Study area

The RUT District is in the southwestern region of Colombia, in the northern part of the Valle del Cauca Department, within the RUT watershed. The district spans the area between the foothills of the Western Andes and the left bank of the Cauca River, encompassing territories under the jurisdiction of the municipalities of Roldanillo, La Unión, and Toro (Figure 1). The region has a moderately warm climate with an average temperature of 24 °C and a bimodal rainfall distribution, with a mean annual precipitation of 1,015 mm and an average annual evaporation of 1,145 mm. The total area of the RUT District is approximately 10,243 hectares, bounded to the west by the open Interceptor Canal, to the east by a marginal protection levee, and bisected by the Main Drainage Canal. The district is managed by the User Association (ASORUT) and benefits approximately 1,200 farmers (Echeverri-Sánchez et al. 2017). Land use is predominantly sugarcane cultivation (50.4%), followed by maize (14.9%), with smaller areas allocated to chili pepper, soybean, cacao, grape, papaya, melon, guava, passion fruit, and other fruit crops.

Figure 1: Study site and sampling points in the RUT district.

Sampling site selection, temporality, and water quality indicators

The seven sampling points were located along the Interceptor Canal of the RUT District (Figure 1). Site selection was based on hydraulic infrastructure, discharge points, and land use characteristics. Additionally, historical precipitation data were used to identify representative sampling periods: November 2021 (high rainfall), June 2022 (low rainfall) and October 2022 (transitional conditions), thereby capturing maximum, minimum, and average hydrological scenarios. The RUT Irrigation District was subdivided into Southern, Central, and Northern zones, corresponding to the municipalities of Roldanillo, La Unión, and Toro, respectively (Table 1).

Table 1: Sampling points description.

Sampling procedure

Surface water samples were collected as composite samples at all sampling points, following the protocols established by the Instituto de Hidrología, Meteorología y Estudios Ambientales (IDEAM 2002). An aliquot was collected every 20 minutes over a 1-hour sampling period. Samples collected during the first campaign were analyzed by Análisis Ambiental S.A.S., accredited by IDEAM under Resolution No. 0710 of 2019. Samples from subsequent campaigns were analyzed by AGQ Colombia S.A.S., an IDEAM-accredited laboratory under Resolution No. 1726 of 2022. Table 2 summarizes the analytical methods used by the laboratories to determine the evaluated parameters.

Table 2: Analytical methods used for the determination of irrigation water quality parameters.

Sampling was carried out by identifying and georeferencing each sampling point using a GPS receiver. Water samples were collected using a calibrated bucket, rinsed prior to each collection, from elevated structures (bridges), which allowed sampling at the center of the canal and at mid-depth of the water column.

For physicochemical analyses, 1-L pre-cleaned plastic containers were used; in the case of hardness determination, sulfuric acid was added as a preservative.

For fecal coliform analysis, samples were collected in sterilized glass containers with the addition of 0.2 mL of 3% sodium thiosulfate (Na₂S₂O₃). All samples were kept under refrigeration and transported to the laboratory for analysis within the established holding times.

Parameter selection

Tables 3 - 6 present the selected hazard types for evaluation, along with the associated parameters, hazard categories, and risk levels. This information was organized based on a review of the scientific literature and criteria established by the authors, as well as relevant studies pertaining to each hazard type.

Table 3: Methods, categories, and levels for estimating the potential impact on soil from the quality of irrigation water.

Table 4: Methods, categories, and levels for estimating the potential impact on crops from the quality of irrigation water.

Table 5: Methods, categories, and levels for estimating the potential impact on irrigation infrastructure from the quality of irrigation water.

Table 6: Methods, categories, and levels for estimating the potential impact on human health from the quality of irrigation water.

It should be noted that two methods were selected to evaluate the hazard of soil salinization: one based on the FAO guidelines (Drechsel et al. 2023) and another proposed by Echeverri-Sánchez (2016). The FAO methodology classifies the salinity risk of irrigation water solely on the basis of ECw, whereas the approach developed by Echeverri-Sánchez (2016) jointly integrates ECw and salt solubility, the latter estimated according to the methodology proposed by Villafañe (2011). Under this framework, ECw reflects the total amount of dissolved salts present in the water, while salt solubility indicates the ease with which these salts can be leached from the soil through lixiviation processes.

