periods. Therefore, the analyses did not assume persistent snow coverage on the panels; rather, it was acknowledged that at a 35° tilt, snow would naturally slide off the surface and not obstruct panel operation. Heidari et al. examined the performance of PV systems in a heavy snowfall climate and found that increasing the panel tilt from 0° to 45° reduced the annual snow-related energy loss from approximately 34% to 5% [41]. Van Noord et al. analyzed multiple sites in northern Sweden over several winters and demonstrated through their modeling and observations that higher tilt angles significantly reduce the duration of snow coverage on PV panels [42,43,44]. The study by Cooper et al. presents an innovative method for detecting snow-related energy losses using inverter data and validates it across PV systems with different tilt angles. Their results show that panels with higher tilt angles shed snow more quickly, and that snow losses are inversely correlated with the tilt angle [45]. Moreover, in real-world applications, following heavy snowfall, snow removal vehicles or manual interventions are typically employed to clear snow from both the panel surfaces and the mounting structures. On sunny days, it has also been observed that snow melts quickly due to both the panel’s own heat and the high albedo effect of the reflective ground surface. This supports the assumption that the positive albedo contribution used in the simulations is realistically achievable under actual field conditions.

In addition, regional differences are also important factors affecting energy generation. The eastern region of the country exhibits approximately 60% higher energy production than regions with low insolation, such as the Black Sea region, due to the prolonged duration of snow cover. Furthermore, the Eastern Anatolia region demonstrates a 30% increase in energy production over the southeastern region, despite its high insolation. These findings show that the presence of snow is an important potential for bifacial panels, even in regions with low irradiation.

One of the key findings of this study is that natural snow cover, particularly under fresh and dry snow conditions, can increase rear-side reflectivity (albedo) and boost annual energy production by up to 7% in high-altitude regions like Erzurum. In large-scale power plants, this represents a substantial energy gain. Similarly, this situation is applicable to other cities around the world that share similar climatic characteristics with Erzurum, such as Quebec City in Canada, Duluth in the U.S. state of Minnesota, Lviv in Ukraine, Sibiu in Romania, Sion in Switzerland, and Lillehammer in Norway—all of which experience long, snowy winters and cool, short summers. This demonstrates that snow is not a hindrance to panel performance but rather a natural advantage that can enhance production. In this context, the need for active snow removal is often eliminated, which, in turn, reduces the complexity and cost of operation and maintenance (O&M) processes. Indeed, field applications in cold climates such as Canada and Norway have shown that when a system’s design is optimized (e.g., proper panel height, tilt, orientation), passive solutions are generally sufficient to cope with snow. The 1 m panel height used in our simulations is based on this design philosophy, serving as a passive strategy to prevent surface obstruction due to snow accumulation. Although our study does not include a direct cost analysis, the productivity benefits of snow cover and the absence of a need for regular snow cleaning significantly reduce the active maintenance burden on the system.

In this study, albedo-based optical contributions to the performance of bifacial photovoltaic (PV) systems under different snow climate conditions have been thoroughly analyzed. However, snow removal requirements and the associated operational costs were excluded from the scope of this research. In real-world conditions, especially in regions with heavy snowfall, snow removal is carried out periodically as an operational practice to maintain system efficiency [46]. These operations are planned within the long-term maintenance strategy of the plant and typically do not require frequent or high-cost interventions. According to a report published by the National Laboratory of the U.S. Department of Energy [47] on operation and maintenance (O&M, operation and maintenance) costs, periodic maintenance activities such as snow removal represent only a limited portion of the total O&M expenses. This indicates that the impact of snow removal on total investment costs is minimal. Therefore, the economic effect of snow removal operations is limited, and the energy gains achieved through albedo enhancement are more than sufficient to offset such operational expenses.