Aerosol–PAR Interactions: Critical Insights from a Systematic Review (2021–2025)

Abstract

Atmospheric aerosols significantly influence photosynthetically active radiation (PAR), critical for plant photosynthesis and ecosystem functioning. This study systematically reviewed recent research (2021–2025) on aerosol–PAR interactions. Using targeted keywords, 22 open-access articles from Scopus and Google Scholar were analyzed via VOSviewer for thematic, methodological, and geographic trends. Analysis revealed a strong concentration in Earth and Environmental Sciences, showcasing significant advances in radiative transfer modeling, remote sensing, and machine learning for estimating aerosol impacts on PAR. Studies primarily utilized satellite data and models (e.g., DART, SCOPE) to assess diffuse/direct radiation changes. The literature consistently demonstrates how aerosols modulate PAR, influencing canopy light penetration and photosynthetic efficiency. However, critical gaps persist, including limited field validation in tropical biomes (e.g., Amazon, Cerrado, Pantanal) and a lack of studies differentiating aerosol types like black and brown carbon. This synthesis underscores the need for expanded monitoring and integrated modeling efforts to improve understanding of aerosol–PAR interactions, particularly in underrepresented tropical regions.

1. Introduction

Prior to 2021, the impact of atmospheric aerosols on photosynthetically active radiation (PAR) had been extensively documented in the scientific literature, particularly regarding their capacity to modulate the proportion between direct and diffuse radiation [1,2,3,4,5]. This phenomenon, known as the diffuse fertilization effect, demonstrated the potential to increase photosynthetic efficiency in various ecosystems, particularly favoring environments with high leaf area density and complex canopy structure [4,6,7]. However, significant gaps persisted until that time, mainly related to limited empirical validation in tropical and subtropical biomes, such as the Amazon, Cerrado, and Pantanal. In these regions, specific conditions like intense seasonal biomass burning and different aerosol types, such as black and brown carbon, complicate precise predictions of their effects on PAR and vegetation productivity [8,9,10]. The exact quantification of these impacts at regional scales remained an open challenge in the scientific community [4,11].

Building upon this established understanding and addressing these persistent challenges, atmospheric pollution has emerged as a major driver of environmental change in recent climate scenarios, directly influencing the propagation of solar radiation through absorption and scattering mechanisms [12,13]. These interactions modify both the quantity and quality of radiation, understood as the balance between direct and diffuse components and the spectral distribution of incident light, particularly within the PAR range, that reaches the Earth’s surface, thereby regulating fundamental processes such as photosynthesis and the ecosystem energy balance.

The choice to highlight the Brazilian tropical biomes—Amazon, Cerrado, and Pantanal—is based on three main factors. First, these regions host one of the planet’s highest biodiversities and feature ecosystems highly dependent on solar radiation for maintaining primary productivity. Second, they are seasonally impacted by intense biomass burning, responsible for massive emissions of anthropogenically derived aerosols, especially black and brown carbon. Third, they remain underrepresented in field observational validation studies. While temperate and Asian regions have been the focus of aerosol and PAR monitoring with more consolidated time series, Brazilian tropical biomes still lack adequate coverage by flux towers, AERONET networks, and integrated measurement campaigns [9,10,14]. These gaps compromise the capacity for accurate regional modeling and limit the direct transfer of results obtained in other ecological contexts.

Among the affected spectral ranges, photosynthetically active radiation (PAR) stands out as the band between 400 and 700 nm that is used by plants for photosynthesis [15]. PAR is one of the main factors determining the primary productivity of terrestrial ecosystems [16] and serves as an essential parameter in ecosystem models that incorporate the physiological, biological, and physical processes of plants [17].

Atmospheric aerosols play a relevant role in modulating the partitioning between direct and diffuse radiation, especially under clear skies, where particle nuclei significantly scatter solar photons [18]. However, clouds and the solar zenith angle are considered the primary modulators of radiation transmission on a global scale. Studies show that clouds are responsible for attenuating solar radiation much more intensely than aerosols, directly blocking solar rays and reflecting a large part of the incident energy [19,20]. Aerosols also alter the balance between direct and diffuse radiation components, which can substantially affect photosynthetic efficiency and, consequently, influence vegetation productivity [16].

Aerosols originate from both natural and anthropogenic sources and exhibit varied chemical compositions, including sulfates, nitrates, organic carbon, black carbon, and mineral dust. This diversity determines their optical properties and radiative impacts, allowing them to act either as cooling agents by reflecting solar radiation or as warming agents by absorbing it [21,22,23]. Beyond their composition and types, the cooling or warming of the surface is also related to the vertical structure of aerosols [24].

