Influence of Optically Active Substances on Light Attenuation in a Tropical Eutrophic Urban Reservoir

Abstract

This study investigated the impact of optically active substances on light attenuation in a tropical eutrophic urban reservoir under different seasonal conditions. Diffuse attenuation coefficients for photosynthetically active radiation (Kd_PAR) and ultraviolet radiation (Kd_UVA and Kd_UVB) were measured at three representative sites and correlated with water quality parameters (chlorophyll-a, total suspended solids [TSS], dissolved organic carbon, and colored dissolved organic matter [CDOM]). The results revealed significant spatial and seasonal differences, with the highest attenuation observed during the rainy season. The Ilha site exhibited the greatest coefficients (Kd_PAR = 6.0 m⁻¹, Kd_UVA = 17.9 m⁻¹, Kd_UVB = 19.0 m⁻¹), while lower values were recorded at Barragem (Kd_PAR = 2.4 m⁻¹, Kd_UVA = 9.1 m⁻¹, Kd_UVB = 12.0 m⁻¹) and Igrejinha (Kd_PAR = 3.1 m⁻¹, Kd_UVA = 10.8 m⁻¹, Kd_UVB = 11.9 m⁻¹). Statistical analyses showed strong correlations between TSS and Kd_PAR (r = 0.66) and between CDOM and both Kd_UVA (r = 0.66) and Kd_UVB (r = 0.59), with regression models confirming TSS and CDOM as key predictors of light attenuation. These findings underscore the pivotal role of particulate and dissolved organic matter in underwater light dynamics, emphasizing the need to reduce their input during periods of heavy rainfall.

1. Introduction

Solar radiation, encompassing ultraviolet radiation (UVR, 200–400 nm) and photosynthetically active radiation (PAR, 400–700 nm), is a fundamental driver of physical, chemical, and biological processes in aquatic ecosystems [1]. The attenuation of light in water depends on the presence and interaction of optically active substances (OAS), such as chromophoric dissolved organic matter (CDOM), non-algal particles (NAPs), and phytoplankton. These substances affect the underwater light environment by absorbing and scattering light, and their influence varies across regions and types of aquatic systems [2]. Widely used indicators, such as the Secchi depth (ZSD) and the light extinction coefficient (Kd), provide valuable insights into water quality and transparency and serve as critical parameters for managing aquatic environments, especially in eutrophic and turbid systems [3].

The relative importance of OAS in controlling light attenuation varies significantly among ecosystems. For instance, in reservoirs, such as Sau in Spain, total suspended solids (TSS) are the primary determinants of light extinction, whereas chlorophyll-a (Chla) plays a minor role because of sediment resuspension [4]. In contrast, in the Baltic Sea, CDOM strongly governs light attenuation in some regions, with varying contributions from Chla and inorganic suspended solids depending on the environmental conditions [3]. These examples underline the need for localized studies to understand the interactions among TSS, Chla, and CDOM in modulating light dynamics.

In this context, the Pampulha reservoir, located in southeastern Brazil, exemplifies the challenges faced by urban eutrophic water bodies. Rapid urbanization has led to significant ecological challenges, including nutrient overloading, sedimentation, and frequent cyanobacterial blooms, which severely affect water quality and ecosystem health [5]. These conditions directly affect light penetration, altering primary productivity and biogeochemical cycles. Light attenuation not only affects primary production but also profoundly alters the structure and functioning of aquatic ecosystems, impacting zooplankton, benthic organisms, and trophic dynamics. Reduced light can modify zooplankton migration patterns, hinder visual predation by fish, and favor phytoplankton species adapted to low irradiance, such as filamentous cyanobacteria, resulting in lower diversity and changes in ecological stability [6,7]. In benthic environments, lower light penetration decreases the production of epipelic algae, affecting energy availability for macroinvertebrates and compromising nutrient cycling and the base of the food web [8,9].