Validation of results and hazard estimation

The laboratory results were validated through ion balance estimation and comparison with ECw values (Mageshkumar and Vennila 2020). Once the data were verified-within an acceptable error margin of 15%-the calculation of hazard indicators related to water quality was carried out, as presented in Tables 3 through 6.

Microbiological Hazard

Across the three zones identified, the microbiological hazard associated with FC showed varying levels, evidenced by pronounced differences in median values and P25-P75 ranges (Table 7), with a marked tendency toward a "High" hazard classification at sampling points located downstream from the wastewater treatment plant discharges of the three municipalities (points 3, 6 and 7) (Figure 2). During the second sampling campaign, the highest concentrations were recorded at these points-particularly at point 6, where a "Very High" hazard level was observed with a value of 920,000 MPN 100 mL^-1, likely due to the campaign being conducted during a period of lower dilution. This behavior can be explained by the fact that, during dry periods, the flow in the Interceptor Canal is reduced and pumping from the Cauca River is sometimes restricted, thereby increasing the influence of the treatment plant discharges on the system. In addition, these FC concentrations are also attributable to untreated sewage discharged directly into the canal throughout the year from approximately 600 households.

Table 7: Descriptive statistics (median, P25, and P75) of water quality parameters associated with risk categories by monitoring campaign.

Figure 2: Results of microbiological hazard from FC associated with water quality.

The crops in these zones do not come into direct contact with irrigation water. Notably, sugarcane-the predominant crop-is processed industrially, and its final product undergoes treatments that can enhance food safety (Madera et al. 2009). However, concern remains for fruit crops that are consumed with their peel and for short-stature vegetable crops, as Colombian regulations establish a maximum limit of 1,000 MPN 100 mL^-1 for fecal coliforms. Non-compliance with these standards compromises food safety and restricts the potential of export-oriented agricultural production.

Soil Degradation Hazard

Two different methodologies were applied to evaluate the salinity hazard (Figure 3): the approach proposed by Drechsel et al. (2023) and the method developed by Echeverri-Sánchez (2016). According to the Drechsel et al. (2023) methodology, 95% of the samples analyzed across the three campaigns were classified within the "Very Low" and "Low" hazard categories. The only exception was observed (dS m^-1) (Figure 3A). In contrast, Echeverri-Sánchez (2016) method demonstrated greater sensitivity in hazard classification (Figure 3B). Under this approach, 38% of the sampling points increased to a "Moderate" hazard level, while the remaining 57% remained classified as "Low."

Figure 3: Results of soil salinity hazard associated with irrigation water quality. Background shading indicates salinity hazard classes according to A. Ayers and Westcot (1985), and B. Echeverri (2016).

Post-sampling analysis revealed that the salinity hazard to soils associated with irrigation water quality increases from "Low" to "Moderate" when moving from the Central Zone toward the Northern Zone of the district. This zone is characterized by gentler slopes, poor drainage, and heavy-textured soils that are prone to physicochemical degradation (Echeverri-Sánchez 2022). Although the solubility of salts in this zone is low (<1,000 mmolc L^-1), the ECw of the irrigation water exceeds 1,000 dS m^-1.

The pH, generally slightly alkaline and classified as a "Moderate" hazard at the sampling points, indicates a relative reduction of calcium and magnesium in comparison to Na⁺ and the SAR. This imbalance promotes the formation of compounds that reduce the availability of essential nutrients for plants (Kundu et al. 2022). Among the common salts detected was NaCl, which may disrupt water uptake and nutrient absorption in plants (Lu and Fricke 2023). Similarly, points 2 and 7 were consistently characterized by the presence of MgCl₂, which at high concentrations can significantly impair protein synthesis in plants (Geilfus 2018).

