1. Introduction
Treatment wetlands (TW) have proven to be a suitable alternative for domestic wastewater treatment [
1,
2,
3] and have been successfully implemented for wastewater treatment in various countries around the world [
4,
5,
6,
7,
8,
9,
10,
11,
12,
13]. In many of these countries, wastewater management systems are carefully planned and regulated. Although TW systems have low construction and operating costs, they are easy to construct and maintain, they contribute to biodiversity restoration and provide ecosystem services to the public when designed for this purpose [
14].
Despite these advantages, the adoption of TW in developing countries remains limited, even though many of these regions have significant wastewater treatment deficiencies. In Ibero-America and Caribbean, 36 million people in urban areas and 31 million in rural regions lack access to sanitation services [
15]. Although government efforts have been substantial, they have not been sufficient to close the gap compared to high-income countries. Furthermore, the main factors limiting the use of TW are the lack of knowledge about the technology, the land requirements for its implementation, the low esthetic value of vegetation and its large-scale costs in urban and interurban areas [
16].
Vymazal [
17] has highlighted that in recent years there has been a growing trend toward short-term, laboratory-scale studies on treatment wetlands (TWs), often utilizing synthetic wastewater that is easily degradable and not representative of real wastewater. This approach results in the development of a bacterial community that is significantly different from that found in full-scale TWs. Moreover, due to the limited duration of these studies, nutrient uptake is often overestimated, as plants in their early growth stages require higher nutrient uptake. As a result, these studies cannot be considered as a basis for developing design, construction, maintenance, and operational criteria for large-scale TWs, which must address real-world wastewater treatment challenges. Moreover, this situation is more complicated in tropical areas, since most of the research on TW has been carried out in temperate climates and information in tropical climates is limited [
18]. Ibero-American and Caribbean countries have major problems in terms of wastewater treatment, especially developing countries such as Mexico and Colombia, where the development of water treatment is slow due to the costs generated by decentralized management, given the territorial dispersion of the population throughout the length and breadth of their territories.
Tropical climates introduce additional challenges to the performance of wastewater treatment systems due to high temperatures, significant precipitation, and elevated humidity levels. These climatic factors significantly influence the biogeochemical processes within wastewater treatment systems, affecting microbial activity, plant growth, and pollutant removal efficiency [
19]. High temperatures can accelerate the decomposition of organic matter and nitrogen transformations, while intense rainfall may lead to hydraulic overload, potentially reducing treatment efficiency [
20,
21,
22]. In addition, the vigorous plant growth stimulated by tropical conditions may require more frequent maintenance to prevent clogging and ensure optimal system performance. The lack of large-scale applications and real-world monitoring highlights the need for further research to address these challenges and refine the design and operational strategies of wastewater treatment systems in tropical environments [
23]. A recent trend in the design of wastewater treatment systems is the incorporation of ornamental plants, which enhances the esthetic appeal of these systems and makes them more acceptable in urban and peri-urban areas [
24]. The use of ornamental plants not only contributes to landscape integration but also fosters public acceptance and promotes biodiversity. However, most studies involving ornamental plants in wastewater treatment systems have been conducted under control conditions, such as microcosms or mesocosms. Their adaptability to real-world climatic conditions, especially in tropical climates remains largely unexplored due to the absence of large-scale studies [
25].
In this study, the pollution removal capacity and vegetation development were evaluated under real operating conditions of a large-scale turtle-shaped treatment wetland with polycultures of ornamental plants for the treatment of municipal wastewater in a tropical climate. This research is crucial for decision-makers and the scientific community, as it provides essential information for adapting the design and operational criteria of such systems to effectively control pollution in estuaries of the Gulf of Mexico. In addition, this study contributes to bridging the knowledge gap regarding the implementation of treatment wetlands in tropical regions by providing insights into how climate-related variables influence their performance and demonstrating the potential benefits of integrating ornamental plants for urban wastewater management.