While previous studies have examined seasonal changes in water quality parameters and nutrient dynamics within this reservoir [10], limited attention has been paid to understanding the contributions of optically active substances, such as TSS, Chla, and CDOM, to light attenuation under varying seasonal conditions. The optical properties of the Pampulha reservoir were investigated, and its dynamics were elucidated [11]. However, their study did not explore the intricate interactions between these substances in modulating light attenuation, which is the core focus of this research.

This study aimed to investigate the contributions of TSS, Chla, and CDOM to light attenuation in the Pampulha reservoir under different seasonal conditions. Specifically, we addressed the following research questions: (1) What are the primary factors affecting light attenuation in the Pampulha reservoir? (2) How do seasonal variations, particularly differences in rainfall, influence the attenuation coefficients of PAR, UVA, and UVB radiation? By addressing these questions, this study seeks to advance the understanding of light attenuation dynamics in urban eutrophic reservoirs and to provide a scientific basis for targeted interventions that promote ecosystem resilience and sustainable water resource management.

2. Materials and Methods

2.1. Study Area

The Pampulha reservoir (WGS84 19°51′09″ S, 43°58′42″ W) (Figure 1), located in the city of Belo Horizonte in southeastern Brazil, is part of the “Pampulha Architectonic Complex”, recognized by UNESCO in 2016 as a Cultural Heritage of Humanity. This tropical, warm, monomictic reservoir exhibits thermal stratification between October and April during the rainy season and complete water mixing between May and September in the dry season [12]. It has a total area of 2.7 km², an original perimeter of 21 km, and a maximum depth of 14 m [11]. The reservoir drains a catchment area of 98 km² and has eight tributaries: (1) Olhos d’Agua, (2) “AABB″ creek, (3) Braúnas, (4) Bom Jesus, (5) Tijuco, (6) Mergulhão, (7) Sarandi, and (8) Ressaca; the last two are the major contributors to external nutrient and sediment input [13]. The catchment area covers two cities, Belo Horizonte and Contagem; the main land cover in both cities is residential and industrial urban area, which corresponds to more than 70% of the total coverage [14].

The reservoir was initially constructed in 1938 to supply water and was reinforced in 1958 following dam failure. Since the 1970s, the watershed has suffered from rapid and unplanned urbanization. The domestic and industrial pollution that the reservoir receives triggers severe eutrophication and sedimentation issues, with frequent algal blooms leading to a rapid decline in water quality [13,15]. Since the 1980s, due to its hypereutrophic conditions and recurrent cyanobacterial blooms [16], the reservoir has ceased to function as a water source. Although it contributes to flood control and serves as a tourist attraction, direct contact with the water through aquatic sports or navigation is prohibited.

For this study, three distinct sampling sites were chosen, Barragem (WGS84 19°50′50.56″ S, 43°58′7.28″ W), Igrejinha (WGS84 19°51′25.24″ S, 43°58′43.94″ W), and Ilha (WGS84 19°50′56.65″ S, 43°59′19.08″ W), in the Pampulha reservoir (Figure 1). The Barragem site is the closest to the dam, is the deepest, and does not have any close tributaries. The Igrejinha site has a medium depth and suffers influence from the tributaries Tijuco and Mergulhão. The first tributary is covered by a cellular concrete gallery and receives various forms of contamination; the second one is covered by a concrete gallery in part of its length has has moderate urbanization, and riparian vegetation is partially present. Despite the water quality of those tributaries, both present a low outflow, even more in the dry period. The Ilha site is shallower and has a direct influence from the tributaries Ressaca and Sarandi, both of which present the highest flows and drainage area and the poorest water quality, with high nutrients and DOC concentrations. The contamination of these two steams is due to industrial and domestic discharges, garbage, and soil erosion [10,12,17].

The region’s climate is classified as tropical highland, characterized by two distinct seasons: a warm and rainy season between October and March, which accounts for 90% of the annual rainfall (averaging 1500 mm), and a cooler dry season from April to September. Air temperatures exhibit relatively low variations throughout the year, with a minimum monthly mean of 18 °C in July and a maximum monthly mean of 23 °C in February [18].