Regarding sodicity hazard (Figure 4), "High" hazard levels were observed, particularly in the Southern Zone, at sampling points 1 and 2, where values exceeded 325 g m^-3 of pure gypsum required for water treatment (Villafañe 2011). It should be noted that the concentration of this compound exhibited marked contrasts between Campaign 1 and the subsequent campaigns, as the median value in Campaign 1 was approximately twice that observed in the others (Table 7). Nevertheless, data dispersion in Campaign 1 was relatively low, whereas Campaigns 2 and 3 displayed wider interquartile ranges, indicating greater variability during these latter periods. Sodicity leads to the dispersion of soil aggregates, which reduces water infiltration and hydraulic conductivity, and promotes the formation of surface crusts. These conditions hinder root growth and limit the uptake of water and nutrients by plants (Hailu and Mehari 2021).

Figure 4: Results of soil sodicity hazard associated with irrigation water quality.

Specific ion toxicity hazard

Results related to specific ion toxicity indicated that Cl⁻ concentrations were below 213 mg L^-1, corresponding to a hazard "Low" level, and that Na⁺ values, evaluated using SAR, were lower than 3 meq L^-1 (Figures 5C and 5D). Consequently, none of these parameters reached levels considered detrimental to the dominant crops in the RUT Irrigation District, as these thresholds correspond to general low-risk conditions for agricultural irrigation according to the reference limits applied.

However, analysis of Fe²⁺ concentrations, based on the adopted risk thresholds (<5 mg L^-1 low, 5-20 mg L^-1 medium, and >20 mg L^-1 high), made it possible to identify sampling points with consistently critical behavior. While Campaign 2 exhibited concentrations within the "Low" hazard range, Campaigns 1 and 3 showed medium-risk conditions, with P75 values approaching or exceeding the "High" hazard threshold, indicating the occurrence of localized critical episodes that do not uniformly affect the entire system (Table 7). Sampling point 1 emerged as the most problematic within the system, recording "High" hazard concentrations during Campaign 1 (24.5 mg L^-1) and Campaign 3 (102.0 mg L^-1), evidencing the recurrence of severe events with a high likelihood of Fe²⁺ precipitation and emitter clogging. In contrast, sampling points 2 and 4 exhibited concentrations predominantly within the "Medium" or "Low" hazard range, without exceeding the critical threshold, and sampling point 5 consistently remained under "Low" hazard conditions across all campaigns (Figure 5A).

Figure 5: Concentrations of parameters associated with specific ion toxicity hazard: A. iron, B. manganese, C. chlorides, and D. sodium, for each sampling campaign.

Although soil pH in the RUT District ranges from slightly acidic (pH=6) to slightly alkaline (just above pH=7), this factor alone would not significantly limit Fe²⁺ availability. Nevertheless, the Northern Zone of the district was identified as having poor drainage conditions, which could favor the accumulation of this ion in crops (Lei et al. 2014). Similarly, excess Mn²⁺ may induce toxic effects in plants, manifested as interveinal chlorosis, necrosis in young leaves, and impaired root development (Lee et al. 2011).

Similarly, considering the risk thresholds for Mn²⁺(<0.2 mg L^-1 Low, 0.2-10 mg L^-1 Medium, and >10 mg L^-1 High), the results show a progressive increase in hazard across campaigns, with median values shifting from predominantly "Low" hazard conditions in Campaign 1 to "Medium" hazard conditions in Campaigns 2 and 3 (Table 3). Sampling points 1 and 2 were identified as the most critical within the system, as they consistently exhibited concentrations within the "Medium" hazard range in all or most campaigns, reaching maximum values of 1.28 and 1.54 mg L^-1, respectively (Figure 5B). It should be noted that excess Mn²⁺ may induce toxic effects in plants, manifested as interveinal chlorosis, necrosis in young leaves, and impaired root development (Lee et al. 2011).

Scaling, corrosion and emitter obstruction hazard

The LSI was negative in all campaigns, indicating corrosive potential (Figure 6F). Campaign 1 exhibited the highest corrosivity and variability, whereas Campaigns 2 and 3 showed more homogeneous corrosive conditions, as reflected by the median and interquartile range (Table 7).

Figure 6: Results of hazards to irrigation infrastructure and emitter clogging associated with A. TSS, B. total suspended solids, C. pH, D. iron, E. manganese, and F. the Langelier Saturation Index.

