Quantitative effects of PM concentrations on spectral distribution of global normal irradiance

Highlights

• The influences of PM₁₀ concentrations on SPD values are investigated.

• A strong influence of PM₁₀ on SPD could be noticed ranging from 380 to 540 nm.

• A band of 510 nm is adopted to establish models and its models has been validated.

• Other bands are also validated and the band ranges from 480 to 520 nm.

• Quantitative study on SPD under PM₁₀ levels promotes building energy-efficient design.

Abstract

Daylight refers to a free and valuable resource exhibiting photometric and radiometric features, and the physical parameters commonly used to represent its features are irradiance and illuminance. They are both obtained by the spectral power distribution (SPD) of global irradiance, demonstrating that values noticeably affect indoor thermal and luminous environments. Nevertheless, outdoor particulate matter (PM) significantly affects SPD conditions, and two relative spectral notions, i.e., percentage of spectral irradiance (PSI_r) and percentage of spectral illuminance (PSI_l), were proposed to study the effects of PM₁₀ concentrations in radiometry and photometry with a field measurement, Beijing. As a result, the strong impact of PM₁₀ on PSI_r/ PSI_l can be noticed in the range (380 to 540 nm) (P-value < 0.01) by correlation analysis, and a wavelength of 510 nm was taken to build models between PM₁₀ and PSI_r/ PSI_l based on regression analysis. With this wavelength, PSI_r varies from 0.29 to 0.25%, and PSI_l ranges from 0.48 to 0.42% with the increase of PM₁₀ concentrations from 0.01 to 0.098 mg/m³. With the verification of the model, several bands were further analyzed to comprehensively verify the monotonic correlations, and the band ranging 480 ~ 520 nm were reasonably proved that PSI_r/PSI_l and PM₁₀ can be well indicated by each other. As influenced by the complexity of spectral transmittance through the glazing, the mutual characterization between solar spectrum and particulate matters can contribute to the building energy-efficient design and develop indoor thermal and luminous environment research in the future.

Introduction

The variation of the solar radiation directly or indirectly affects numerous meteorological conditions and biological processes on Earth. Such resource is constantly valuable, clean, and abundant in most places (Kreider and Kreith, 1981), and it is available on the earth’s surface (Leckner, 1978) for beneficial applications, particularly in the building sector that accounts for over one-third of total energy consumptions in most countries worldwide (Pérez et al., 2008).

The daylight through the building envelope refers to a major element that affects buildings’ energy performance and occupant’s thermal comfort for its radiometric features (La Gennusa et al., 2007). The solar radiation on building envelope is critical to assessing the active and passive energy-saving practices (Xie et al., 2017). However, the daylight through the windows and skylights specifically affects the occupants’ visual comfort (Xue et al., 2014) and non-visual effect (Dai et al., 2017) (e.g., circadian rhythms (Rea and Figueiro, 2018)). In photometric features, it is generally known to be able to affect physical, physiological (Xue et al., 2016) and psychological characteristics (Vetter et al., 2011). Thus, researchers and residents increasingly focus on the solar radiation.

The solar radiation acts as a vital factor in indoor thermal and luminous environment with photometry and radiometry. It is also noticeably known that the irradiance (Gueymard, 2004) and illuminance (Gueymard, 2008) are the physical parameters commonly adopted to represent the features of solar radiation; they are both calculated based on the spectral power distribution of global irradiance (abbreviated as SPD) (Chain et al., 1999), which has been a hotspot in the built environment with extensive studies.

As fueled by the advancement of building materials technology (Cuce and Riffat, 2015), the spectral selectivity and active control of solar radiation transmittance (Gueymard, 2009) and absorptance (Moreno and Hernández, 2018) have been more complicated. The SPD will affect the building cooling load (Gueymard and Thevenard, 2009) with the different transmittance on different wavelengths, especially in the application of photovoltaic technologies (Nazeeruddin et al., 2011) as well as the calculation for glazing’s secondary heat transfer (Piccolo et al., 2018). Except for radiometry features, SPD values also create a specific daylighting environment and determine the color rendering index (CRI) (Aste et al., 2015), correlated color temperature (CCT) (Ghosh et al., 2018), and luminous intensity (Bellia et al., 2011) for the building environment in photometry features. However, atmospheric turbidity is a main parameter controlling the attenuation of spectral distribution of global irradiance reaching the Earth’s surface under cloudless sky conditions (Gueymard and Garrison, 1998). Its reduction mechanism results both from the scattering of daylight by tiny particles suspended in the air (e.g., water droplets, dust, and particulate matters) and from the absorption of daylight by water vapor and carbon dioxide (Lopez and Batlles, 2004). In the studies of the relation between SPD and atmospheric turbidity, some researchers have been paying more attention to the scattering effects of PM (Fig. 1.) on solar spectrum in the atmosphere (Chaâbane et al., 2004).