2.2. Data Collection

Sampling was conducted monthly from September 2022 to October 2024 in the Pampulha reservoir (Figure 1). For each site, vertical profiles of photosynthetically active radiation (PAR) and ultraviolet radiation (UV) were obtained between 10:00 and 14:00 using a radiometer (Biospherical Instruments, San Diego, CA, USA). Profiles were made on the sunside of the boat to avoid shadows, and data collection was not performed on heavily overcast or rainy days to ensure consistent light conditions. The radiometer was attached to a notebook with the software LoggerLightTM (Biospherical Instruments, San Diego, CA, USA) while collecting radiation data PAR and UV (305 and 340 nm) and depth during the descent and ascent of the probe. Meanwhile, water samples were collected with 5 L prerinsed (acid-cleaned) plastic bottles at depths of 0.5 m to measure the absorption coefficients and optically active substance concentrations. The samples were then stored in dark, temperature-controlled conditions and transferred to the laboratory for further analysis.

2.3. Calculation of the Optical Properties

The diffuse attenuation coefficients for photosynthetically active radiation (PAR) and UV (Kd_PAR, Kd_UVA, and Kd_UVB) were calculated based on the exponential light attenuation law [19]:

$$Kd\left(\lambda \right)={\displaystyle \frac{-\mathrm{l}\mathrm{n}\left(I_d/I_o\right)}{d}}$$

(1)

where I_d is the light intensity at depth d; I_o is the surface light intensity; and d is the corresponding depth. Kd(λ) values were considered when R² > 0.90.

The 1% attenuation depth (Z_1%) is inversely proportional to Kd(λ) as follows:

$$Z_{1\%}\left(\lambda \right)={\displaystyle \frac{4.605}{{Kd}_{\left(\lambda \right)}}}$$

(2)

In this study, Z_1%(305) and Z_1%(340) represent the 1% attenuation depths for the UVB and UVA wavelengths, respectively. The 1% attenuation depth for PAR was used as a proxy for the euphotic depth.

2.4. Measurements of Optically Active Substances Concentration

Water samples were filtered using GF/F filters 0.7 μm porosity (Whatman, Amersham, UK) for Chla and TSS, and the filters were frozen until analysis. Water samples were also filtered for the analysis of dissolved organic carbon (DOC) and colored dissolved organic matter (CDOM) (0.22 μm filter, Millipore, Burlington, NJ, USA) and stored in amber glass bottles (pre-washed with distilled water and hydrochloric acid 10%) at 4 °C in the dark. The Chla concentration corrected by pheophytin (Chl-a) was obtained by acetone extraction (90%), measured using a spectrophotometer UV–VIS 2600 (Shimadzu, Kyoto, Japan) at 665 and 750 nm and calculated according to APHA (1998). The DOC concentration was obtained by catalytic oxidation at high temperatures using a TOC VCPN Analyzer (Shimadzu, Kyoto, Japan). Total suspended solids (TSS) were determined using the gravimetric method, considering the difference between the dry weight of GF/F Millipore filters (Whatman, Amersham, UK) (105 °C, 2 h) before and after the filtration of water samples [20].

We measured the absorbance spectra of CDOM for each sample using a spectrophotometer (UV–VIS 2600, Shimadzu, Kyoto, Japan) between 250 and 750 nm (1 nm intervals), with a 5 cm quartz cuvette and Milli-Q water (Millipore, Burlington, NJ, USA) as a blank. The absorbance data were used to calculate the absorption coefficients (α_CDOM(λ); m⁻¹) according to Equation (1):

$$a_{CDOM}\left(\lambda \right)=2.303A\left(\lambda \right)/r$$

(3)

where A is the absorbance measurement at wavelength λ and r is the cuvette path length [12]. The absorption coefficients were corrected by subtracting the value of the coefficient at 700 nm to avoid scattering, refractive effects, and instrument baseline drift [21]. Here, the spectrophotometric absorption coefficient at 254 nm (CDOM_254nm) was chosen as an indicator of CDOM concentration.

