Community participation and technical design for wastewater management in Cassava Agro-Industry in Cauca, Colombia Open Access

Only 30% of the world's wastewater is treated (ONU-Habitat & OMS 2021). Wastewater generated by agro-industries has a significant global environmental, health, and economic impact (ONU-Habitat & OMS 2021). The discharge of pollutants, such as pesticides, fertilizers, and organic matter, decreases water quality to the detriment of the receiving waters (Aguiar Novillo et al. 2022). In Colombia, the agro-industry is key to the economy of the Cauca department. The Cauca River basin, which functions as a water regulator and a natural reservoir, is seriously polluted in some areas due to agro-industrial activities. In the north of the Cauca department, the most significant agro-industry is the production of sour cassava starch, commonly used in baking traditional products.

The production of sour cassava starch is a longstanding (over 100 years) family-led agro-industrial practice, and the processing factories are commonly referred to as ‘rallanderías’. They produce starch from cassava roots, generating solid and liquid residues at various process stages. Very briefly, the root is washed and peeled. Then, the cassava is grated, and the solid residues (afrecho) are separated from the water containing the starch. Finally, a sedimentation process occurs where the starch is recovered in sedimentation channels and separated from the ‘mancha’, a protein-rich by-product. Rallanderías use low to medium technology, mostly artisanal. The development and use of industrialized machinery have arisen mainly from empirical innovations made by the owners, employees, and regional technicians.

Cassava processing requires high-quality water. In the Cauca region, this water is obtained from the upper Cauca River basin, specifically from the sub-basins of the Quinamayó and Ovejas rivers, whose average flows are 2.1 and 6.3 m³/s, respectively (Ministerio de Ambiente y Desarrollo Sostenible 2020). This type of water is characterized by low levels of biological oxygen demand (BOD₅, <3 mg/L), chemical oxygen demand (COD, <10 mg/L), and total suspended solids (TSS, <5 mg/L) (IDEAM 2023). The rallanderías of Cauca pose important challenges for effective water management, not only due to the overexploitation (abstraction) of water but also due to direct discharges of untreated wastewater into nearby rivers. This wastewater carries a high organic load and contains cyanide, a toxic compound released through the hydrolysis of cyanogenic glycosides naturally present in cassava root (Torres et al. 2003; Dorca dos Santos et al. 2018; Taborda 2018). This widespread practice has consequences for ecosystems and those communities that depend on these water sources for their livelihoods and cultural practices. Furthermore, it exacerbates socio-cultural tensions by fostering conflicting perspectives on the value and use of water resources.

Addressing this problem requires research that analyzes the interdependence of these factors and promotes integrated solutions for socio-ecological systems (SES). SES are adaptive systems in which humans and ecosystems interact continuously, forming complex and dynamic relationships (Folke et al. 2016; Berkes 2017). In this sense, community trust and participation are key in places with such a complex socio-ecological situation. Participatory research emerges from the participatory action research (PAR) paradigm, which was developed to help community groups identify their problems, opportunities, and solutions collaboratively (Lewin et al. 1990). These tools encourage effective dialogue and stimulate active participation, ensuring that community perspectives and knowledge contribute meaningfully to the decision-making process. In the context of a Colombian rallandería, participatory research could represent a valuable tool to understand, promote, and implement, a water use and conservation plan in the Cauca region.

This study aimed to develop technical recommendations for a water management plan focused on treating wastewater from a Colombian rallandería. This was achieved by characterizing the SES of a rallandería and proposing participatory strategies with the community to improve water management. These strategies were evaluated through social multi-criteria evaluation (SMCE), and one was selected as it meets the technical, social, and environmental criteria relevant to the context of the rallanderías in the northern Cauca region.

The research focused on five rallanderías. These rallanderías were selected to represent the geographical and operational diversity in the Cauca region. The study included one large and four medium-sized rallanderías, four located in Santander de Quilichao (the area where most rallanderías are located), and one in Caldono. To respect the community's request for confidentiality, we identified them as R1, R2, R3, R4, and R5.

A mixed methodology integrating quantitative and qualitative data was implemented. For community work, participatory research (PI) tools were used (Mayrhofer et al. 2019). The collection and analysis of the information were carried out following the SES approach guidelines (Berkes 2017; Schluter et al. 2019) and Munda's SMCE (Munda 2004). The working group consisted of academics from engineering and social sciences, five rallanderías owners, 17 employees and collaborators, community experts, and 20 members of the public living near the rallanderías.