3. Results
3.1. Results of Environmental Parameters
During the treatment period, the ambient temperature (
Figure 3a) varied from 13.5 to 33 °C, with the maximum temperature in May (33 °C). This temperature range is considered adequate since at temperatures below 20 °C there is a decrease in microbial activity involved in pollutant removal [
35]. The average temperature documented for the 12 months of monitoring was 23.28 °C, which is within the optimal range for vegetation development in this type of subtropical climate, as reported by Majumder [
36], which is between 12 and 25 °C. The average light intensity was 42,993.75 Lx (
Figure 3b), which was within the optimal range (40,000–60,000 Lx) for the ornamental plants [
36]. On the other hand, the relative humidity (
Figure 3c), values over the entire monitoring cycle range from 72 to 100% with an average of 86.3%. The precipitation (
Figure 3d) over the monitoring cycle ranged from 17 to 96 mm; the months with the highest precipitation were August and September; the average precipitation over the monitoring cycle was 51.83 mm.
3.2. Results of Water Quality Parameters Measured in the Field
pH values were within the limits between 6 and 9, which had an appropriate pH range for treatment of wastewater [
37].
Figure 4 shows the development of pH over the 12 months of monitoring the system with an average of 7.21. The values ranged from 6.95 to 7.6, which is within the mentioned limits, and these are common pH values for wastewater (
Figure 4), being within the typical values for domestic wastewater [
38]. In general, values close to neutrality were found during the 12 months duration of the study.
The pH distribution across the different zones exhibited significant differences (p < 0.05). The sedimentation zone revealed notable disparities in pH behavior in comparison to Zone B (p < 0.05) Zone C (p < 0.05) and Zone D (p < 0.05). Furthermore, Zone A exhibited significant differences when compared to Zone C (p < 0.05) and Zone D (p < 0.05).
The correlation analysis of pH concerning environmental variables (light intensity, ambient temperature, relative humidity, and precipitation) commenced with a normality test, which indicated that the data followed a normal distribution. Consequently, the correlation was determined using Pearson’s coefficient. The results indicated no correlation between pH and light intensity and between pH and ambient temperature (p > 0.05). However, a significant inverse correlation was observed between pH and relative humidity (p < 0.05), with a Pearson coefficient of −0.6158 and a confidence interval ranging from −0.8791 to −0.06486, indicating a moderate to strong association. Similarly, the relationship between pH and precipitation demonstrated an analogous inverse correlation of moderate to strong strength (p < 0.05) with a Pearson coefficient of −0.6371 and a confidence interval from −0.8868 to −0.09958.
The next field parameters: DO, water temperature, EC and TDS are shown in
Table 2. The TDS ranged from 159.84 ± 1.76 and 171.41 ± 2.00 mg L
−1, values which were never above the TDS permissive limit (1000 mg L
−1) indicated in the Mexican standard NOM-127-SSA1-2021.
The average DO in the areas, in general, varied from 2.56 ± 0.31 to 3.53 ± 0.29 mg L−1. The highest DO concentration was found in the influent zone and gradually decreased until reaching Zone D where the effluent was located, indicating a 27.5% decrease in DO concentration at the end of the treatment. The temperature remained relatively constant in all zones of the WT, ranging from 24.4 to 25.4 °C.
The DO exhibited significant differences among the different zones (p < 0.05), with the only exceptions being the comparisons from the Settler to Zone A and from Zone D to Zone C, which did not show significant differences (p > 0.99). All other comparisons between zones showed strong significant differences (p < 0.0001). The EC showed differences, especially between the settler and Zone B (p < 0.05) and between the settler and Zone D (p < 0.05). The TDS did not show significant differences between the zones (p > 0.05). The water temperature variation between the zones was statistically significant, specially from the Settler to Zone B (p < 0.0001), from the Settler to Zone D (p < 0.0001), between Zone A and Zone D (p < 0.05) and between Zone C and Zone D (p < 0.05).
Regarding the correlation of DO, EC, and TDS with environmental variables, no significant correlations were identified (p > 0.05) with any variable. Conversely, water temperature did show correlation with ambient temperature and relative humidity (p < 0.05) However, the data did not follow a normal distribution; hence, the Spearman correlation coefficient was employed. The Spearman correlation coefficient of ambient temperature was 0.6480, with a confidence interval ranging from 0.6166 to 0.8946, indicating a direct correlation with a moderate to strong association. In contrast, the correlation with relative humidity yielded a Spearman coefficient of −0.7273, with a confidence interval of −0.9210 to −0.2452, reflecting an inverse correlation with a moderate to strong association.