Beyond their direct radiative effects, aerosols also exert indirect influences on PAR by acting as cloud condensation nuclei (CCN). By modifying cloud microphysical properties, such as droplet size and cloud lifetime, aerosols can alter cloud albedo and optical thickness, thereby indirectly affecting the amount and distribution of solar radiation, including PAR, reaching the Earth’s surface [25,26].

Notably, all aerosols increase the proportion of diffuse radiation relative to direct radiation reaching plant canopies, a phenomenon observed across different aerosol types, varying with their composition and concentration [27]. Aerosols from biomass burning, specifically, are a prominent source in many tropical regions and contribute significantly to this effect. Under certain conditions, this phenomenon can enhance photosynthetic potential by promoting a more uniform light distribution within the canopy layers, known as the diffuse fertilization effect [28].

While methodological advances in satellite remote sensing, radiative transfer modeling, and machine learning have significantly improved our capacity to study aerosol–PAR interactions globally, the persistent shortage of field-based validation studies in tropical regions remains a critical challenge. This limitation is especially pronounced in Brazil’s diverse biomes such as the Amazon, Cerrado, and Pantanal, which face intense biomass burning and land-use changes that affect aerosol loads and radiation regimes. Without detailed local measurements and validations, the impacts of aerosols on PAR remain poorly quantified, hindering accurate predictions of their ecological and climatic consequences.

This study aims to conduct a systematic review of the scientific literature published between 2021 and 2025, mapping thematic trends, methodological approaches, and key knowledge gaps in research on the interactions between atmospheric aerosols and photosynthetically active radiation. By synthesizing these recent findings, we seek to identify the specific advances made in understanding aerosol–PAR interactions, particularly in addressing the persistent challenges in tropical biomes, and to inform future interdisciplinary investigations that integrate atmospheric sciences, ecology, and climate change research.

2. Materials and Methods

2.1. Databases and Search Strategy

The systematic literature search was conducted using the Scopus and Google Scholar databases. Scopus was selected as the primary source due to its extensive coverage of the peer-reviewed scientific literature across multiple disciplines and its robust citation indexing [29], ensuring comprehensive capture of established research. Google Scholar was utilized as a supplementary source to identify potentially relevant studies, including those not yet fully indexed in Scopus or from the grey literature, thereby minimizing publication bias.

The search strategy employed specific English-language terms designed to focus on studies examining the interactions between atmospheric aerosols and photosynthetically active radiation (PAR). The precise search string used in Scopus was TITLE-ABS-KEY ((“radiative transfer model” OR “aerosol optical depth”) AND “photosynthetically active radiation”).

Filters were applied to restrict results to the document type “Article” and language “English”. The temporal scope of the search was limited to studies published between 2021 and 2025. This time frame was chosen to ensure the review captured the most recent developments and emerging trends in radiative transfer modeling, aerosol optical properties, and their impacts on PAR, reflecting the dynamic nature of research in this field [29,30].

2.2. Inclusion and Exclusion Criteria

Articles were selected based on predefined inclusion and exclusion criteria to ensure relevance and consistency. Inclusion criteria required that articles (a) contained the descriptors “radiative transfer model” or “aerosol optical depth” and “photosynthetically active radiation” in the title, abstract, or keywords (in English); (b) were published within the defined time frame (2021–2025); and (c) were open-access articles available for full review. The inclusion and exclusion criteria applied in this review are summarized in

Table 1. Article selection criteria in the Scopus database.

Item	Description
Search terms	TITLE-ABS-KEY ((“radiative transfer model” OR “aerosol optical depth”) AND “photosynthetically active radiation”)
Databases	Scopus and Google Scholar
Time frame	Last five years (2021–2025)
Inclusion criteria	(a) Articles containing in the title, abstract, and keywords the descriptors “radiative transfer model” or “aerosol optical depth” and “photosynthetically active radiation” in English; (b) articles published within the last five years (2021 to 2025); (c) open-access articles available online in full.
Exclusion criteria	(a) Articles without open-access availability; (b) articles published outside the defined time frame (2021 to 2025); (c) review articles, book chapters, data papers, and other document types not corresponding to research articles.
Document type	Research articles
Number of articles found	Scopus (67); Google Scholar (3)
Articles included	22
Analysis of selected data	Microsoft Excel; VOSviewer version 1.6.20

.

Exclusion criteria removed any articles (a) without open-access availability; (b) published outside the 2021–2025 time window; and (c) document types not corresponding to primary research articles, such as reviews, book chapters, data papers, and other non-research formats. These criteria were designed to focus the review on recent, accessible, peer-reviewed primary research directly addressing aerosol and PAR interactions [29].