2.5. Statistical Analysis

Statistical analyses were conducted using R software 4.4.3 [22] to evaluate the variation in diffuse light attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB) and their modulators. The normality of the data was assessed using the Shapiro–Wilk test, which revealed non-normal distributions. Consequently, non-parametric methods were employed, including the Kruskal–Wallis test to compare sampling points and Spearman’s rank correlation to examine the relationships between Kd coefficients and optically active substances (TSS, Chla, DOC, and CDOM at 254 nm).

Stepwise multiple regression was employed to identify significant predictors of Kd_PAR, Kd_UVA, and Kd_UVB. Independent variables were iteratively included or excluded based on p-values (threshold: 0.05). The assumptions of linear regression were examined, including assessments of linearity, normality of residuals, and homoscedasticity. In cases of heteroscedasticity or autocorrelation, Newey–West (HAC) robust standard errors were applied to ensure reliability. Only the statistically significant predictors (p < 0.05) were retained in the final model.

3. Results

3.1. Descriptive Statistics

The summary statistics of water quality and light attenuation parameters for the three sampling sites—Barragem, Igrejinha, and Ilha—during the study period from September 2022 to October 2024 are presented in

Table 1. Summary statistics of water quality and light attenuation parameters measured at three points in the Pampulha Reservoir (September 2022–October 2024), including mean, median, minimum, maximum, and standard deviation (SD) for Chla, TSS, DOC, CDOM_254nm, Z_1% PAR, Z_1% UVA, Z_1% UVB, Secchi depth, Kd_PAR, Kd_UVB, and Kd_UVA.

Parameters	Min	Median	Mean	Max	SD
Chla (mg/m3)	1.28	48.11	57.07	177.06	41.51
TSS (mg/L)	0.40	19.33	24.12	105.00	16.64
DOC (mg/L)	2.34	19.62	21.73	90.11	15.10
CDOM254nm (m−1)	11.15	15.13	16.62	30.76	4.15
Z1% PAR	0.60	1.90	1.97	5.48	0.95
Z1% UVA	0.18	0.53	0.52	0.94	0.16
Z1% UVB	0.18	0.45	0.42	0.88	0.15
Secchi (m)	0.20	0.58	0.60	1.40	0.29
KdPAR (m−1)	0.84	2.42	3.03	7.67	1.75
KdUVB (m−1)	4.88	8.77	9.97	26.09	3.90
KdUVA (m−1)	5.22	10.16	12.36	26.07	4.84

. The data demonstrate considerable variability in parameters, such as chlorophyll-a (Chla), total suspended solids (TSS), dissolved organic carbon (DOC), and light attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB).

Chlorophyll-a concentrations ranged from 1.28 to 177.06 mg/m³, with a mean of 57.07 mg/m³, reflecting significant fluctuations in phytoplankton biomass, which highlights dynamic ecological conditions in the reservoir. TSS values ranged from 0.40 to 105.00 mg/L (mean of 24.12 mg/L), indicating substantial variability in the particulate load. DOC concentrations varied between 2.34 and 90.11 mg/L (mean of 21.73 mg/L), pointing to differing levels of organic matter across the sampling points and over time.

Light attenuation coefficients also exhibited significant variability: Kd_PAR values ranged from 0.84 to 7.67 m⁻¹, Kd_UVA values ranged from 4.88 to 26.09 m⁻¹, and Kd_UVB values ranged from 5.22 to 26.07 m⁻¹. These differences indicate variations in how photosynthetically active and ultraviolet radiation penetrate the water column, affecting photosynthesis and primary production processes.

The 1% light attenuation depth (Z_1%) was used as a proxy for the euphotic zone, indicating the depth to which sufficient light penetrated to support photosynthesis. Z_1% PAR values ranged from 0.60 to 5.48 m (mean of 1.97 m), whereas Z_1% UVA and Z_1% UVB were significantly shallower, underscoring the reduced penetration capacity for UV radiation (see Table 1).

3.2. Temporal Variability of Light Attenuation and Rainfall Influence

The temporal variability of the light attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB) and their correlation with rainfall patterns from September 2022 to October 2024 are shown in Figure 2, Figure 3 and Figure 4. These data highlight the significant impact of seasonal changes in hydrological conditions on the underwater light dynamics in the Pampulha reservoir.