The study consisted of four phases: (1) characterization of the production system, (2) physicochemical characterization of water discharges, (3) socio-ecological characterization of rallanderías through participatory research, and (4) socio-technical evaluation of water management alternatives. Each phase is detailed below.

For the characterization of the production system, a survey was conducted in each rallandería, targeting owners, employees, and collaborators, to collect the following data in two sessions:

• (a) Social welfare: Number of employees, working conditions, kinship between employees and/or owners, level of education, training, experience, and safety conditions.

• (b) Starch production system: Type of infrastructure (machinery), description of the process, traceability, cassava source and price, water consumption, and discharges.

The wastewater from rallanderías was characterized at three locations: (1) after the washing, peeling, and grating of cassava, (2) after the separation and recovery of the starch slurry (from afrecho tanks), and (3) after starch recovery from the mancha channels. At each sampling point, a composite sample was taken every 20 min over a period of 4 h. The samples were kept at 4 °C until analysis at the laboratory the same day. Each sample was analyzed in triplicate to conduct physicochemical analyses according to the Standard Methods 23rd edition (Baird et al. 2017). The parameters analyzed were: BOD₅, COD, settleable solids, TSS, total cyanide, hydrogen sulfide, total arsenic, conductivity, pH, temperature, and dissolved oxygen. A list of the specific methods and their detection limits are included in Supplementary Table S1. Additionally, the wastewater volumetric flow rate of each sample point was calculated by the volumetric method (DGIAR 2015). The flow measurements and analyses were performed by the community services laboratory of the Escuela de Ingeniería de los Recursos Naturales y el Ambiente (EIDENAR), Universidad del Valle.

The PAR paradigm proposed by Lewin et al. (1990) was used as a methodological framework to establish a relationship with the community and characterize the SES. Participatory workshops, structured (for academics and experts) and semi-structured (for the general public) interviews, and field visits were used as tools, and documentary material was collected to provide technical and historical context. The inputs, outputs, participants, and their interaction within the system were defined.

Water management alternatives were proposed to minimize water consumption and discharge while enabling different levels of water treatment for a more sustainable operation. Participatory research (workshops and interviews with academics and community experts) was used to collaborate with the community and evaluate water management alternatives in the context of Colombian rallanderías. It consisted of three stages:

• (a) Definition of the alternatives. A thorough literature review on cassava starch production was conducted, including studies from Colombia and other countries such as Nigeria, Vietnam, China, Brazil, and Thailand. These studies provided valuable insights into management strategies that served as references for developing alternatives tailored to the Cauca region. Expert consultations with both community members and academics, combined with the SES characterization, guided the development of the final water management alternatives.

• (b) Selection of an alternative. Five alternatives were evaluated through a participatory process using SMCE (Munda 2004). The evaluation criteria were collaboratively defined by the community and researchers from the Universidad del Valle (Duque-Achipiz 2025). It considered the technical feasibility of the technologies, alongside their social, economic, and environmental impacts on the environment and community. The analytical hierarchy process (AHP) was employed to rank the alternatives, ensuring a comprehensive and systematic assessment (Folke et al. 2016). Finally, the results were discussed with community, social, and technical experts to identify the most suitable alternative for the context of Colombian rallanderías.

• (c) Technical evaluation of the selected water management alternative. A technical evaluation was conducted to assess and recommend specific technologies suitable for implementation in Colombian rallanderías. This analysis, based on an extensive literature review, considered key criteria such as implementation and operational costs, as well as the required level of technical expertise for effective operation.

This section provides a comprehensive characterization of the key socio-environmental factors and details the processing stages, water consumption patterns, and waste generation. These aspects are crucial in understanding the challenges related to wastewater management, and the long-term sustainability of these facilities.

• (a) Social welfare

The rallanderías analyzed in this study are classified as medium-sized, employing 4–16 people (see Supplementary Table S2). Employees typically have a secondary school education and do not require professional technical training for their roles. Knowledge is acquired through experience, with occasional training in machine operation, quality control, and other relevant areas.

• (b) Starch production system

Cassava processing involves the generation of three different wastewater effluents: wastewater 1 comes from washing, peeling, and grating; wastewater 2 from straining and afrecho separation; and wastewater 3 from mancha separation. Wastewater from washing and peeling is treated using ‘sandals’ (filters) to separate the solid residue, husk, and soil particles from the water to reuse it for further washing, thereby reducing the amount of water used. On the other hand, wastewater generation from the subsequent stages relies mainly on sedimentation of the afrecho and mancha by-products, which inefficiently separates the solids (with commercial value) from the liquid effluent. Finally, in some cases, additional treatment of the wastewater is carried out by passing it through coarse filters before being discharged into nearby rivers.