Overall, none of the parameters analyzed (pH, DO, EC, TDS, and water temperature) exhibited correlations with each other (p > 0.05).
3.3. Results of Parameters Measured in the Laboratory
Figure 5 and
Figure 6 show the average concentrations and removal efficiency of COD, NH
3-N, NH
4-N, NO
2-N, NO
3-N, TN, PO
4-P and TP in the TW system, which showed a gradual decrease in their concentrations throughout the process. The average COD concentration decreased considerably, from an initial concentration of 443.56 ± 14.47 to 29.39 ± 3.61 mg L
−1, obtaining an efficiency of 93.37%, which indicates a high reduction in organic matter (
Table 3).
The distribution of COD across the various zones exhibited significant differences (p < 0.05). The Settler showed notable differences compared to Zone B (p < 0.05), Zone C (p < 0.001) and Zone D (p < 0.001). Furthermore, Zone A demonstrated significant differences when compared to Zone C (p < 0.05) and Zone D (p < 0.05).
The TW system reduced the initial average NH
3-N concentration from 22.81 ± 0.9 to 1.97 ± 0.57 mg L
−1, which implies a removal efficiency of 91.36% (
Table 3). However, the average annual NH
3-N removal concentration was above the permissible limit of 0.50 mg L
−1 indicated by the Mexican official standard NOM-127-SSA1-2021 for NH
3-N. Meanwhile, the annual average concentration of NH
4-N decreased from 57.85 ± 1.61 to 5.04 ± 1.24 mg L
−1, obtaining a removal efficiency of 91.29%. However, there is no official Mexican standard for NH
4-N that indicates a permissible limit, so the international standard was considered, where the World Health Organization (WHO) indicates that in surface waters the maximum natural concentration of NH
4-N is 12 mg L
−1, which means that the TW system reduces it to acceptable levels.
The initial concentrations of NO
2-N (9.93 ± 0.63 mg L
−1) and NO
3-N (23.77 ± 1.34 mg L
−1) in the TW system were reduced to 0.45 ± 0.08 and 2.05 ± 0.60 mg L
−1, respectively. Both final concentrations are below the permissible limit indicated by the Mexican official standard NOM-127-SSA1-2021, 0.90 mg L
−1 for NO
2-N and 11.00 mg L
−1 for NO
3-N, indicating a removal efficiency of 95.74% for NO
2-N and 97.36% for NO
3-N (
Table 3).
On the other hand, TN had a removal efficiency of 91.37%, which reduced the 35.71 ± 0.87 to 3.08 ± 0.77 mg L
−1 of the annual average TN concentration, which is below the 15.00 mg L
−1 discharge limit for the protection of aquatic life, estuaries and urban public use stipulated according to the Mexican official standard [
33]. Likewise, the initial average concentrations of TP and PO
4-P were reduced from 16.73 ± 0.66 to 1.43 ± 0.36 mg L
−1 and from 19.13 ± 0.57 to 1.48 ± 0.64 mg L
−1, respectively. Obtaining a removal efficiency of 91.45% and 92.26% for TP and PO
4-P, respectively. In addition, the TP and PO
4-P concentrations were conformed with the permissible limit of 5.00 mg L
−1 that the Mexican official standard [
33] indicates for the security and protection of aquatic life in rivers, estuaries and for public use.
3.4. Seasonal Variability in Pollutant Removal
The statistical analysis revealed significant seasonal fluctuations (p < 0.05) in COD and nitrogenous contaminants, with distinct behaviors depending on the season. In summer, COD was lower, likely due to increased microbial activity, while in winter, lower temperatures favor more efficient removal of pollutants. Autumn showed stability in COD, with no significant differences (p > 0.05) compared to other seasons. Regarding nitrogen compounds, concentrations of NH3-N and NH4-N were higher in winter due to reduced microbial activity, while no significant differences were observed between the other seasons. NO2-N and NO3-N, levels remained stable throughout the year, reflecting a consistent removal process. TN concentrations were higher in winter, possibly due to environmental conditions that favor nitrogen accumulation.