During the dry season, the attenuation coefficients reached their lowest values, indicating clear water conditions and greater light penetration. This pattern aligns with the reduced input of suspended solids and dissolved organic matter. Conversely, the rainy season exhibited a marked increase in Kd values across all wavelengths, which was attributed to increased turbidity and organic matter influx, mainly from surface runoff. These seasonal dynamics directly influence the depth of the euphotic zone, with Z_1% for PAR, UVA, and UVB exhibiting substantial reductions during periods of heavy rainfall.

3.3. Spatial Variability Among Sampling Sites

The Kruskal–Wallis test revealed significant spatial differences in light attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB) among Barragem, Igrejinha, and Ilha (Kd_PAR: H = 67.58, p < 0.001; Kd_UVA: H = 53.28, p < 0.001; Kd_UVB: H = 40.13, p < 0.001). Ilha consistently exhibited the highest attenuation values, likely due to elevated levels of suspended and dissolved organic matter exacerbated by its geographic positioning in the mouth of the main tributary and its shallower depth. In contrast, Barragem and Igrejinha demonstrated comparatively lower values, reflecting their greater depths and locations that are more distant from the main tributary.

3.4. Influence of Rainfall on Light Attenuation

The attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB) exhibited significant spatial differences among the three sampling sites during both the low and high rainfall periods, as confirmed by the Kruskal–Wallis test (

Table 2. Results of the Kruskal–Wallis test for light attenuation coefficients (Kd_PAR, Kd_UVA, and Kd_UVB) across sampling points Barragem, Igrejinha, and Ilha during low and high rainfall periods. The H statistic of the Kruskal–Walis H-test and p-values indicate statistically significant differences (p < 0.05) for all parameters, highlighting the influence of rainfall on light attenuation.

Sampling Point	Parameter	High Rainfall Mean	Low Rainfall Mean	H Statistic	p-Value
Barragem	KdPAR	2.4	1.8	10.5	0.001
	KdUVA	9.1	7.3	12.3	0.002
	KdUVB	12.0	9.4	15.8	0.003
Igrejinha	KdPAR	3.1	2.5	10.5	0.001
	KdUVA	10.8	8.7	12.3	0.002
	KdUVB	11.9	10.3	15.8	0.003
Ilha	KdPAR	6.0	6.6	10.5	0.001
	KdUVA	17.9	12.9	12.3	0.002
	KdUVB	19.0	18.9	15.8	0.003

). During periods of high rainfall, the mean Kd_PAR, Kd_UVA, and Kd_UVB values increased substantially across all sites. For instance, during high rainfall periods, the mean Kd_PAR at Barragem increased from 1.77 m⁻¹ to 2.43 m⁻¹, while at Igrejinha, it rose from 2.46 m⁻¹ to 3.07 m⁻¹. Ilha showed even more pronounced differences, with Kd_PAR remaining elevated across all conditions. Similar trends were observed for Kd_UVA and Kd_UVB, reinforcing the dominant role of rainfall-driven inputs in reducing the water transparency.

Figure 5 illustrates these spatial and temporal variations using box plots, highlighting the distribution of Kd values at the three sites under low and high rainfall conditions. These findings underscore the combined influence of localized environmental characteristics and rainfall-driven turbidity on light attenuation.

3.5. The Relationships Between Kd(λ) and Optically Active Substances

Spearman’s rank correlation analysis revealed significant associations between light attenuation coefficients and optically active substances (

Table 3. Spearman’s rank correlation coefficient analysis to examine relationships between Kd coefficients and optically active substances (TSS, Chla, DOC, and CDOM_254nm). * Indicate statistically significant differences (p < 0.05).