2.3. Selection Process

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines [31]. These guidelines provide a structured approach for identifying, selecting, and reporting the relevant literature in a transparent and reproducible manner.

The Scopus and Google Scholar databases were used as information sources. Scopus, in particular, accounted for the majority of identified publications and is widely recognized as one of the most comprehensive databases of the peer-reviewed scientific literature, covering a broad spectrum of academic disciplines [29].

In the first stage, after defining the search terms, time window, and document type, a total of 70 records were retrieved (67 from Scopus and 3 from Google Scholar). After removing 1 duplicate, 69 records were screened by title and abstract.

During the second stage, two records were excluded for not meeting the thematic or methodological inclusion criteria. The remaining 67 records were assessed for full-text availability, and 4 could not be retrieved. As a result, 63 articles were selected for full-text review.

In the third and final screening stage, a detailed reading of the 63 full-text articles was performed. A total of 41 studies were excluded because, while meeting initial inclusion criteria, they did not introduce novel methodological approaches or distinct thematic insights beyond those already identified and sufficiently represented by other included studies. This refinement process ensured that the final selection provided a comprehensive overview of the diverse methodological and thematic trends within the defined scope of aerosol–PAR interactions, focusing on studies that offered unique contributions to the mapping of recent advances and persistent gaps. Selected studies were subsequently categorized into thematic groups to aid in qualitative synthesis and discussion.

Ultimately, 22 studies met all eligibility criteria and were included in the final analysis. The complete selection process is summarized in the PRISMA 2020 flow diagram (Figure 1). The categorization of these studies into two main thematic areas—(a) effects of aerosols on radiation and vegetation productivity, and (b) radiative modeling and remote sensing—is presented in

Table 2. Thematic categorization based on shared methodological approaches among the selected articles. Source: Authors.

Category	References
(a) Effects of Aerosols on Radiation and Vegetation Productivity	Ma et al. [16]; Zhang et al. [17]; Shu et al. [32]; Gui et al. [33]; Zhang et al. [34]; Crivelari-Costa et al. [35]; Franco et al. [21]; Rodrigues et al. [14]; Curado et al. [36]; Lozano et al. [18]; Yuan et al. [12]; Li et al. [37]; Bai et al. [38]; Zuo and Wang [39]; Quintanilla-Albornoz et al. [40]
(b) Radiative Modeling and Remote Sensing	Boitard et al. [41]; Wang et al. [42]; Liu et al. [43]; Zhang et al. [44]; Zhang et al. [22]; Foyo-Moreno et al. [15]; Regaieg et al. [45]

.

2.4. Data Analysis

Data from the selected articles were organized and analyzed using Microsoft Excel for management and descriptive statistics. For bibliometric and thematic mapping, the VOSviewer software was employed. VOSviewer is an open-source Java-based tool that generates network-based maps of bibliometric data, enabling the visualization of co-authorship networks, institutional collaborations, and keyword co-occurrence patterns [30].

Keyword co-occurrence networks were constructed to identify prominent thematic clusters and emerging areas of research related to atmospheric aerosols and photosynthetically active radiation. This approach allowed the detection of interconnected research trends and provided insights into methodological advances in the field.

It is important to note that this review did not include a formal appraisal of methodological quality or risk of bias of the included studies, as its primary aim was to map recent thematic and methodological trends rather than to synthesize quantitative effect sizes. Additionally, the search was restricted to articles published in English and available in open-access format, which may have excluded relevant studies published in other languages or behind paywalls. These limitations should be considered when interpreting the findings of this scoping review.

3. Results

3.1. Descriptive Overview of Included Studies

3.1.1. Subject Area Distribution of Publications

The analysis of the subject areas of the 22 articles included in this systematic review reveals a strong concentration in the Earth and Planetary Sciences, which account for 50% of the publications analyzed (Figure 2). Significant contributions also came from Agricultural and Biological Sciences (28%) and Environmental Science (24%). The remaining subject areas show markedly lower representation, including Engineering (6%) and smaller shares (1–2% each) in Energy, Physics and Astronomy, Chemistry, Computer Science, Decision Sciences, Multidisciplinary, and Social Sciences.

3.1.2. Journal and Source Analysis

Regarding the distribution of studies across scientific journals (Figure 3), Remote Sensing of Environment stands out with 13 publications, followed by Remote Sensing (7), Agricultural and Forest Meteorology, and IEEE Transactions on Geoscience and Remote Sensing (5 each).

Other journals with lower but still relevant frequency include Atmospheric Environment, Atmospheric Research, Ecological Modelling, Journal of Quantitative Spectroscopy and Radiative Transfer, and Science of the Total Environment (each with two publications). The remaining journals contributed only a single article and are grouped under “Other” in Figure 3.