Considering the tropical environment of the study area, as illustrated in
Figure 5 and
Figure 6, the concentrations of COD, NH
3-N, NH
4-N, NO
2-N, NO
3-N, TN, PO
4-P, and TP exhibited behavior influenced by the seasons. During the wet season (May–September), contaminant concentrations slightly increased, likely due to the increased hydraulic load caused by higher precipitation. This resulted in shorter hydraulic retention times, which primarily affected COD removal efficiency, as reflected by the comparatively higher effluent concentrations during these months. In contrast, during the dry season (October-April), lower precipitation and higher ambient temperatures likely enhanced microbial activity and nutrient absorption by vegetation, improving the removal of nitrogen compounds such as NH
3-N and NO
3-N. Phosphorus removal (PO
4-P, and TP) exhibited greater stability during this period, with effluent concentrations consistently below the maximum thresholds established by Mexican official standard [
33]. The results suggest that the tropical environment and seasonal variations play a key role in the efficiency of water treatment, significantly affecting contaminant removal.
The official Mexican standard aims to establish the permissible limits of pollutants in wastewater discharges. In order to ensure compliance with this standard, monthly monitoring was carried out for wastewater discharges to receive bodies such as rivers, streams, canals and drains. The reports on water sampling and analysis results were compared with the permissible limits (
Table 3) of the monthly average based on the effluent.
3.5. Correlations Between Pollutants vs. (Water Quality Parameters, Environmental Variables and the Time)
The correlation analysis between pollutant removal and environmental variables shows significant associations, highlighting the influence of seasonal changes on system performance. COD removal efficiency showed strong negative correlations with light intensity and ambient temperature, with Pearson coefficients ranging from −0.7716 to −0.8027, suggesting reduced microbial activity during the wet season. A positive correlation with DO emphasizes the importance of aerobic conditions for organic matter degradation.
Nitrogen species removal (NH3-N, NH4-N, NO2-N, NO3-N, and TN) exhibited similar trends, with strong inverse correlations to light intensity and water temperature and positive correlations to DO. Pearson coefficients for these correlations ranged from −0.6774 to −0.8457 (light intensity and water temperature) and 0.6617 to 0.7066 (DO), indicating that aerobic conditions and favorable pH levels are crucial for nitrification and nitrogen transformation processes. NO2-N removal also showed a strong inverse Spearman correlation with temperature (−0.6655), suggesting efficient conversion to nitrate under higher temperatures during the dry season.
Phosphorus removal (PO4-P and TP) demonstrated strong negative correlations with light intensity and precipitation, with Pearson coefficients ranging from −0.6574 to −0.6770, while positive correlations with DO and pH (0.6325 to 0.6975) highlighted the role of aerobic and pH-favorable conditions in phosphorus uptake and transformation. Overall, the findings underscore the importance of integrating environmental variables into the design and management of treatment wetlands to enhance pollutant removal under variable climatic conditions.
The correlation analysis indicates a significant decrease in contaminants over time, demonstrating an improvement in water quality within the treatment system. The reduction in organic matter (COD) is especially notable (r = −0.9598), facilitated by microbial activity under aerobic conditions. Likewise, nitrogen species (NH
3-N, NH
4-N, NO
2-N, NO
3-N, TN) exhibit effective removal (−0.66 to −0.97), influenced by temperature and dissolved oxygen levels. Regarding phosphorus (PO
4-P, TP), its reduction is moderate (−0.57 to −0.74), suggesting the influence of chemical precipitation and biological absorption, both favored by aerobic conditions and optimal pH levels (
Table 4).