	TSS (mg/L)	Chla (mg/m3)	DOC (mg/L)	CDOM254nm
KdPAR (m−1)	0.66 *	0.50 *	0.01	0.44 *
KdUVA (m−1)	0.58 *	0.38 *	0.10	0.59 *
KdUVB (m−1)	0.65 *	0.39 *	0.02	0.66 *

). TSS emerged as the strongest predictor for Kd values across all wavelengths, with correlation coefficients of R² = 0.66 for Kd_PAR, R² = 0.65 for Kd_UVA, and R² = 0.58 for Kd_UVB. CDOM also played a pivotal role, particularly for UV radiation, with R² = 0.66 for Kd_UVA and R² = 0.59 for Kd_UVB. Chlorophyll-a (Chla) showed a moderate correlation with Kd_PAR (R² = 0.50) and weaker relationships with the UV coefficients, indicating its primary influence on PAR attenuation.

In contrast, DOC exhibited negligible correlations with any Kd value, suggesting that it did not significantly influence light attenuation under the study conditions. These results highlight the dominant contributions of TSS and CDOM to underwater light attenuation dynamics, with Chla playing a secondary role.

Multiple regression models confirmed these findings, with equations demonstrating high predictive power (R² > 0.90):

• Kd_PAR = 0.0852 × TSS + 0.0085 × Chla + 0.0508 × CDOM_254nm (R² = 0.91, p < 0.05), indicating that TSS is the primary driver of PAR attenuation, followed by CDOM and Chla.
• Kd_UVA = 0.1478 × TSS + 0.3872 × CDOM_254nm (R² = 0.94, p < 0.05): CDOM was the dominant factor, absorbing UVA radiation, with TSS as a secondary contributor.
• Kd_UVB = 0.1688 × TSS + 0.4983 × CDOM_254nm (R² = 0.94, p < 0.05), and CDOM dominated UVB attenuation, with a significant input from TSS.

Each coefficient explains the factor of light attenuation in water, measured in m⁻¹, caused by its respective parameters of TSS, Chla, and CDOM. These models collectively reinforce the critical influence of suspended particles and CDOM on light attenuation dynamics in the reservoir.

4. Discussion

This study provides an analysis of the optical properties of the highly eutrophic Lagoa da Pampulha reservoir, emphasizing the significant role of optically active substances (OAS) in modulating light attenuation. The OAS include components such as dissolved organic matter, phytoplankton, and suspended particles, which absorb and scatter light, influencing water transparency. These findings underscore the complex interplay of physical, biological, and anthropogenic factors that influence underwater light dynamics, which vary significantly across temporal and spatial scales [23,24].

4.1. Key Drivers of Light Attenuation

The analysis highlights that total suspended solids (TSS) and colored dissolved organic matter (CDOM) are the dominant contributors to light attenuation, whereas chlorophyll-a (Chla) plays a secondary role. Statistical analyses revealed strong correlations between TSS and Kd_PAR (R² = 0.66), indicating a substantial impact on photosynthetically active radiation (PAR) scattering. This is consistent with the high sediment inputs during rainy periods, driven by surface runoff and sediment resuspension, which amplify turbidity. Sediment resuspension is particularly relevant in shallow areas, as indicated by the consistently high Kd values at Ilha, where localized geomorphological and hydrodynamic conditions likely promote sediment deposition.

CDOM strongly influenced UV light attenuation, with R² = 0.66 for Kd_UVA and R² = 0.59 for Kd_UVB. This highlights its capacity to absorb UV radiation stemming from terrestrial organic matter inputs, which are intensified during rainfall events. Similarly, studies in the Sau Reservoir (Spain) and the Baltic Sea [3] have also demonstrated the strong influence of terrestrial CDOM on light absorption. However, unlike these temperate systems, our findings suggest that, in a tropical urban reservoir, the combination of high sediment loads and anthropogenic influences results in a distinct optical environment where CDOM and TSS exert a more pronounced impact on light attenuation. In addition to light attenuation, CDOM plays a key ecological role by modulating UV radiation exposure, which affects microbial activity and photochemical processes [2].

Chlorophyll-a exhibited moderate correlations with Kd_PAR (R² = 0.50) and negligible effects on UV attenuation, suggesting that its influence was primarily confined to the PAR spectrum. The modest contribution of Chla reflects the secondary role of phytoplankton compared with particulate and dissolved organic matter in this eutrophic system. However, episodic algal blooms during periods of peak productivity may transiently increase their influence, further reducing light penetration.