Wastewater characterization for each rallandería is presented in Table 1, alongside the corresponding cassava processing volume on the sampling day and the volumetric flow rates for each wastewater discharge.

All wastewater samples exhibited pH values below 5, except for wastewater 1 from R3, which recorded a pH of 6.1. The prevalence of acidic pH levels has been extensively reported in the literature (Oghenejoboh 2015; Amorim et al. 2018; Costa et al. 2022). Hydrogen sulfide and total arsenic concentrations were also analyzed in the samples; however, both remained below the detection limits, with arsenic measured at <1 mg/L and sulfide at <0.0025 mg/L. The BOD₅/COD ratio found in all the samples evaluated ranged from 0.32 to 0.67, indicating their potential to be treated biologically (Anupong et al. 2022). Comparable findings have been reported in similar cassava processing facilities, which revealed BOD₅/COD ratios between 0.50 and 0.75 (Amorim et al. 2018; Costa et al. 2022).

Wastewater 1 from the washing, peeling, and grating of cassava presented the highest values for TSS, above 500 and up to 2,880 mg/L in R4. In comparison, Costa et al. (2022) reported an average TSS of 712 mg/L, with values ranging from 30 to 3,540. On the other hand, when wastewater is conveyed in a single effluent, TSS measurement can present values as high as 6,500 mg/L (Fleck et al. 2017). R4 has the particularity of reusing water from the strainers (grating of cassava) to the washing of the peeled cassava, which increases the total amount of solids in the wastewater. Ultimately, it also results in an increase in the cyanide content, reaching a value of ∼16 mg/L, compared with the range of 1.1–5.1 mg/L found in wastewater 1 of the other four rallanderías.

In general, wastewater 2 (afrecho tanks) from the five rallanderías was the least polluted and had an acceptable biodegradability index as shown by the BOD₅/COD ratio. R4 sample was the only exception, exhibiting a COD concentration of 2,591 mg/L, nearly double that of the other wastewater 2 samples, which ranged from 697 to 1,395 mg/L. Usually, a rallandería includes two steps to separate the solids (afrecho) from the water, which consists of a filtration step followed by sedimentation in tanks, but R4 was the only rallandería included in this study that used sedimentation channels instead of the conventional two-step process to separate the solid particles from the wastewater.

Previous studies carried out in cassava processing plants in Cauca have shown that the mancha sedimentation stage generates the most polluted wastewater, with levels of organic matter and cyanide reaching up to 80% of the total pollution generated by discharges (Pérez et al. 2009; Taborda 2018). In our study, wastewater 3 presented the highest COD concentration reaching up to 7,252 mg/L, with total cyanide concentrations ranging from 1.3 up to 101 mg/L. Similarly, Dorca dos Santos et al. (2018) reported COD and cyanide levels of 7,732 and 1.27 mg/L, respectively. Additionally, laboratory-scale experiments showed that cyanide concentrations as high as 52 mg/L can be easily reached during cassava processing (Chukwuneke et al. 2024).

In all the cases, the wastewater exceeded the permissible limits according to the Colombian regulation, the 0631 Resolution of 2015, as shown in Table 2 (Ministerio de Ambiente y Desarrollo Sostenible 2015). Our findings highlight the need for improved water management in rallanderías. These improvements are essential not only to comply with Colombian regulations but also to improve socio-ecological interrelationships within the territory.

Within the SES, two main subsystems were identified: (a) social (orange-dotted circle, to the left) and (b) ecological (green-dotted circle, to the right). Social subsystems are mainly comprised of various social agents influencing water use, such as state entities, cassava producers, customers, and academic institutions through research. The ecological subsystem includes the main elements of the environment, such as water, soil, and living organisms. The interdependence between these subsystems includes bidirectional interactions. Changes in the environment affect water availability and quality, which in turn impacts cassava starch production and stake-holder decision-making, whereas the ecological subsystem is affected by the disturbance of water use by rallanderías and impacted by wastewater discharges, leading to the ecosystem degradation. That is, the ecological subsystem provides various ecosystem services for the development of cassava agro-industry, while the social subsystem is considered to increase entropy, reflected by the amount of dispersed or disordered energy and resources in the ecological subsystem.