3.1.3. Geographic Distribution of Studies

In terms of geographic distribution, China was the country with the highest number of studies (17 articles), followed by Spain (5), and both the United States and Brazil (4 each). Other countries with lower representation include India, France, and the United Kingdom (two each), while Switzerland, Japan, Italy, Germany, Cyprus, Greece, and the Netherlands each accounted for a single publication (Figure 4).

3.2. Keyword Co-Occurrence Network Analysis

To identify the main themes addressed in the selected publications, a keyword co-occurrence analysis was performed using the VOSviewer software. The generated map showed an interconnected structure of topics in which the term “photosynthetically active radiation” occupies a central position (Figure 5). The co-occurrence analysis results highlighted the formation of five major thematic areas.

A prominent cluster of studies focused on radiative transfer modeling and remote sensing, with a strong presence of terms such as radiative transfer modelling, radiometers, surface albedo, and top of atmosphere. These works, mostly based on satellite data like MODIS and Sentinel-2, described the behavior of solar radiation along its path to the surface, as well as the effects of its interaction with atmospheric particles and surface properties.

Another identified axis referred to ecological and agricultural applications of PAR, marked by terms related to cultivated vegetation and canopy structure, such as plants, crops, canopy architecture, and laser method. Studies in this line, such as Boitard [41], explored the use of multispectral sensors mounted on unmanned aerial platforms to calibrate three-dimensional models (e.g., DART-3D), enabling spatially detailed estimates of PAR distribution across different vegetation types [45].

In the atmospheric context, the effects of aerosols on the partitioning of solar radiation into its direct and diffuse components were frequently addressed. The presence of terms such as aerosol, AOD, air pollutant, and sunlight indicated a focus on how suspended particles alter the diffuse fraction of radiation and influence photosynthetic processes. Zuo and Wang [39] investigated the effects of reduced atmospheric aerosols (AOD) during the COVID-19 pandemic on the gross primary productivity (GPP) and water use efficiency (WUE) of winter wheat plantations. The study demonstrated that the decrease in aerosol concentrations had a positive impact on these indicators, linking cleaner atmospheric conditions to the agronomic performance of winter wheat. Although aerosol reduction tends to decrease the diffuse fraction of radiation, often associated with increased photosynthetic efficiency in dense canopies, the study observed a significant increase in total irradiance, which, in this specific context, mitigated the impact of the diffuse radiation reduction on productivity.

Additionally, the relationship between radiation and meteorological and hydrological variables was evident, especially with terms such as evapotranspiration, soil moisture, and drought.

Finally, the analysis showed the emergence of a data science-focused thematic cluster, with terms such as mapping, machine learning, and budget control, reflecting the incorporation of artificial intelligence techniques in processing and analyzing large volumes of environmental data. Some terms appeared isolated on the map, such as atmospheric aerosol, aquatic ecosystems, and temperature.

3.3. Thematic Categorization of Research Approaches

3.3.1. Effects of Aerosols on PAR and Vegetation Productivity

The analysis of the 22 selected articles reveals consistent evidence that variations in atmospheric aerosol concentrations influence PAR availability, radiation quality, and consequently vegetation productivity and carbon fluxes. Multiple studies report that changes in aerosol levels modify gross primary productivity (GPP), water use efficiency, and carbon sequestration in agricultural systems. Under certain conditions, aerosols increase the fraction of diffuse PAR, enhancing canopy light penetration and photosynthetic efficiency [16,17].

Several works reported the need to explicitly incorporate aerosol effects into GPP models to improve carbon cycle assessments [32]. For example, research modeled diffuse fertilization effects in Chinese ecosystems using MODIS-based AOD and GPP data, showing stronger responses in areas with high vegetation cover and adequate water availability, while arid regions exhibited limited effects [33].

Other studies analyzed the relationship in Central Asia over long time periods, combining MODIS products and geographically weighted regression to demonstrate spatially variable GPP responses to AOD, with agricultural lands and shrublands displaying the most significant positive diffuse fertilization effects [16]. Some researchers also highlighted the importance of considering intra-daily variation in FAPAR under clear-sky conditions, noting that hourly dynamics in radiation diffusion can significantly affect productivity estimates [17].

Investigations in Brazil explored these interactions in tropical contexts. One study used MODIS and GOSAT data to estimate CO₂ fluxes, finding that land-use changes increase net emissions and highlighting the importance of monitoring regional carbon balance [35]. Other work using AERONET sun photometer data in the central Amazon characterized seasonal variations in aerosol refractive indices, showing how compositional changes influence the regional radiative balance [46].