According to the index classification, 67% of the points fell within the "Severe Corrosion" category, 24% in "No scale formation but slight corrosion," and 10% in "Intolerable Corrosion." In Campaign 1, the most negative values were recorded in the Southern Zone at points 1 (-2.92) and 2 (-4.59), indicating intolerable corrosion. In Campaign 2, corrosivity decreased at several points, such as point 5 (-0.17) and point 3 (-0.54). For Campaign 3, hazard levels remained predominantly in the "Severe Corrosion" range. This corrosive trend in the water could primarily affect metallic components (Panday et al. 2021), resulting in hydraulic and economic losses, as well as challenges in the logistics and application of scheduled irrigation volumes. In the western zone of the RUT District-the focus area of this study-drip, sprinkler, and gravity irrigation systems are used, each requiring at least one pumping unit to ensure water delivery due to the low terrain slope. Therefore, this hazard must be considered when selecting equipment or implementing water quality improvement systems on farms.

On the other hand, the hazard of clogging in high-frequency irrigation emitters due to physical parameters revealed that the hazard from TDS was predominantly classified as "Low" (Figure 6A). In contrast, the hazard from TSS showed significant variability in particulate load across campaigns. Campaign 1 registered the highest levels, particularly at point 7 (9,400 mg L^-1). In Campaign 2, a drastic reduction was observed at most points, with values below 120 mg L^-1 except for point 1 (558 mg L^-1) and point 7 (473 mg L^-1). In Campaign 3, concentration remained low compared to the first campaign, with the lowest recorded value at point 7 in the Northern Zone (23 mg L^-1).

The Cauca River water exhibited a "High" hazard level for TSS at point 1 due to its high sediment load, reflected in the elevated sedimentation rate within its canal network. In a section of the Interceptor Canal in the Southern Zone, where two rivers and two streams converge, approximately 12,000 m³ of sediment are removed quarterly from the canal bed deposited along the canal bank. Downstream from this point, the hazard level decreased to "Moderate," likely due to a desilting structure located in the main canal that reduces the solid load. This decreasing trend persisted through the Central Zone, but toward the Northern Zone, the hazard level increased again to "High."

The hazard of clogging in high-frequency emitters due to chemical parameters during Campaign 1 revealed chemically aggressive conditions, with acidic pH values at points 2 (4.03) and 1 (5.71) (Figure 6C), as well as elevated concentrations of Fe²⁺ (Figure 6D) and medium concentration of Mn²⁺ (Figure 6E). In Campaign 2, pH values rose toward neutral, but Mn²⁺ reached its highest concentration (1.53 mg L^-1 at point 2), maintaining a "High" hazard level. Fe²⁺ concentrations, though still present, were relatively lower, mostly within the "Moderate" hazard range. In Campaign 3, pH stabilized within neutral to slightly alkaline ranges (up to 7.72 at point 7), while Fe²⁺ levels increased critically, reaching up to 102 mg L^-1 at point 1 and 28.4 mg L^-1 at point 3, maintaining the "High" hazard classification. These Fe²⁺ concentrations far exceed the recommended thresholds for avoiding encrustation in high frequency localized irrigation systems which are set at concentrations lower than 1.5 mg L^-1. Under neutral pH and in the presence of oxygen, Fe²⁺ oxidizes to form Fe (OH)₃ precipitates, which can severely clog emitters and conveyance lines (Muniz et al. 2023). These issues can significantly reduce the efficiency of on-farm irrigation systems and lead to high maintenance costs for users.

The results obtained provide a basis for guiding operational measures for irrigation water management according to the main hazard factors identified. TSS concentrations show an increase toward the downstream sampling points, where the risk of emitter clogging becomes more critical; therefore, the implementation of sedimentation and/or pre-filtration processes is recommended to reduce the particulate load. In addition, periodic flushing of pipelines is advised in these areas as a preventive measure. When elevated concentrations of Fe and Mn are recorded, especially at the upstream sampling points, the use of oxidation followed by filtration helps to minimize precipitation and clogging problems. Furthermore, during temporary peaks in ECw, blending with higher-quality water sources can be employed as a strategy to reduce impacts on the irrigation system and the soil.