This participatory characterization of rallanderías provided key information on challenges and opportunities, laying the groundwork for the proposal for water management strategies that balances the social and ecological components of the system.

• (a) Definition of the alternatives

Based on the data collected from the participatory research, four alternatives in addition to the traditional option (Alternative 1, conventional process) were proposed to the rallandería owners and experts (Table 3). Water consumption reduction and wastewater treatment technologies were the main technical improvements included in the proposed alternatives.

Alternative 1 was the starting point, where all processes are traditional and artisanal, with low technological advancement, high water consumption, and insufficient wastewater treatment. Alternative 2 incorporated technical and technological innovations already adopted by some rallanderías owners, such as machinery optimization to reduce water and energy consumption, dry pre-peeling equipment, pressurized water, improved settling tanks for afrecho, starch, and mancha, and integration of channels and settling tanks for wastewater separation and partial treatment.

Alternative 3 introduced water management technologies such as leak prevention, dry pre-peeling, pressurized water for peeling and washing, sludge tanks, wastewater tank filters, an anaerobic reactor for wastewater treatment, and cleaner production (CP) throughout the process. Alternative 4, unlike alternative 3, implemented centrifuges and hydrocyclones in the straining phase, reducing water use, operating times and increasing starch productivity. Alternative 5 proposed a primary pretreatment for the wastewater and sending the effluent to a centralized treatment plant serving multiple rallanderías.

From Alternatives 3 to 5, the activities proposed throughout the process promote pollution prevention, which is why the implementation of CP is recommended. CP is defined as a continuous prevention strategy applied to all processes, services, and products of the agro-industry, whose application allows the improvement of environmental performance by minimizing waste and risks to the environment and human health (De Mello Santos et al. 2022).

• (b) Selection of an alternative for water management

The selection of the favored alternative resulted from a multi-criteria evaluation process that prioritized the solution that most effectively integrated the technical, social, and environmental requirements of a rallandería. The active participation of the different stakeholders ensured that the final decision was legitimate and representative of the perspectives of all those involved.

Alternative 3 was selected as the most viable alternative following the evaluation process in consultation with the community and experts. This alternative stood out for its comprehensive approach, which included reducing water consumption and implementing basic anaerobic wastewater treatment. In detail, the proposal included pipeline renovations to prevent water leaks, and then, the implementation of machinery for more efficient water use. For the straining step, ‘recollectors’ were recommended. Recollectors use pressurized water during cassava grating, reducing the water used and increasing starch recovery efficiency. In terms of the generated wastewater in the first half of the process, filters (to separate most of the bagasse) and water recirculation were proposed. Additionally, biological reactors for wastewater treatment were incorporated.

Moreover, this alternative also provided a cost-effective solution in terms of implementation and long-term operation. The criteria above ensured that the selected alternative was technically feasible and socially accepted by the rallandería owners and community.

• (c) Technical evaluation of the selected water management alternative

Based on our physicochemical characterization findings, we propose a three-step wastewater treatment system for Colombian rallanderías. In this approach, the three wastewater streams are collected and directed to an equalization tank before treatment. These tanks will act as reservoirs to regulate wastewater flow, ensuring consistent hydraulic retention times in the biological reactors, even when rallanderías are not processing cassava. Additionally, they will enable early pH regulation strategies. Following the equalization tank, a pretreatment stage will be implemented to facilitate the separation and recovery of the mancha by-product. This will be followed by a secondary treatment using biological reactors, and finally, a post-treatment step will aim to reduce the cyanide content in the final effluent.

In almost all cases, wastewater 3 from mancha separation presented the highest level of cyanide (except for R5). The physicochemical properties of the mancha can contribute to the obstruction of pipes and reactors. Torres et al. (2014) reported that mancha particles range in size from 2 to 8 μm and settle as discrete, flocculent particles. This means that, to achieve effective particle removal, a critical flow of 1.5 L/s, a surface area of 1,288 m², a height of 1 m, and a hydraulic retention time (HRT) of 10 days is required. However, these requirements are economically, technically, and spatially unfeasible, highlighting the need for alternative management strategies.