3.6. Result of Vegetation Growth
The growth evaluation of the 13 plants utilized in the TW system was conducted by comparing plant growth, stem diameters, and the quantities of flowers, leaves, and stems (
Figure 7). Among the 13 plants used in the TW system,
C. papyrus was the largest, although it did not present the greatest growth during the study period. The plant that grew the most during the study was
C. hybrids, which increased to 15 times its size during the study, followed by
C. indica which increased to 12 times its initial size and
C. papyrus and
C. esculenta which increased to 4 and 3.7 times their initial sizes, respectively. Stem thickening presented more discrete increases than plant size, with
H. psittacorum being the plant that presented the greatest stem thickening, to 2.5 times its initial diameter.
C. papyrus showed a similar stem thickening, 2.2 times its initial diameter, and
C. esculenta had the third largest stem thickening, 1.5 times its initial diameter.
Among the 13 plants employed in the TW system only seven presented flowers, S. gramina, H. psittacorum, Ruellia sp., C. hybrids, C. indica, S. graminea and E. crassipes. Among seven species, only two species had >10 flowers, S. graminea and Ruellia sp., while the rest had 5–10 flowers per plant, except for S. graminea which presented <5 flowers. Also, only four species presented >60 leaves at the end of the study, C. papyrus the one with the highest number of leaves, followed by Berberis sp., Croposma sp. and Ruellia sp. Finally, C. papyrus was the plant with the highest number of stems, presenting >20 stems per plant at the end of the study, while P. cordata, Berberis sp., B. thunbergii and E. crassipes presented >10 stems and the rest of the species presented <10 stems per plant, except for Sansevieria trifasciata which remained with the same number of stems until the end of the study.
The correlation analysis of vegetation in relation to environmental variables demonstrated statistically significant differences (p < 0.05) concerning luminosity and ambient temperature. No correlation was observed for the environmental variables of relative humidity and precipitation. The correlations between vegetation and both light intensity and ambient temperature were all direct (positive) and showed moderate (0.3–0.5) to strong (0.5–1.0) associations for Sansevieria trifasciata, Canna indica, Canna hybrids, Colocasia esculenta, Ruellia sp. Sagittaria graminea, Pontederia cordata, and Heliconia psittacorum, Berberis sp., and Eichhornia crassipes and reflected significant associations. All correlations present confidence intervals that suggest a robust association.
The Pearson correlation analysis for various plant species (Sansevieria trifasciata, Canna indica, Canna hybrida, Colocasia esculenta, Ruellia sp., Sagittaria graminea, Pontederia cordata, Heliconia psittacorum Berberis sp., Eichhornia crassipes, and Cyperus papyrus) revealed significant negative correlations (p < 0.05) with most water quality parameters, especially nitrogenous compounds (NH3-N, NH4-N, TN) and COD. The negative correlations for these parameters were strong (r > −0.9) for several species, including Eichhornia crassipes, Canna indica, and Colocasia esculenta, indicating their effectiveness in reducing these pollutants. Moderate correlations (−0.6 and −0.7) were observed for NO2-N, NO3-N, and TP across most species. The correlation with PO4-P was generally weak or non-significant. Confidence intervals for most parameters suggested robust associations, with significant reductions in nitrogen compounds and COD. These results highlight the potential of these species in improving water quality especially through the removal of nitrogen and organic matter, while their effect on phosphorus removal was less consistent.
4. Discussions
In the present study, a comprehensive evaluation of the performance of treatment wetlands (TW) utilizing ornamental plants on a large scale for the treatment of municipal wastewater in a contaminated estuary of the Gulf of Mexico was conducted. The findings are promising and underscore the potential of treatment of wetlands as a viable and sustainable option in similar contexts.
The COD removal efficiency reached 93.37%. However, it is pertinent to compare these results with those from other studies. For instance, Rizzo et al. [
39] reported an average removal efficiency of 97.50%, utilizing a wetland with more than double the surface area of the one in this study. In contrast, Sandoval et al. [
16] documented a removal efficiency of 86.95% in a wetland in PastorÃas, Actopan, which treated 18.3 m
3/day of domestic wastewater. Similarly, a wetland in China, with almost three times the surface area of our TW, exhibited removal efficiencies of 22.29% [
40]. Expanding TWs for broader applications, especially in urban environments, presents challenges such as the need for significant land area, vegetation maintenance, and potential seasonal variations in plant growth.
The selection of ornamental plants may have contributed to the high removal efficiency in our study. In addition, the flowering ornamental plants in constructed wetlands, in monocultures or mixed species, enhance the esthetic appeal while treating wastewater [
41,
42].