The multivariate regression models reinforced these findings. The equation for Kd_PAR = 0.0852 × TSS + 0.0085 × Chla + 0.0508 × CDOM_254nm underscores the dominance of TSS, followed by CDOM and Chla, in PAR attenuation. For UV radiation, CDOM emerged as the primary contributor, as demonstrated by the equations for Kd_UVA (R² = 0.94) and Kd_UVB (R² = 0.94), with TSS acting as the secondary driver.

4.2. Temporal and Spatial Variability

Seasonal hydrological changes significantly influence the optical properties of reservoirs. During the rainy season, increased surface runoff elevates total suspended solids (TSS) and colored dissolved organic matter (CDOM) levels, leading to higher light attenuation coefficients and diminished water clarity. For instance, at Barragem, the Kd_PAR rose from an average of 1.77 m⁻¹ during low rainfall to 2.43 m⁻¹ during high rainfall, with similar patterns observed for Kd_UVA and Kd_UVB. These results underscore the reservoir’s sensitivity to precipitation-driven inputs, which enhance turbidity and organic matter loads. In contrast to temperate reservoirs—such as the Sau Reservoir, where seasonal stratification primarily governs light attenuation—our findings demonstrate that sediment resuspension and runoff-induced CDOM inputs are the dominant factors shaping optical properties in this system. This distinction highlights the need to consider local hydrodynamic and climatological conditions when assessing light dynamics in eutrophic environments.

Spatially, Ilha consistently exhibited the highest Kd values across all wavelengths, reflecting its exposure to localized sedimentation and runoff inputs. In contrast, Barragem and Igrejinha showed comparatively lower attenuation coefficients, suggesting a less direct influence from runoff and sediment resuspension. These spatial variations emphasize the need for localized management strategies that address site-specific hydrological and ecological conditions.

4.3. Ecological Implications

The attenuation of photosynthetically active radiation (PAR) and ultraviolet (UV) light triggers cascading effects throughout the reservoir’s ecosystem. As light penetration decreases, the euphotic zone becomes shallower, which directly limits primary productivity and reshapes the spatial distribution of phytoplankton and submerged macrophytes. Moreover, reduced light availability alters zooplankton behaviors—such as diel vertical migration and predator–prey interactions—and influences benthic community composition by affecting periphytic algae, a key energy source in the system. In Lagoa da Pampulha, these light conditions likely drive shifts in phytoplankton composition, favoring bloom-forming cyanobacteria, which further exacerbate light limitation and promote hypoxic conditions during decomposition.

The UV attenuation by CDOM has additional ecological consequences. Although UV radiation can inhibit microbial activity and phytoplankton growth, excessive attenuation may reduce the photochemical degradation of organic pollutants, altering the biogeochemical cycles of the reservoir. This dynamic interplay between light attenuation and ecological processes underscores the need for an integrated understanding of these factors to predict ecosystem responses to changing environmental conditions.

4.4. Management and Monitoring Implications

The findings of this study underscore the critical influence of total suspended solids (TSS) on light attenuation in Lagoa da Pampulha, highlighting the importance of managing sediment inputs to improve water quality and underwater light conditions. In an urbanized watershed, where tributaries are canalized and subject to runoff from dense residential areas, sediment dynamics play a key role in shaping the reservoir’s optical environment.

The effective management of sediment inputs requires a comprehensive understanding of the interactions between hydrological processes, urban infrastructure, and anthropogenic impacts. Addressing these challenges involves integrating sediment dynamics into broader watershed management strategies, particularly in tributaries such as Ressaca and Sarandi, which contribute disproportionately to the reservoir sediment load.

Given the urbanized nature of the basin, monitoring programs should prioritize the assessment of sediment transport and deposition patterns, focusing on areas with the highest inputs. These efforts can provide a robust basis for identifying critical sources of sediment and evaluating the effectiveness of interventions aimed at reducing sedimentation and improving water quality.