Different approaches can be discussed to achieve higher mancha separation from wastewater. Currently, the so-called mancha sedimentation channels are the most often used process. It consists of interconnected channels filled with starch slurry for the recovery step. Depending on the site-specific conditions, the water retention times can be adjusted from a few hours to a day to allow an effective phase separation. Extending the surface area or changing the geometry of the sedimentation channels could also represent valuable alternatives for a more efficient by-product separation and recovery. Finally, upgraded machinery can be incorporated into the process. For example, the effect of gravity can be improved by using shear forces exhibited in centrifugal extractors (Hamamah & Grützner 2022). The difference in densities would allow a reasonable separation for the mancha recovery. However, the major drawback is the capital cost for this type of machinery compared with other technologies.

Anaerobic reactors have been proposed for the treatment of cassava production wastewater. As seen in Table 1, the biodegradability index denoted by the BOD₅/COD ratio demonstrated the feasibility of biological processes to treat the wastewater generated during cassava starch production. Different anaerobic reactor configurations have been previously designed and tested at the laboratory and full-scale for this approach. Torres et al. (2005) evaluated a horizontal flow anaerobic filter with integrated baffles, which showed removal efficiencies (REs) of 79% for COD and 90% for TSS while operating at an HRT below 14 h. The study emphasized the importance of an inoculum to reduce the start-up time and ensure a stable reactor operation. Meanwhile, Ferraz et al. (2009) tested a compartmentalized anaerobic baffled reactor. The results demonstrated that incorporating multiple chambers enhances pH and alkalinity control, as acidogenic reactions predominantly occur in the first chamber, while methanogenic activity is established in the last compartment. Sun et al. (2012) investigated using a novel up-flow multistage anaerobic reactor using a fixed HRT of 6 h. The reactor removed up to 90% of the COD when the organic loading rate was 40 kg_COD/m³d. On the other hand, Watthier et al. (2019) evaluated the feasibility of a fixed-bed reactor for pollution reduction while producing biogas. Their study showed COD RE above 90%, whereas, for the TSS, the RE was 85%. Biogas production ranged from 1 to 1.4 L/d.

In alignment with these previous studies, the installation of a multistage high-rate anaerobic reactor, highlighting the critical role of an initial inoculum for the successful implementation of this technology, is recommended. A key challenge associated with this technology is the need for trained technical personnel to ensure proper operation. Notably, this study found that similar technologies had already been installed in some Colombian rallanderías, but failed primarily due to a lack of knowledge regarding their operation, maintenance, and troubleshooting.

Various physicochemical, chemical, and biological methods have been developed for cyanide removal from water. While most of the studies focus on applications in the mining and chemical industries, these technologies could be implemented at cassava processing plants. Some technologies investigated to date are bioremediation, coagulation/electrocoagulation, chlorination, adsorption, chemical oxidation, and photocatalysis (Martínková et al. 2023). In this study, we will focus on bioremediation and photocatalysis due to their minimal personnel training requirements for operation and low reliance on chemical additives, contributing to reduced operational costs.

Bioremediation offers a practical and environmentally friendly solution to reduce cyanide concentration in wastewater. Its approach relies on the natural ability of microorganisms to degrade contaminants, providing a sustainable strategy. Some of the genera with reported species for cyanide removal are Rhizobium, Acinetobacter, Cupriavidus, Methylobacterium, Bacillus, Rhodococcus, Pseudomonas, Agrobacterium, and Variovorax. Studies conducted by Anupong et al. (2022) and Maciel et al. (2023) demonstrated that some strains from these genera achieved cyanide removal greater than 95% after 10 days.

Cyanide biodegradation can occur under aerobic or anaerobic conditions, with notable differences in efficiency and requirements. Under aerobic conditions, microorganisms metabolize hydrogen cyanide (HCN), generating hydrogen cyanate, which is subsequently hydrolyzed into ammonia and carbon dioxide. In contrast, anaerobic cyanide biodegradation requires hydrogen sulfide (H₂S) and a pH above 7.0 (Maciel et al. 2023). However, anaerobic cyanide removal occurs at slower rates and is ineffective at degrading certain cyanide species, such as thiocyanate (Dash et al. 2009).

An effective strategy is to enrich indigenous microorganisms already adapted to high cyanide concentrations. The latter will reduce operational costs and ensure a rapid start-up of the treatment process. While bioremediation typically involves high capital costs, its significantly lower operational costs make it an attractive long-term option. Additionally, cyanide biodegradation does not produce toxic by-products, further supporting its viability.

From a technological perspective, cyanide treatment can be implemented using either suspended or fixed biomass, including rotating biological contactors, packed beds, biofilters, sequencing batch reactors, facultative lagoons, and activated sludge systems (Dash et al. 2009).