Additional research focused on semi-deciduous forests in southern Amazonia, combining eddy covariance data with MODIS AOD and AERONET measurements to demonstrate that high aerosol loads during the dry season increase diffuse radiation and enhance CO₂ uptake, while reducing air temperature and vapor pressure deficit (VPD), thus lowering thermal and water stress [14]. Similarly, studies in the Pantanal biome using eddy covariance and PAR data showed that high black carbon levels during biomass burning reduce available PAR, increase air temperature, and diminish carbon assimilation [36].

Studies in agricultural contexts also revealed interesting responses to aerosol reduction during the COVID-19 lockdown in China. During this period, an atypical situation was observed: although aerosol concentration decreased, leading to an increase in total irradiance, there was also an increase in plant productivity, despite the expected reduction in the diffuse fraction. In a rice crop in Northeast China, Zuo and Wang [39] identified that a reduction in AOD of approximately 30% resulted in 5% increases in gross and net primary productivity (GPP and NPP), FPAR, and LAI. In Chinese urban areas, NPP significantly increased with the decrease in AOD and the increase in PAR, with AOD and temperature emerging as the main factors explaining this response [37]. Although an increase in the diffuse fraction under lower aerosol loads is unexpected, as physically the presence of aerosols is expected to intensify this fraction, the findings, by indicating greater radiation availability even under lower aerosol concentrations, highlight the importance of considering specific meteorological or seasonal conditions as potential contributors to this effect.

3.3.2. Radiative Transfer Modeling and Remote Sensing Methods

Several studies employed advanced radiative transfer models (RTMs) and remote sensing approaches to investigate aerosol and PAR interactions. One line of research focused on calibrating 3D models such as DART using UAV and Sentinel-2 data to assess radiative budgets in maize systems under conventional and agroecological management [41].

Other investigations combined SCOPE modeling with sun-induced chlorophyll fluorescence (SIF) measurements to separate physiological from optical responses to water stress in sugar beet crops, using UAV platforms for data acquisition [42]. There was also work applying terrestrial laser scanning (TLS) to create bitemporal 3D forest reconstructions, which were then analyzed with DART 3D simulations to assess changes in LAI, FAPAR, and canopy light extinction over time [43].

In addition, research employed machine learning algorithms trained on synthetic datasets generated by 6S RTM simulations to estimate variables such as LAI, FAPAR, PAR, albedo, ISR, and AOD from VIIRS TOA radiances, demonstrating high accuracy when validated against MODIS, GLASS, and AERONET products [17]. Another study refined this approach by using Random Forest models trained on RTM outputs to produce high-resolution FAPAR estimates from Landsat surface reflectance without the need for in situ measurements [47].

Simplified modeling approaches were also proposed, including methods for estimating aerosol radiative forcing on PAR using only global irradiance and solar position data under clear skies. Such methods provide a cost-effective alternative to complex RTMs and showed good agreement with SBDART and CPCR2 model outputs [15].

Overall, these methodologies illustrate how combining RTMs, satellite imagery, UAV data, and machine learning enables more accurate estimation of variables such as PAR, FAPAR, albedo, and SIF under varying atmospheric aerosol loads. Overall, these methodologies demonstrated the combination of RTMs, satellite imagery, UAV data, and machine learning for the estimation of variables such as PAR, FAPAR, albedo, and SIF under varying atmospheric aerosol loads.

4. Discussion

4.1. Synthesis and Interpretation of Identified Research Trends

The systematic review of the recent literature on aerosol and PAR interactions reveals a clear convergence of research efforts on modeling radiative transfer, monitoring aerosol properties, and assessing implications for vegetation productivity. This convergence underscores a maturing field, with a consistent emphasis on the pivotal role of aerosols as modulators of solar radiation quality, particularly their capacity to shift the balance between direct and diffuse PAR fractions. This modulation is widely recognized as a key factor influencing canopy light penetration and photosynthetic efficiency, especially in densely vegetated or agricultural systems, thereby directly impacting carbon uptake and ecosystem functioning [16,17,32].

Researchers have increasingly adopted advanced radiative transfer models and remote sensing techniques to better quantify these complex interactions. The frequent use of satellite data from sensors such as MODIS, Sentinel-2, and VIIRS highlights a reliance on large-scale observational capabilities to support assessments of aerosol optical depth (AOD) and its impact on surface radiation budgets, enabling spatially explicit evaluations of diffuse fertilization effects [16,33,44]. However, the reviewed studies also demonstrate that such effects are highly context-dependent, varying significantly with canopy structure, soil moisture availability, and regional climate patterns [12,33], underscoring the need for localized and nuanced interpretations.