The results also emphasize the need for ongoing monitoring of optically active substances, including TSS, to support the development of bio-optical models tailored for urban reservoirs. Such models can improve the accuracy of water quality assessments and inform management decisions, particularly for systems with complex sediment and nutrient dynamics.

The unique challenges posed by the Lagoa da Pampulha urban watershed require innovative approaches that align with its realities: canalized tributaries, dense urbanization, and a limited but persistent issue of clandestine sewage discharge. Sediment mitigation efforts should focus on integrating stormwater management with sediment control, improving the urban infrastructure, and fostering community participation. These measures combined with targeted tributary interventions can enhance light penetration and water quality in the reservoir while addressing the root causes of sediment and pollutant inputs.

4.5. Broader Relevance and Future Research Directions

In contrast to previous studies on temperate reservoirs that primarily emphasize seasonal stratification, our research provides novel insights into the light attenuation dynamics of a tropical urban reservoir. In this system, sediment resuspension and anthropogenic inputs are the dominant factors influencing optical properties. By integrating bio-optical modeling with hydrodynamic variability, our work advances the methodological framework for understanding urban aquatic ecosystems, underscores the unique challenges faced by tropical reservoirs—characterized by high sediment loads, nutrient enrichment, and substantial human pressures—that extend beyond Lagoa da Pampulha, and offers valuable perspectives on eutrophic systems worldwide [24,25].

Effective management of urban eutrophic reservoirs, such as Pampulha, requires integrated strategies that address both pollution sources and hydrological processes. Implementing watershed management practices to reduce sediment and nutrient loads, alongside urban infrastructure improvements, is critical [13]. Moreover, establishing buffer zones with native vegetation can effectively filter sediments and nutrients, thereby enhancing water clarity and mitigating eutrophication [18].

In addition, incorporating light attenuation data into bio-optical models improves predictions of euphotic zone depth and primary productivity, supporting real-time decision-making. When coupled with early warning systems, these models can be instrumental in preventing harmful algal blooms [2,3].

Future research should investigate the interactive effects of light attenuation and nutrient dynamics on phytoplankton composition; assess the impacts of climate change on rainfall patterns, sediment transport, and CDOM dynamics in tropical reservoirs; and develop integrated hydrological and bio-optical models to simulate ecosystem responses under various management scenarios. Building on our findings, further studies should also explore how urbanization-driven hydrological modifications influence light regimes—a factor that has received limited attention in research on temperate and subtropical reservoirs.

5. Conclusions

This study provides a comprehensive analysis of the factors driving light attenuation in Lagoa da Pampulha, a eutrophic urban reservoir significantly influenced by seasonal and spatial variations in optically active substances (OAS). The results underline the dominant roles of total suspended solids (TSS) and colored dissolved organic matter (CDOM) in modulating light attenuation, particularly during the rainy season, when inputs from runoff are most pronounced. Chlorophyll-a (Chla), which is important for photosynthetically active radiation (PAR) attenuation, plays a secondary role compared with TSS and CDOM.

The significant correlations and predictive regression models developed in this study reinforce the importance of TSS and CDOM as key modulators of underwater light dynamics. These findings highlight the sensitivity of reservoirs to hydrological processes, emphasizing the critical impact of sediment and organic matter inputs on light availability, primary productivity, and ecological processes.

Given the urbanized nature of the Pampulha watershed, this study underscores the need for continuous monitoring of OAS and their contribution to light attenuation. This is particularly relevant for the development of bio-optical models tailored to urban reservoirs that can enhance the accuracy of water quality assessments and support evidence-based management strategies.

By advancing the understanding of the interplay between OAS and light dynamics, this study provides a scientific basis for interventions aimed at mitigating sedimentation and organic matter input. These efforts are essential for improving water clarity, supporting ecosystem resilience, and addressing the broader challenges faced by urban reservoirs worldwide.

Future research should explore the interactive effects of light attenuation and nutrient dynamics on hydrobiological community composition as well as the potential implications of climate change on hydrological processes and OAS dynamics in urban aquatic systems.