In recent years, the world has witnessed China’s rapid economic and industrial development, which has triggered tremendous rises in energy consumptions (Ebohon, 1996) and emissions of air pollutants (Zhao et al., 2015). Air pollution originates from multiple sources (Guo et al., 2020), and it has imposed progressively serious effects on public human health (Löndahl et al., 2010). As revealed from the data harvested from the air quality monitoring stations in 338 cities in China, the annual average concentrations of the PM_2.5 (cutoff sizes ≤ 2.5$\mu$ m) ranged from 11 to 125$\mu$ g/m³ in 2015 and from 12 to 158$\mu$ g/m³ in 2016 with an annual average value of 50 and 47$\mu$ g/m³, respectively and the PM was the primary pollutants for over 80.3% of the days with huge pollution (Lin et al., 2018). It has been concluded that the particulate matter (PM), especially PM_2.5, refers to the primary pollutant in most cities and adversely affects public health for the energy-consumed structure that the fossil resources are still applied in China extensively (Wu et al., 2015).

Given the critical impact of the solar spectrum on the built environment and the scattering effects of PM, the outdoor SPD values should be studied under a range of PM concentrations. However, since the relation between PM levels and SPD has been rarely studied, the following review will only present the investigations for solar radiation with the effects of PM concentrations.

In studies on the relation between solar radiation and PM values, the scattering of direct solar radiation should be primarily considered (Ladjevardi et al. 2013). Yang et al. (2016) measured direction solar radiation at various PM concentrations; they reported that PM values significantly affected the direct solar radiation available at the surface in China. Haywood et al. (1997) suggested that the rising atmospheric pollution, especially anthropogenic particulate matter, has split sunlight into scattering light and reduced direct surface solar radiation. Other researchers found that various additional indexes that can present the attenuation of solar radiation. Based on solar radiation and PM data measured in Beijing, Yao et al. (2017) built novel daily diffuse solar radiation models. Zhao et al. (2013) statistically analyzed the relations between visibility and PM concentrations in northeast China; as revealed from their results, the PM noticeably affected visibility and solar radiation.

The effects of PM values on solar radiation have been investigated based on the aforementioned studies. However, solar radiation is affected not only by the scattering of particulate matter, but also by other factors. Consequently, considerable researchers have begun to accurately analyze the effect of PM values on solar radiation under complex conditions. Furlan C et al. (2012) assessed the surface hourly values of diffuse solar radiation with PM concentrations considering the effects of clouds. Galindo et al. (2011) measured daily concentrations of PM₁, PM_2.5, PM₁₀ with solar radiation and other meteorological parameters (e.g., wind speed and rainfall rate). Liu et al. (2015) developed a model that considers air pollution, temperature, humidity, and wind speed data; they concluded that PM in the atmosphere displayed a robust linear correlation with solar radiation.

Given our review of the literature, the analysis was based on the data of monitoring solar radiation and PM concentrations. Accordingly, some statistical models were adopted to effectively conduct the study. To delve into the scattering effects of PM on solar radiation simplifying the effects of clouds, moisture, and other atmospheric factors, Liou et al. (2002) yielded the radiative equation with the numerical solution. Furlan C et al. (2012) presented a novel regression model based on clearness index and diffuse fraction to assess territory diffuse solar radiation with PM values considering conventional meteorological variables, and its assessing results were better than the developed ones. In addition to the mentioned parametric models, several non-parametric models were also applied. Sun et al. (2016) exploited meteorological and solar radiation data to develop random forest models for estimation of solar radiation. Vakili et al. (2017) illustrated that using novel methods (e.g., Artificial Neural Network modeling) can enable researchers to make more accurate predictions in solar radiation. Fan et al. (2018) proposed a method by considering seven single air pollution parameters and 15 combinations of two parameters with Support Vector Machine; as revealed from their results, the PM_2.5 was the most relevant air pollution parameter for diffuse solar radiation.

In the development of building façade energy-efficient design and daylighting conditions, solar radiation acts as a crucial factor for researchers and residents. While SPDs’ values more significantly affect our indoor environment for the complexity in spectral selectivity based on the emerging glazing. Numerous existing studies analyzed the relation between PM concentrations and solar radiation, whereas the effects of PM values on SPD have been rarely studied.

This paper elucidates the relation between PM concentrations and SPD based on the experiments conducted in Beijing. The indication with PM concentration on the solar spectrum will vary. To be specific, 1) the difference between global horizontal irradiance (GHI) and global normal irradiance (GNI) measurement is demonstrated; 2) the stronger relation, PM concentrations with irradiance and illuminance, is revealed in two measurement; 3) effects of PM₁₀ on SPD are investigated, and its efficiency and accuracy are verified.