A further significant trend has been the integration of machine learning approaches with physically based models. This synergy aims to improve the retrieval of key variables such as LAI, FAPAR, and AOD from satellite observations, allowing for higher-resolution mapping of surface and atmospheric properties while reducing the reliance on extensive in situ measurements [44,47]. Similarly, the growing prominence of UAV-based data collection as a strategy for validating radiative transfer models at finer scales reflects an increasing demand for detailed representation of canopy structure and light distribution, bridging the gap between ground-based and satellite observations [41,42,43].

Ecological and agricultural applications have received notable attention, with studies highlighting the capacity of aerosols to enhance crop productivity under certain conditions by increasing the diffuse radiation fraction [37,38,39]. Concurrently, there is a growing recognition of the complex interactions among aerosol composition, biomass burning emissions, and seasonal climatic variability. These interactions can either enhance or suppress photosynthetic processes depending on local conditions [14,36], suggesting that the impact of aerosols is not linear but rather a result of intricate biophysical feedbacks.

Together, these trends underscore the inherently interdisciplinary and multi-scale nature of research in this area. The sustained interest in improving low-cost, simplified approaches to estimating aerosol radiative forcing on PAR, particularly in data-scarce regions [15], further emphasizes the importance of integrating atmospheric physics, remote sensing, ecological modeling, and data science to advance a comprehensive understanding of aerosol and PAR interactions and their broader environmental implications.

4.2. Critical Perspectives on Aerosol–PAR Interactions and Ecosystem Responses

4.2.1. Modulation of Direct and Diffuse Radiation: Nuances and Complexities

A recurring finding across the reviewed literature is the crucial role aerosols play in altering the balance between direct and diffuse components of PAR. Generally, increased aerosol concentrations, often linked to biomass burning and urban pollution, reduce the intensity of direct radiation while enhancing the proportion of diffuse light reaching the canopy [16,32,34]. This shift has important implications because diffuse light penetrates more effectively into dense canopies, promoting photosynthesis in shaded leaves and potentially increasing overall carbon uptake.

However, the review also highlights instances where the relationship between aerosol load and diffuse radiation is not straightforward, particularly under specific atmospheric conditions or in response to rapid changes in aerosol concentrations. For example, studies during the COVID-19 pandemic lockdown in China observed an atypical situation where, despite a decrease in aerosol concentration (and thus an expected reduction in diffuse fraction), plant productivity increased [39]. This seemingly contradictory finding can be attributed to a significant increase in total irradiance, which, in specific contexts like crops with more open architecture or under energy limitation, can compensate for the potential negative impact of reduced diffuse radiation on productivity. This underscores that the response of photosynthesis to atmospheric changes is neither linear nor universal, but rather depends on factors such as canopy type and local atmospheric conditions [39,48].

Furthermore, efforts to incorporate these processes into radiative transfer models and large-scale land surface models underscore the complexity of simulating the spectral distribution and angular characteristics of radiation under varying aerosol loads [44,47]. This complexity makes it critical to better quantify not only the amount of PAR but also its direct/diffuse partitioning under realistic atmospheric scenarios, considering the synergistic effects of aerosols with meteorological variables (such as relative humidity and temperature) and vegetation characteristics [48].

4.2.2. Implications for Photosynthetic Efficiency and Ecosystem Productivity

Beyond radiation partitioning, several studies have emphasized the profound impact of aerosols on ecosystem carbon balance and agricultural productivity. Under specific conditions, increased diffuse PAR can enhance gross primary productivity (GPP) by improving light use efficiency, particularly in high-LAI crops or forests [16,33]. This diffuse fertilization effect is a key mechanism by which aerosols can positively influence terrestrial ecosystems.

However, research has consistently demonstrated that responses to aerosol-induced changes in radiation quality are not uniform. Factors such as canopy architecture, soil moisture, and nutrient availability mediate the magnitude of the diffuse fertilization effect [12,33]. For instance, analyses in Central Asia have shown spatially heterogeneous GPP responses to aerosol variations, with agricultural lands often showing more pronounced benefits [16]. Conversely, in arid regions, where photosynthesis is limited by water scarcity, the impact of aerosols on GPP is often reduced, aligning with environmental expectations [33].

At the same time, modeling approaches have identified that ignoring aerosol effects can lead to substantial biases in estimating GPP and carbon sequestration, underscoring the critical need to integrate these processes into ecosystem productivity models [32]. Studies using eddy covariance towers and satellite data have also begun to validate these model predictions, showing that increases in diffuse radiation during periods of high aerosol loading can indeed coincide with enhanced carbon uptake under certain conditions [14,36].