In the context of a Colombian rallandería, sequencing batch reactors could be a suitable option due to their flexibility, efficient cyanide degradation, and adaptability to fluctuating wastewater loads, a crucial factor given the seasonal nature of cassava processing. However, maintaining a stable microbial community capable of handling high cyanide concentrations remains a challenge, as it requires specialized expertise and controlled environmental conditions.

Photocatalysis stands out for its ability to purify polluted streams using solar energy during the process. The Cauca Region exhibits a high annual average irradiance value of 4.3 kWh/m² compared with the global average of 3.9 kWh/m² (IDEAM 2024). Unlike other photochemical processes, this method is non-selective, making it suitable for the treatment of complex mixtures of pollutants. One option is to use a relatively affordable and non-toxic catalyst, such as titanium dioxide (TiO₂) in its crystalline form, which has also demonstrated biological compatibility (Vargas & Cuesta 2009; Wei et al. 2023).

A previous study used laboratory photoreactors equipped with 100–200 W ultraviolet light lamps to purify water, reaching pH values close to 10.0 units (Vargas & Cuesta 2009). Under UV radiation alone, cyanide RE reached 24% in 90 min. However, when H₂O₂ was added, RE increased to 93% over the same period, demonstrating a direct correlation between degradation time and H₂O₂ concentration, alongside a proportional decrease in pH. The combined TiO₂/UV/H₂O₂ system demonstrated high efficiency in cyanide degradation, reaching an elimination of 94% in 40 min and 99% in 60 min, with less alteration of the pH of the solution. These results highlight the potential for rapid and efficient cyanide elimination (Quispe et al. 2011). However, while TiO₂ offers advantages as a photocatalytic material, it is hindered by limitations such as poor light absorption capacity and low selectivity toward organic compounds, which compromises its stability and efficiency. To overcome these problems, studies have been conducted on the development of composite materials. Experiments have shown that adding MnO₂ and doping TiO₂ with lanthanum significantly improves its photoelectric conversion efficiency, achieving a cyanide degradation of over 96% (Jaramillo-Fierro & León 2023).

It is important to note that implementing photocatalysis systems may present challenges. The operation and maintenance will require specialized technical and scientific personnel, particularly when reagents need to be added; besides, their acquisition may not be easy in the Cauca region. While photo-oxidation does occur naturally in waste stabilization ponds (Curtis et al. 1992), we do not know if the oxidative capacity in these systems will be sufficient to effectively remove cyanide. Given these considerations, we recommend that further studies should be conducted to assess the feasibility of photocatalysis in Colombian rallanderías. Specifically, evaluating the cost-effectiveness and operational (including training) requirements will be crucial for determining their practical applicability.

This study provides insight into wastewater management in the cassava agro-industry in Cauca, Colombia, highlighting the importance of integrating community participation with technical design to develop sustainable solutions. A successful technology must be socially, economically, and technically acceptable to its users and society more broadly. To determine the technical requirements, we characterized the wastewater from five rallanderías, identifying problems with organic matter, suspended solids, and cyanide.

Given these challenges, including rallanderías in wastewater regulations with specific guidelines for cyanide management is crucial, as current legislation (Resolution 0631 of 2015) does not account for the unique challenges posed by cassava processing. Establishing targeted policies, alongside technical and financial support from government authorities, would ensure more effective and context-specific water management strategies for these agro-industries.

To ensure social and economic viability, we completed a participatory study in collaboration with the community. The participatory research allowed us to identify and select water management alternatives based on technical, social, and environmental criteria. The proposed solution includes preventing leakages, reducing water use by using novel equipment, and integrating wastewater treatment technologies before discharge into the environment.

Integrating community knowledge with technical expertise is essential to develop sustainable environmental solutions. Participatory research not only enhances wastewater management but also fosters local ownership and empowerment, ensuring the long-term sustainability of the implemented strategies. Furthermore, this collaborative methodology highlights the importance of socio-ecological considerations in engineering practices and the need for holistic approaches to address environmental challenges in the agro-industrial sector.

This work was supported by the Water Security and Sustainable Development Hub funded by the UK Research and Innovation's Global Challenges Research Fund (GCRF) [grant number: ES/S008179/1]. We are grateful to the rallanderías owners and their collaborators in the regions of Santander de Quilichao, Caldono, and Piendamó, in Cauca, Colombia, who participated in this research project.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.