Canna indica and
Cyperus papyrus have demonstrated robust growth and high nutrient absorption capacity under wastewater treatment conditions [
42,
43,
44,
45]. These plants also promote the development of microorganisms that facilitate nitrification and denitrification processes in the wetland substrate [
46,
47]. Furthermore, these findings agree with results observed in other tropical environments, where the selection of plant species and local climatic conditions play a significant role in determining the performance efficiency of constructed wetlands.
Furthermore, previous studies have shown that wetlands without vegetation, or with limited vegetative cover, exhibit significantly lower removal rates. For example, investigations in Nigeria revealed that non-vegetated systems achieved COD removals between 41.6% and 44.85% [
48].
The NH
4-N removal was also significant, with an efficiency of 91.29%, surpassing reports by Li et al. [
49] and Wang et al. [
50], which documented efficiencies of 44.3% and 28.41%, respectively. This enhancement in NH
4-N removal can be attributed to the synergistic interactions between the plants and microorganisms within the system, which are favored by local tropical climatic conditions that support more vigorous vegetative growth and increased microbial activity.
Furthermore, the study indicated that the system achieved removal efficiencies for NO
2-N and NO
3-N of 95.74% and 97.36%, respectively. These results are significantly higher than those reported by Li et al. [
49] in systems treating industrial wastewater, where efficiencies were 84.70% for NO
2-N and only 39.90% for NO
3-N.
The TN removal efficiency observed in our study exceeded the 42.9% and 52.48% reported by Li et al. [
49] and Wang et al. [
50], respectively. However, the TN removal efficiency was intermediate compared to the efficiencies reported by Zhu et al. [
51], which ranged from 35.4% to 81.3% over a decade of operation.
Regarding TP removal, an efficiency of 91.45% was achieved, which is comparable to other quality parameters. This is noteworthy, as other studies, despite reporting high efficiencies in COD, NO
2-N, and NO
3-N, showed TP removal efficiencies of less than 51% [
39,
49]. The capacity of the plants to absorb phosphorus, combined with retention in the substrate, contributes to improving this aspect of treatment.
The incorporation of ornamental plants in treatment wetlands enhances water quality while offering potential economic benefits. In tourism-dependent regions, esthetically appealing wetlands can increase landscape value and attract visitors, reinforcing their dual function as wastewater treatment systems and economic assets. Moreover, vegetation management in tropical climates presents challenges, as excessive plant growth positively influences system performance but necessitates regular pruning to maintain efficiency and prevent flow obstruction—a critical factor in pollutant removal.
The TWs represent a sustainable approach that can be implemented in other tropical regions with similar wastewater treatment challenges. To optimize wetland performance in these contexts, we recommend further research to establish reliable design parameters, assess long-term economic benefits, and address the scalability of the system for urban applications.
5. Conclusions
Most of the extant research on the use of large-scale constructed wetlands has been conducted in template climates, and there is a lack of information on the use of ornamental plants from the tropical and intertropical regions of Mexico. Warmer temperatures and sunlight intensity can provide favorable conditions for the development of ornamental plants, achieving significant nutrient uptake from municipal wastewater. In this study, the optimal plant species was identified as C. papyrus, which exhibited increased height and a substantial increase in the number of stems compared to other species.
This study underscores the potential for integrating ornamental and thematic landscape plants into treatment wetlands, promoting their adaptation to urban environments and fostering public acceptance. Beyond the enhancement of water quality, these systems can augment the esthetic value of urban landscapes, offering economic and social benefits, especially in regions where tourism is a primary economic driver.
Regarding the removal of pollutants, the system was found to be in accordance with the Mexican standard NOM-001-SEMARNAT-2021 for the evaluated parameters, achieving high removal efficiencies for COD (93.37%), TN (91.37%), and TP (91.45%) during the monitoring period. Nevertheless, it is imperative to continue monitoring over extended periods to refine design parameters and assess system performance under seasonal variations. The establishment of reliable operational guidelines, including optimized pruning and maintenance schedules, is imperative for ensuring the long-term effectiveness of large-scale ornamental treatment wetlands in tropical regions. The findings emphasize the importance of ornamental treatment wetlands for wastewater management in other tropical areas facing similar challenges, providing a foundation for future research and environmental policies.