4.2.3. Regional Case Studies and the Tropical Biome Gap

Despite global advances in understanding aerosol–PAR interactions, a critical insight from this review is the limited number of field-validated studies focusing on tropical biomes, particularly in South America. Research in the Amazon, Cerrado, and Pantanal has begun to highlight the unique dynamics of aerosol and PAR interactions in these regions, where biomass burning contributes substantial seasonal aerosol loads [14,21,36]. The spatial distribution of studies suggests that the mechanisms by which aerosols affect PAR vary significantly with regional context, not only due to direct scattering and absorption effects but also due to aerosols’ capacity to act as cloud condensation nuclei (CCN), altering cloud properties such as reflectivity, lifetime, and cover, which indirectly impacts surface radiation [49,50]. This highlights the non-uniformity and significant spatial and climatic variability of aerosol impacts.

For example, studies combining eddy covariance measurements with satellite-derived AOD have documented that high aerosol concentrations during the dry season can reduce direct PAR while increasing the diffuse fraction, enhancing CO₂ uptake in certain forest systems. However, this benefit may be offset by increases in air temperature and reductions in overall PAR when black carbon concentrations are particularly high [14,36]. Furthermore, changes in land use and deforestation intensify these dynamics by altering local emission patterns and atmospheric composition [35], and it is important to highlight that land use and land cover change can substantially modify aerosol fluxes to the atmosphere, both through direct emissions and through the formation of secondary particles, potentially having direct implications for PAR modulation and surface energy balance [51].

Overall, the limited number of tropical case studies highlights a significant research gap, suggesting the urgent need for expanded field measurements, improved ground-based PAR monitoring, and integration with remote sensing and modeling efforts tailored to tropical conditions. This gap is particularly concerning given the high biodiversity and critical role of tropical ecosystems in global carbon and water cycles.

4.3. Advances and Persistent Limitations in Radiative Transfer Modeling and Remote Sensing

4.3.1. Satellite Observations and Data Products: Progress and Challenges

One of the most significant advances identified in the literature is the expanded use of satellite observations to quantify aerosol optical depth (AOD) and model its impact on PAR. Studies have increasingly relied on data from sensors such as MODIS, Sentinel-2, VIIRS, and Himawari-8 to provide spatially and temporally consistent estimates of atmospheric aerosol properties and surface radiation budgets [17,33].

Such satellite products enable the assessment of aerosol–radiation interactions over large regions, offering critical insights into how AOD variability modifies the balance between direct and diffuse PAR. However, research has also consistently highlighted the limitations of these datasets in resolving fine-scale variations in aerosol composition, especially in complex atmospheric conditions or regions with high cloud cover [12,18]. Furthermore, studies underscore that satellite-derived products often require careful validation against ground-based measurements, such as those from AERONET sun photometers, to ensure accuracy and reliability when used in ecological and climate modeling [21].

4.3.2. Radiative Transfer Models and Machine Learning Applications: Synergies and Future Directions

Radiative transfer models (RTMs) have become essential tools for simulating the interactions between solar radiation and atmospheric aerosols, enabling detailed exploration of how these particles alter the spectral and angular distribution of PAR. Research has applied sophisticated RTMs, such as DART 3D, SCOPE, and 6S, to better understand these processes under varying environmental conditions [41,42,43,44,47].

In addition, the integration of machine learning approaches with physically based models has emerged as a prominent trend. Several studies have trained algorithms using synthetic datasets generated from RTM simulations, improving the retrieval of variables such as LAI, FAPAR, PAR, albedo, ISR, and AOD from satellite radiance data [44,47]. This approach has enhanced the spatial resolution and accuracy of parameter estimation while reducing dependence on labor-intensive field campaigns. Nevertheless, while these methods show promise, they also introduce new challenges, including the need to ensure that training datasets adequately capture the diversity of real-world conditions, especially in under-studied regions with unique atmospheric properties or land-cover types.

4.3.3. Challenges in Field Validation and Ground Truthing: The Persistent Need for In Situ Data

Despite methodological advances in modeling and remote sensing, a critical limitation consistently identified in the reviewed studies is the persistent lack of robust field-based validation, particularly in tropical biomes. Research has pointed out that many models and satellite products have been developed and calibrated primarily in temperate regions or agricultural systems in East Asia, potentially limiting their transferability and accuracy when applied to ecosystems such as the Amazon, Cerrado, and Pantanal [14,21,36].

Field campaigns using eddy covariance towers, PAR sensors, and sun photometers remain essential for improving the accuracy of aerosol and PAR interaction models and for validating diffuse/direct partitioning under realistic conditions. However, logistical challenges, limited sensor networks, and funding constraints continue to hamper such efforts in many tropical countries. Additionally, the variability in aerosol types, such as black carbon from biomass burning and mineral dust, requires targeted research to understand their specific optical properties and radiative effects. While some studies have begun to address these issues, there remains a pressing need for expanded, coordinated field measurement networks and collaborative efforts to improve model parameterization for tropical and subtropical environments.

4.4. Key Knowledge Gaps and Future Research Needs

Despite important methodological advances and an expanding body of research, this review highlights several persistent knowledge gaps that constrain a comprehensive understanding of aerosol and PAR interactions, particularly in tropical regions. A consistent limitation identified is the severe underrepresentation of tropical and subtropical ecosystems in existing datasets and model calibrations, as many remote sensing algorithms and radiative transfer models have been developed and tested primarily in temperate agricultural landscapes or urban environments in East Asia. This focus limits their applicability to tropical forests and savannas, where aerosol composition, emission sources, and atmospheric dynamics can differ substantially [16,33,34], underscoring an urgent need for more field-based measurements to validate satellite-derived products and radiative transfer model outputs under conditions typical of these biomes [14,21,36].

Furthermore, research has pointed out the need for better characterization of aerosol optical properties, including black carbon and brown carbon components associated with biomass burning, as variations in refractive indices, particle size distributions, and hygroscopic behavior significantly affect solar radiation scattering and absorption [18,21]. While some studies have begun to address these issues, field campaigns specifically targeting these diverse aerosol types in tropical regions remain scarce. There is also a critical need for integrated approaches combining satellite observations, ground-based measurements, radiative transfer modeling, and machine learning to improve predictive capabilities, as current models depend heavily on the availability of representative training datasets that include diverse land cover types and atmospheric conditions [44,47].

Finally, beyond these central themes, the co-occurrence analysis revealed isolated terms such as “atmospheric aerosol,” “aquatic ecosystems,” and “temperature,” suggesting emerging or less integrated lines of research. Specifically, the study of atmospheric deposition effects on aquatic environments represents a significant underexplored area, highlighting the need to expand investigations into coastal and aquatic ecosystems where the impacts of aerosols on carbon cycling, nutrient dynamics, and photosynthesis are still poorly understood [52]. Addressing these knowledge gaps is crucial not only for advancing scientific understanding but also for informing land management, climate mitigation strategies, and agricultural planning, as improved models of aerosol and PAR interactions can help predict changes in ecosystem productivity under varying pollution and climate scenarios, supporting decision-making in regions particularly vulnerable to deforestation, fire regimes, and climate variability such as the Amazon, Cerrado, and Pantanal.

5. Conclusions

This systematic review unequivocally highlights the complex and multifaceted nature of interactions between atmospheric aerosols and photosynthetically active radiation (PAR). Over the past five years, significant methodological advances have been achieved in modeling radiative transfer processes, leveraging satellite and UAV-based remote sensing, and applying machine learning techniques. These innovations have substantially improved our capacity to estimate PAR, FAPAR, and related ecological variables, thereby deepening our understanding of how aerosols modulate the balance between direct and diffuse radiation and their subsequent effects on vegetation productivity and carbon fluxes.

Nonetheless, our analysis reveals critical and persistent knowledge gaps that severely limit the generalizability of current models, particularly within tropical biomes. Field-validated studies remain conspicuously scarce in regions like the Amazon, Cerrado, and Pantanal, despite their profound sensitivity to seasonal biomass burning and deforestation-driven changes in aerosol emissions. Existing models and remote sensing products, often calibrated with data from temperate or monsoonal systems, demonstrate questionable applicability to tropical forests and savannas, which possess distinct aerosol sources, optical properties, and canopy structures.

Addressing these fundamental gaps necessitates coordinated and substantial investments in ground-based PAR monitoring, expanded eddy covariance flux tower networks, and a more comprehensive characterization of aerosol optical properties, including black carbon and brown carbon components, specific to these understudied regions. Integrating these robust field data with advanced remote sensing observations and sophisticated radiative transfer models, further supported by cutting-edge machine learning methods, is essential for developing accurate, scalable, and globally representative predictions of ecosystem productivity under varying aerosol and climate conditions.

Ultimately, such concerted efforts are not merely critical for advancing fundamental scientific understanding; they are imperative for informing effective policy and management strategies in regions facing escalating pressures from land-use change, fire regimes, and climate variability. By strengthening the interdisciplinary integration of atmospheric physics, remote sensing, ecological modeling, and data science, future research can provide the indispensable tools required to better anticipate and mitigate the ecological impacts of atmospheric aerosols on both tropical and global scales.